Computingedge (ISSN 2469-7087) Is Published Monthly by the IEEE Computer Society

HOT TOPICS AND EMERGING TECHNOLOGIES

In this issue: > Future Automotive Architecture and the Impact of IT Trends > Cancer’s Big Data Problem

JUNE 2017 www.computer.org Are Enemy Hackers Slipping Through Your Team’s Defenses?

Protect Your Organization From Hackers By Thinking Like Them

Take Our E-Learning Courses in the Art of Hacking

You and your staff can take these courses where you are and at your own pace, getting hands-on, real-world training that you can put to work immediately.

www.computer.org/artofhacking IEEE COMPUTER SOCIETY http://computer.org • +1 714 821 8380

STAFF

Editor Manager, Editorial Content Lee Garber Carrie Clark

Contributing Staff Senior Manager, Editorial Services Christine Anthony, Brian Brannon, Lori Cameron, Cathy Martin, Robin Baldwin Chris Nelson, Meghan O’Dell, Dennis Taylor, Rebecca Torres, Bonnie Wylie Director, Products and Services Evan Butterfield Production & Design Carmen Flores-Garvey, Monette Velasco, Jennie Zhu-Mai, Senior Advertising Coordinator Mark Bartosik Debbie Sims

Circulation: ComputingEdge (ISSN 2469-7087) is published monthly by the IEEE Computer Society. IEEE Headquarters, Three Park Avenue, 17th Floor, New York, NY 10016-5997; IEEE Computer Society Publications Office, 10662 Los Vaqueros Circle, Los Alamitos, CA 90720; voice +1 714 821 8380; fax +1 714 821 4010; IEEE Computer Society Headquarters, 2001 L Street NW, Suite 700, Washington, DC 20036. Postmaster: Send address changes to ComputingEdge-IEEE Membership Processing Dept., 445 Hoes Lane, Piscataway, NJ 08855. Periodicals Postage Paid at New York, New York, and at additional mailing offices. Printed in USA. Editorial: Unless otherwise stated, bylined articles, as well as product and service descriptions, reflect the author’s or firm’s opinion. Inclusion in ComputingEdge does not necessarily constitute endorsement by the IEEE or the Computer Society. All submissions are subject to editing for style, clarity, and space. Reuse Rights and Reprint Permissions: Educational or personal use of this material is permitted without fee, provided such use: 1) is not made for profit; 2) includes this notice and a full citation to the original work on the first page of the copy; and 3) does not imply IEEE endorsement of any third-party products or services. Authors and their companies are permitted to post the accepted version of IEEE-copyrighted material on their own Web servers without permission, provided that the IEEE copyright notice and a full citation to the original work appear on the first scree n of the posted copy. An accepted manuscript is a version which has been revised by the author to incorporate review suggestions, but not the published version with copy-editing, proofreading, and formatting added by IEEE. For more information, please go to: http://www.ieee .org/publications_standards/publications/rights/paperversionpolicy.html. Permission to reprint/republish this material for commercial, advertising, or promotional purposes or for creating new collective works for resale or redistribution must be obtained from IEEE by writing to the IEEE Intellectual Property Rights Office, 445 Hoes Lane, Piscataway, NJ 08854-4141 or pubs-permissions@ieee .org. Copyright © 2017 IEEE. All rights reserved. Abstracting and Library Use: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy for private use of patrons, provided the per- copy fee indicated in the code at the bottom of the first page is paid through the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. Unsubscribe: If you no longer wish to receive this ComputingEdge mailing, please email IEEE Computer Society Customer Service at [email protected] and type “unsubscribe ComputingEdge” in your subject line. IEEE prohibits discrimination, harassment, and bullying. For more information, visit www.ieee.org/web/aboutus/whatis/policies/p9-26.html.

IEEE Computer Society Magazine Editors in Chief

Computer IEEE Micro IEEE Intelligent Systems Sumi Helal, University of Florida Lieven Eeckhout, Ghent V.S. Subrahmanian, University University of Maryland IEEE Software IEEE MultiMedia Diomidis Spinellis, Athens IEEE Computer Graphics Yong Rui, Lenovo Research University of Economics and and Applications and Technology Business L. Miguel Encarnação, ACT, Inc. IEEE Annals of the History IEEE Internet Computing IEEE Pervasive Computing of Computing M. Brian Blake, University of Maria Ebling, IBM T.J. Watson Nathan Ensmenger, Indiana Miami Research Center University Bloomington

IT Professional Computing in Science IEEE Cloud Computing San Murugesan, BRITE & Engineering Mazin Yousif, T-Systems Professional Services Jim X. Chen, George Mason International University IEEE Security & Privacy Ahmad-Reza Sadeghi, Technical University of Darmstadt www.computer.org/computingedge 1 JUNE 2017 • VOLUME 3, NUMBER 6

THEME HERE 17 28 42 Cancer’s Big The Evolving IT and Future Data Problem Role of the Unemployment: CDO Part 2 7 Spotlight on Transactions: A High-level Language for Interactive Data Visualization LEILA DE FLORIANI

8 Editor’s Note: Hot Topics and Emerging Technologies 9 Future Automotive Architecture and the Impact of IT Trends MATTHIAS TRAUB, ALEXANDER MAIER, AND KAI L. BARBEHÖN 15 Fog Computing for Smart Living BRUCE MCMILLIN AND TAO ZHANG 16 The Internet of Things Will Upend Our Industry BRUCE SCHNEIER 17 Cancer’s Big Data Problem JUSTIN H.S. BREAUX 20 From Artificial Intelligence to Augmented Intelligence YONG RUI 22 The Future of Pervasive Health KAY CONNELLY, OSCAR MAYORA, JESUS FAVELA, MAIA JACOBS, ALEKSANDAR MATIC, CHRIS NUGENT, AND STEFAN WAGNER 28 The Evolving Role of the CDO SETH EARLEY 34 PaaS Death Watch? DAVID S. LINTHICUM 38 Predictability or Early Warning: Using Social Media in Modern Emergency Response MARCO AVVENUTI, STEFANO CRESCI, ANDREA MARCHETTI, CARLO MELETTI, AND MAURIZIO TESCONI 42 IT and Future Unemployment: Part 2 GEORGE STRAWN 45 Preparing Tomorrow’s Software Engineers for Work in a Global Environment SARAH BEECHAM, TONY CLEAR, JOHN BARR, MATS DANIELS, MICHAEL OUDSHOORN, AND JOHN NOLL 49 Incorporating Sustainability into Computing Education DOUGLAS H. FISHER, ZIMEI BIAN, AND SELINA CHEN 49 Departments 4 Magazine Roundup Incorporating 56 Computing Careers: Careers in Emerging Technologies for Sustainability People Shifting Jobs from Industry to Academia into Computing 58 Career Opportunities Education Subscribe to ComputingEdge for free at www.computer.org/computingedge. CS FOCUS

Magazine Roundup

in encrypted email, why security tools should be as dynamic as users are, and longitudinal privacy management in social media. There are also articles on the future of force-sensitive authentication and the design he IEEE Computer IEEE Software of eff ective privacy notices and Society’s lineup of 13 controls. Tpeer-reviewed technical Computer and communica- magazines covers cutting-edge tions technology are playing an Computing in Science & topics ranging from software increasingly important role in Engineering design and computer graphics vehicles. IEEE Software’s May/ to Internet computing and secu- June 2017 special issue exam- The number of computing and rity, from scientifi c applications ines automotive IT and soft- STEM-related jobs continues and machine intelligence to ware development, discusses to grow, but the diversity of the cloud migration and microchip which technologies and princi- workforce in these areas has design. Here are highlights from ples deliver value, and explains not. We can measure diversity recent issues. how manufacturers can intro- in many ways, including gen- duce them quickly. der, ethnicity, sexual orienta- Computer tion, socioeconomic standing, IEEE Internet Computing and disability status. Articles Thanks to advances in VLSI in CiSE’s May/June 2017 spe- technology, computing and com- IEEE Internet Computing’s May/ cial issue on the best of the munications are now being used June 2017 special issue dis- IEEE’s Research on Equity in a variety of objects, both fi xed cusses usable security for com- and Sustained Participation and mobile. This makes the tech- puting systems, including the in Engineering, Computing, nology important to the Internet challenges and opportunities it and Technology (RESPECT) of Things (IoT), as addressed by presents. Articles in the issue 2016 conference touch on Computer’s June 2017 special cover topics such as the bal- these diff erent types of diversity issue on VLSI for the IoT. ancing of security and usability and the challenges that exist in

4 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE broadening the pipeline for tech- IEEE Intelligent Systems of a historical group whose mem- nology jobs. bers’ biographies might be largely The amount of Internet of Things untraceable. IEEE Security & Privacy (IoT) data is growing rapidly, but there are important challenges IEEE Pervasive Computing IEEE S&P’s March/April 2017 spe- regarding the quality of this informa- cial issue features selected articles tion and the applications that handle Today’s data revolution is provid- from the IEEE Symposium on it. “IoT Quality Control for Data ing new insights, allowing us to Security and Privacy. The articles and Application Needs,” from IEEE better understand the public’s address vulnerabilities in smart Intelligent Systems’ March/April expectations of smart buildings devices and possible mitigation 2017 issue, discusses these prob- and cities. The articles in IEEE strategies, human behavior in man- lems, as well as potential solutions. Pervasive Computing’s April–June aging devices, code development 2017 special issue explore what for security and privacy, and criti- IEEE MultiMedia infrastructure is necessary to sup- cal infrastructure management. port smart environments, how IEEE MultiMedia’s April–June 2017 cities can measure their return IEEE Cloud Computing special issue is the result of a collab- on these investments, and how orative initiative between the mag- people should interact with these Fog computing, in essence, azine and the IEEE International environments. extends cloud computing to an Symposium on Multimedia (ISM). organization’s network edge, The six theme articles—which cover IT Professional thereby improving the operation of a range of topics on multimedia computing, storage, and network- capturing, mining, and stream- IT Pro’s March/April 2017 spe- ing services between user devices ing—are extended versions of cial issue on IT for a smarter and cloud-based datacenters. The some of ISM 2016’s top papers. world includes articles on oppor- four articles in IEEE Cloud Com- tunistic wireless networking for puting’s March/April 2017 special IEEE Annals of the History of smart dairy farming and quality- issue show that the technology Computing of-service-aware autonomic cloud presents many opportunities and computing. challenges. Published scholarship about race and computing has lagged IEEE Micro IEEE Computer Graphics and because of the absence of archi- Applications val source materials. The author IEEE Micro’s March/April 2017 spe- of “Race and Computing: The cial issue includes papers from Computer graphics research Problem of Sources, the Poten- 2016’s IEEE Hot Chips 28 sympo- is increasingly focused on the tial of Prosopography, and the sium on high-performance chips. high-level analysis and process- Lesson of Ebony Magazine,” from ing of geometric objects, which IEEE Annals’ January–March 2017 Computing Now could help automate some types issue, says archives and libraries of design. The articles in CG&A’s should identify, collect, and make The Computing Now website May/June 2017 special issue con- available relevant materials from (computingnow.computer.org) nect the fabrication process with people of color. The author says features up-to-the-minute com- design- and processing-related scholars could fi nd material in puting news and blogs, along with tasks in new domains including nontraditional sources and might articles ranging from peer-reviewed circuit development, geospatial try prosopography—the investiga- research to opinion pieces by indus- visualization, and 3D scanning. tion of the common characteristics try leaders. www.computer.org/computingedge 5 Move Your Career Forward IEEE Computer Society Membership

Explore These Resources on Hot Topics and Emerging Technologies

IEEE Internet Computing With insightful columns and peer-reviewed feature articles on all aspects of Internet computing—from programming and standards to security and networking—IEEE Internet Computing is the top choice for industry and academic readers.

Rock Stars of Deep Learning/Machine Learning 12 September 2017 Mountain View, CA, USA Join us in Silicon Valley with Fortune 100 C-suite and upper DEEP LEARNING/ management executives and others to learn about deep learning and MACHINE LEARNING machine learning opportunities for 2017. Hear insights on how they use analytics, algorithms, and data structures, and learn how to trust semisupervised or unsupervised analysis and how to contract for or build your deep learning talent pool.

IEEE Big Data Initiative Big data is much more than just data bits and bytes on one side and processing on the other. The scope and direction of IEEE’s new Big Data Initiative is being driven by ongoing efforts to capture all the important perspectives. Join the in-depth discussion.

FOR DIRECT LINKS TO THESE RESOURCES, VISIT The Community for Technology Leaders www.computer.org/edge-resources SPOTLIGHT ON TRANSACTIONS

SPOTLIGHT ON TRANSACTIONS

used to drive conditional encoding A High-level Language logic (for example, to highlight only selected points), parameterize scale mappings (for example, an interval for Interactive might represent a scale domain to be panned and zoomed), and filter data (for example, to perform brushing and Data Visualization linking across multiple views). Users need only specify the desired selection type’s semantics: the VegaLite com Leila De Floriani, University of Maryland, College Park piler automatically synthesizes appro priate eventhandling logic, producing This installment highlighting the work published a complete specification as output. The video at www.computer.org/computer in IEEE Computer Society journals comes from magazine/category/multimedia pro vides a more detailed description of IEEE Transactions on Visualization and Computer Graphics. VegaLite.

or visualization tools to sup query widgets. For custom direct port exploratory data analy manipulation interaction, they must atyanarayan and his coauthors sis, they must allow analysts turn to imperative eventha ndling call received the Best Paper Award to rapidly create and refine vi backs, a stark departure from declara at the 2016 IEEE Information Fsualizations. Some current systems let tive visualencoding grammars. Not SVisualization Conference. VegaLite is developers build custom interactive only does this representational switch now being used to create new visual visualizations for the web, but the time make interactions difficult to specify, analysis systems (including the Voy and effort required to use these tools it also impedes automated reasoning ager visualization recommendation can break the flow of an ad hoc analy over the design space of interactive browser) and enable rapid visualiza sis session. In contrast, tools such charts. tion in datascience environments as Tableau (www.tableau.com) and In “VegaLite: A Grammar of In such as Jupyter Notebooks (using, for ggplot2 (a visualization package for teractive Graphics,” Arvind Satya example, the Altair Python API). the R statistical computing language; narayan, Dominik Moritz, Kanit ggplot2.org) support iterative and in Wongsuphasawat, and Jeffrey Heer teractive construction of graphics us introduce new formalisms for inter LEILA DE FLORIANI is a professor ing a highlevel grammar for concise action techniques and multichart at the University of Maryland and specification. composition (IEEE Trans. Visualization Editor in Chief of IEEE Transactions Existing grammars of graphics are and Computer Graphics, vol. 23, no. 1, on Visualization and Computer effective for creating static charts but pp. 341–350). VegaLite (vega.github Graphics. Contact her at ldf4tvcg@ provide little if any support for con .io/vegalite) extends the traditional umiacs.umd.edu. structing interactive visualizations. grammars of visualencoding rules Analysts can, at most, enable a pre by adding a composition algebra for defined set of common techniques— layered or multiview displays and a linked selections, panning and zoom grammar of interaction. Interactions Visit www.computer.org /computer-multimedia ing, and so on—or para meterize are specified as “selections” over data to learn more about Vega-Lite. their visualization with dynamic points or intervals, which can then be

Hot Topics and Emerging Technologies

he fi elds of computing and communi- fi ve areas related to pervasive-health approaches cations technology are highly dynamic. that researchers will have to address. T Researchers in these areas are con- Some companies are realizing that they need a stantly developing new technologies and tech- new type of executive—the chief data offi cer—to niques, which makes it diffi cult for many people to take responsibility for making the most of the ava- keep up. With this in mind, June’s ComputingEdge lanche of data that customers, suppliers, and others issue examines today’s hot topics and emerging are generating, notes the author of IT Professional’s trends in technology. “The Evolving Role of the CDO.” As IT and consumer-electronics technologies are Platform-as-a-service (PaaS) off erings vary increasingly used in vehicles, they will require adap- greatly among service providers, which has caused tation, according to IEEE Software’s “Future Auto- marketplace confusion and limited the technology’s motive Architecture and the Impact of IT Trends.” adoption, according to “PaaS Death Watch?” from Fog computing will be integral in developing IEEE Cloud Computing. and sustaining smart living, note the authors of IEEE Internet Computing’s “Predictability or “Fog Computing for Smart Living,” from Computer. Early Warning: Using Social Media in Modern The author of IEEE Security & Privacy’s “The Emergency Response” discusses how social media Internet of Things Will Upend Our Industry” says can and can’t help cope with disasters. that as the Internet of Things turns everything into The author of “IT and Future Unemployment: a computer, computer security will become the Part 2,” from IT Professional, describes the rela- security of everything. tionship between long-term, automation-caused The US Department of Energy (DOE) and the unemployment and short-term, off shoring-caused National Cancer Institute will use DOE supercom- unemployment. puters to help fi ght cancer by building sophisti- Educating university software-engineering stu- cated models based on population, patient, and dents in global software engineering is important molecular data, explains “Cancer’s Big Data Prob- because the practice is becoming common, explains lem,” from Computing in Science & Engineering. IEEE Software’s “Preparing Tomorrow’s Software IEEE MultiMedia’s “From Artifi cial Intelli- Engineers for Work in a Global Environment.” gence to Augmented Intelligence” considers AI’s “Incorporating Sustainability into Computing progress and remaining challenges, as well as its Education,” from IEEE Intelligent Systems, pro- potential to augment human performance. vides a survey of the types of college resources and The authors of “The Future of Pervasive courses that integrate sustainability into computer- Health,” from IEEE Pervasive Computing, discuss science curricula.

Future Automotive Archite cture and the Impact of IT Trends

Matthias Traub, Alexander Maier, and Kai L. Barbehön

Architectures for automotive electronics are quickly changing. Energy ef ciency is evolving the classic powertrain toward high- voltage hybrid and electric engines. Autonomous driving demands multisensor fusion away from functionally isolated control units. Connectivity and infotainment have transformed the car into a distributed IT system with cloud access; over-the-air functional upgrades; and high-bandwidth access to map services, media content, other vehicles, and the surrounding infrastructure. Here, Matthias Traub, Alexander Maier, and Kai Barbehön of BMW outline the evolution of automotive electronics and IT. I look forward to hearing from both readers and prospective column authors. —Christof Ebert

DIGITALIZATION IS CREATING op- integration of different manufacturers’ portunities that will change our under- software, in-house software develop- standing of mobility. The automobile ment, and security will become critical as part of the networked world will not success factors, as will higher data rates, only offer new possibilities for inform- support for various data formats, gate- ing and entertaining customers but also way concepts, fail-operational support, lead the way by increasing the automa- and energy ef ciency. tion of comfort and assistance functions, Here, we discuss the major opportu- moving toward autonomous vehicles. nities resulting from transferring IT and These innovations will require an consumer-electronics technologies to the ever-faster increase in electronics, soft- automotive domain. For each opportu- ware, and communication capabilities. nity, we show the current status and the Both the physical architecture and the bene ts it could provide. functional electrical and electronic (E/E) architecture will be the keys to manag- Opportunities for ing the growing complexity. Software the Automotive Industry quality, remote software updates, the Players from the digital world (for example,

Microcontrollers Microcontrollers Systems on chip Computing platform A solution box made of AUTOSAR • homogeneous, scalable hardware platforms and Standard • a common software core GENIVI software platform including basic and system functions Linux

… 2008 2015 2018 2021 …

FIGURE 1. High-performance processors from the consumer-electronics world will enable new integration platforms in automotive- electronics architectures. AUTOSAR = Automotive Open System Architecture.

Google, Apple, and Amazon) are electronics architectures. These pro- manner with a focus on local solu- showing us opportunities and solu- cessors’ use has introduced powerful, tions. That meant four things: tions that we can adapt to the auto- dynamic OSs to the automotive do- motive domain. main—for example, Linux (see Fig- • The partitioning of functions One approach to such solutions is ure 1). Such OSs are used particu- focused on where computing service-oriented architecture (SOA), larly for multimedia and navigation. power was available. including the OS (for example, Ap- The GENIVI Alliance (www • Project-specific and heteroge- ple’s iOS). This approach lets devel- .genivi.org) is working to establish neous development methods opers create efficient new functions a standard for software architec- were common in each E/E devel- (apps) that can be easily integrated in ture, including a dynamic OS. In the opment area. the whole ecosystem of a device (for near future, electronic control units • All ECUs used the general sys- example, the iPhone). This seamless (ECUs) for other applications (for tem requirements of an original integration of new functions can help example, autonomous driving) will equipment manager (OEM), provide personalization for each user. require such software architectures. which often led to overdesigned Furthermore, remote updates enable So, the AUTOSAR (Automotive Open ECUs. optimization, quality enhancements, System Architecture; www.autosar • ECU development focused on and flexible lifecycle management. .org) standard will be enhanced, us- the ECU’s local optima owing to Other opportunities come from the ing the POSIX (Portable Operating strictly separated responsibilities use of high-performance processors, System Interface) standard, to create and objectives. which are already used in IT, and the AUTOSAR Adaptive Platform.1 clear design patterns such as hierar- These new technologies will enable To address these issues, BMW chization and scalability. new ways to structure and design created a hierarchical E/E architec- E/E architectures. ture for the next generation of cars High-Performance Processors (see Figure 2). This approach has Enable Higher Integration A Seamless these benefits: High-performance consumer-electronics Hierarchical Architecture processors are enabling new inte- Until now, E/E architectures were • ECU classification is gration platforms in automotive- designed mostly in an evolutionary requirements-based.

1028 IEEE SOFTWAREComputing | Edge WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE June 2017 SOFTWARE TECHNOLOGY

• Adapted and unified development methods for each ECU Central computing Communication platforms class replace local development ECU ECU Actuator server methods. Differentiation • Each ECU class has specific sys- ECU ECU ECU Integration tem requirements. potential ECUs • System-level optimization is the Sensor Sensor ECU Commodity main focus, driven by the system ECUs, industry standard architects. Sensor ECU Sensor Sensors and actuators In this architecture, the central Actuator Actuator ECU computing platforms (class 1—the top level of Figure 2) partition the main software functions, which are FIGURE 2. Powerful integration platforms enable a seamless hierarchical electrical and developed mainly in-house. These electronic (E/E) architecture for the automotive domain. ECU = electronic control unit. platforms offer high performance and fulfill the highest security and safety requirements. The integration Central communication server ECUs (class 2) close the gap between the central computing platforms and Back-end Diagnostic Logging Manipulation or interface interface interface proxy commodity ECUs (class 3)—for example, for deploying time-critical functions requiring direct access to Central information server and broker for all network architecture information sensors or actuators. For simple and non-OEM-specific functions, commodity ECUs and sensors and actua- Controller Area Local Ethernet FlexRay tors (class 4) are acceptable. Ideally, Network Interconnect front end front end front end Network front end these ECUs and the sensors or actuators are based on common OEM or Tier 1 parts.

A Scalable Architecture Today’s network architectures are developed partly at the cost of in- Network architecture with all ECUs creased variance and complexity. The design paradigms are often based FIGURE 3. A central communication server will enable a scalable E/E architecture. on sender–receiver communication, which frequently leads to high dependencies and limited extensibility. • The E/E architecture will be en- ler Area Network), FlexRay, For the next generation of net- capsulated and scaled with less Ethernet, and the evolving work architectures, a central com- variance from a basic vehicle up machine-to-machine and munication server (CCS) will support to a fully equipped vehicle. vehicle-to-everything wireless encapsulation to handle extensibility • Network architecture commu- protocols. locally. Such architectures will have nication will be structured and • The physical, information, and these main features (see Figure 3): hierarchical. service levels will be separate, • Conversion will be fluid be- including firewall features. • A central information server and tween communication protocols a broker will handle all network such as LIN (Local Intercon- At the system level, the CCS will architecture information. net Network), CAN (Control- offer optimization opportunities for www.computer.org/computingedge MAY/JUNE 2017 | IEEE SOFTWARE 29 11 SOFTWARE TECHNOLOGY

Service-oriented architecture used in the automotive eld

Application layer Framework ... Brake Side-view ... assistant assistant ...

Scene Scene Application services Extended interpretation 1 interpretation 2 Framework A Framework B Framework C Extended services Detailed view layer Road model and localization Basic services OdometryObject fusion ... Fusion of sensor data … … Data processing of sensors

Basic layer

Principal work ow to build a service-oriented architecture

Design of services Functional grouping of services Mapping of services

Denition of services including their Mapping of services to frameworks and Mapping of services to software components interfaces categorization of services and of service interfaces to software interfaces

FIGURE 4. A service-oriented architecture will help BMW master the increasing complexity of E/E function networks.

both the physical and logical topol- ECU level won’t be enough to master can be transferred to the automo- ogy. For example, at the physical the complexity.2 tive domain. level, a subgateway could be re- So, we introduced an SOA ap- Connectivity between the in- placed by the high-performance proach at BMW for a next-generation vehicle network and back end is im- routing engine of the CCS. For op- E/E architecture (see Figure 4). An portant. The main aspects are timization at the logical level, the SOA provides substantial abstracted CCS could decouple the sender– services for the overall system. A • communication standards such receiver dependencies and create the stringent encapsulation and hierar- as LTE (Long-Term Evolution), basis for the stepwise introduction chy enable testing against interfaces Wi-Fi, and 5G in the future; of an SOA. and using agile methods, and they • the provision of high-quality reduce system complexity. Reuse of service; and A Service-Oriented software components between ve- • the availability of sufficient Architecture hicle generations will be significantly bandwidth. Automotive functions must be at the simpler. same level of innovation as func- The back end provides the basis for tions from the IT and consumer- The In-Vehicle various applications. Functions can electronics industries. Furthermore, and Back-End Architectures execute there (for example, for au- automotive functions are becom- The in-vehicle E/E architecture and tonomous driving or multimedia). ing increasingly complex owing to the back-end architecture are mov- Mobility services for the driver can greater dependencies. One example ing closer together. More and more be provided over the back end. Infor- is the evolution from simple cruise automotive functions exchange data mation transferred from the car to control, to active cruise control, to with the back end or execute partly the back end can be analyzed. autopilots. Today, the functional on the back end. So, aspects such The seamless coupling of the in- view of a system is signal-based. as functional separation, encoding, vehicle and back-end architectures, The partitioning of functions is sig- and firewalls are becoming increas- along with the building blocks we nificantly affected by legacy systems, ingly important. These security mentioned before (CCS, SOA, and ECU resources, and the organization patterns already exist in the IT and the hierarchical E/E architecture), building the system. Focusing on the consumer-electronics industries and will constitute a big step toward en-

1230 IEEE SOFTWAREComputing | Edge WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE June 2017 SOFTWARE TECHNOLOGY SOFTWARE TECHNOLOGY

Service-oriented architecture used in the automotive eld Back end Car Application layer Framework ... BMW Brake Side-view ...... assistant assistant AF CD Scene Scene Application services Extended back end back end interpretation 1 interpretation 2 Framework A Framework B Framework C Extended services Detailed view layer Road model and localization 5G Basic services OdometryObject fusion ... WWW Fusion of sensor data External … … GSM Data processing of sensors partners

Basic layer Mobility services Big data and services Principal work ow to build a service-oriented architecture analytics

Design of services Functional grouping of services Mapping of services

FIGURE 4. A service-oriented architecture will help BMW master the increasing complexity of E/E function networks. FIGURE 5. This overall perspective of the in-vehicle and back-end architectures shows that mechanisms such as those for security and privacy must handle both architectures to offer seamless functioning. AF = autonomous-driving functions, CD = crowd data, GSM = Global System for Mobile Communication, NFC = near-field communication, and OBD = onboard diagnostics. both the physical and logical topol- ECU level won’t be enough to master can be transferred to the automo- ogy. For example, at the physical the complexity.2 tive domain. level, a subgateway could be re- So, we introduced an SOA ap- Connectivity between the in- abling our E/E infrastructure to deal current heterogeneous situation, cre- tations, such as seamless remote placed by the high-performance proach at BMW for a next-generation vehicle network and back end is im- with the upcoming innovations. Fig- ating a more homogeneous develop- updates of functions; new mar- routing engine of the CCS. For op- E/E architecture (see Figure 4). An portant. The main aspects are ure 5 shows the overall system struc- ment environment. Figure 6 shows ket players, such as Tesla; and the timization at the logical level, the SOA provides substantial abstracted ture and the design of software for the current and future states. This need for new regulations, such as CCS could decouple the sender– services for the overall system. A • communication standards such functions that can execute on an technological evolution will provide laws covering liability in autono- receiver dependencies and create the stringent encapsulation and hierar- as LTE (Long-Term Evolution), ECU or the back end. mous driving. basis for the stepwise introduction chy enable testing against interfaces Wi-Fi, and 5G in the future; Seamless design and documenta- • a seamless requirements process Automotive-electronics devel- of an SOA. and using agile methods, and they • the provision of high-quality tion of the in-vehicle and back-end from customer interaction to the opment will benefit from IT and reduce system complexity. Reuse of service; and architectures are critical success fac- software architecture; consumer- electronics standards but A Service-Oriented software components between ve- • the availability of sufficient tors. This includes data handling, re- • seamless, complete modeling of must reinvent many of the technol- Architecture hicle generations will be significantly bandwidth. mote updates, and software design E/E architectures based on SOA; ogies so that they meet the strin- Automotive functions must be at the simpler. for functions that can execute on • encapsulation of content for dis- gent demands. Automotive elec- same level of innovation as func- The back end provides the basis for an ECU or the back end. We discuss tributed development using SOA tronics will remain different from tions from the IT and consumer- The In-Vehicle various applications. Functions can these aspects in the next section. design principles; classic software systems. The de- electronics industries. Furthermore, and Back-End Architectures execute there (for example, for au- • the launching of an agile devel- mand for safety, performance, us- automotive functions are becom- The in-vehicle E/E architecture and tonomous driving or multimedia). Changes for Processes, opment process, joint Scrum ability, and security leads to the ing increasingly complex owing to the back-end architecture are mov- Mobility services for the driver can Methods, and Tools teams, and shared code reposito- highest expectations of proficiency greater dependencies. One example ing closer together. More and more be provided over the back end. Infor- Currently, the overall E/E develop- ries; and and quality, which in consumer is the evolution from simple cruise automotive functions exchange data mation transferred from the car to ment process uses a multitude of • continuous integration and electronics wouldn’t be considered control, to active cruise control, to with the back end or execute partly the back end can be analyzed. methods and tools. Often, each de- early use of virtual integration necessary. The intelligent adop- autopilots. Today, the functional on the back end. So, aspects such The seamless coupling of the in- partment in a software project has methods. tion and adaptation of IT and view of a system is signal-based. as functional separation, encoding, vehicle and back-end architectures, its own design principles, work- consumer- electronics methods and The partitioning of functions is sig- and firewalls are becoming increas- along with the building blocks we flow, and tooling. In the future, technologies will enable many op- nificantly affected by legacy systems, ingly important. These security mentioned before (CCS, SOA, and state-of-the-art methods and tools igitalization and con- portunities, while avoiding the ECU resources, and the organization patterns already exist in the IT and the hierarchical E/E architecture), from the IT industry for designing nectivity will result in traps that weakened classic soft- building the system. Focusing on the consumer-electronics industries and will constitute a big step toward en- E/E architectures will eliminate the D changed customer expec- ware systems.

30 IEEE SOFTWARE | WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE www.computer.org/computingedge MAY/JUNE 2017 | IEEE SOFTWARE 31 13 SOFTWARE TECHNOLOGY

Design phase Concept Design phase Concept Architecture tooling Offer phase Agreement Offer phase Agreement Function offer E/E architecture A seamless Overall design of the I&C Function speci cation Domain-speci c tooling ISO 26262 testing requirements process service-oriented DA&S Function speci cation Domain-speci c tooling from the customer P+C Function speci cation ISO 26262 ISO 26262 architecture Continuous integration AIDA Prima HIL HIL I&C B+C Function speci cation DOORS, Boardnet Developer, ADD HIL HIL function to the central diagnosis database, and early use of virtual Fibex Ascent , Simulink… SIL SIL DA&SS software architecture integration methods Franca SIL SIL P&C ... Software logistics, test cases, system integration, ... Encapsulation of XML LabelDB ARXML B&C content for distributed PREEvision development

An agile development process, joint Scrum teams, and joint code repositories

A multitude of methods and tools in the overall software development process State-of-the-art methods and tools in the software development process

(a) (b)

FIGURE 6. Automotive-electronics development (a) today and (b) in the future. Seamless design and documentation of the in-vehicle and back-end architectures are critical success factors. ADD = architecture design document, B&C = body and comfort, DA&S = driver assistance and safety, HIL = hardware in the loop, I&C = information and communication, P&C = powertrain and chassis, and SIL = software in the loop.

References /presentations/EUROFORUM Standards Will In uence the Future 1. S. Fürst, “AUTOSAR Adaptive Plat- _Elektronik-Systeme_im_Automobile E/E Architecture], presentation at 8th form for Connected and Autonomous _2016_-_FUERST_Simon.pdf. Vector Congress, 2016 (in German); Vehicles,” presentation at EURO- 2. M. Traub, “Mitten im Umbruch zur vector.com/congress. FORUM Elektronik-Systeme im Digitalisierung: Wie die Zukunft der Automobil [EUROFORUM Elec- E/E Architektur von It-Standards MATTHIAS TRAUB is the chief architect of tronic Systems in Automobiles], 2016; Beein usst Wird” [In the Midst of E/E-Architecture Next and head of the E/E- w w w . a u t o s a r . o r g / l e a d m i n / l e s Change to Digitalization: How IT Architecture platform at BMW. Contact him at [email protected].

ALEXANDER MAIER is the head of BMW’s E/E-Architecture department. Contact him at [email protected].

KAI L. BARBEHÖN is the vice president for E/E and software architecture, platform management at BMW. Contact him at [email protected].

This article originallyRead your appeared subscriptions in EEE Software,through vol. 34, the no. myCS 3, publications portal at 2017. http://mycs.computer.org

1432 IEEE SOFTWAREComputing | Edge WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE June 2017 This article originally appeared in Computer, COMPUTING: THE NEXT 50 YEARS vol. 50, no. 2, 2017.

computing. One of the biggest risks of Fog Computing massively deploying computational devices is that presented by security compromises and privacy leaks. The for Smart Living recent distributed denial-of-service attack against Dyn that left many users unable to access some of the big- Bruce McMillin, Missouri University of Science and Technology gest sites on the web is an example of Tao Zhang, Cisco large numbers of devices being turned against a large infrastructure. Fog computing will be integral in developing Fog computing providers must stop looking at security as hierarchical lay- and sustaining smart living. ers, but rather as multiple peer domains with often limited resources in potentially hostile environments. Personal he World Health Organiza- to be performed in close proximity to privacy is also at risk. Fog computing tion projects that the number the things it controls. brings resources closer to end users and of people living in cities will Over time, fog and cloud computing might enable ner-grained monitoring nearly double over the next will converge into uni ed end-to-end of human behavior, which could expose Tfew decades, so urban centers need to platforms o ering integrated services a higher degree of private information provide more sustainable solutions and applications along the continuum ranging from utility and transporta- for smart living. New technologies— from the cloud to things. Applications tion usage to personal health. such as materials, sensors, wireless developed and deployed for the cloud communications, and controls—will will be able to run in fog and vice versa. be necessary to manage living envi- og systems can provide a range of ronments that proactively sense be- OPEN ARCHITECTURE new security services that would havioral and health risks and provide An open architecture is needed to de- be di cult for clouds alone to situation- aware responses t o emer- velop a smart living environment with Fprovide. This “closeness” will help gencies or disasters. potentially thousands of vendors. protect the large number of diverse In addition, utility and transporta- The OpenFog Consortium—a global and resource-constrained endpoints. tion networks must adapt to dynamic ecosystem formed by ARM, Cisco, Regard less of approach, no comput- usage, tra c conditions, and user Dell, Intel, Microsoft, and Princeton ing architecture will succeed in our behavior with a minimal carbon foot- University’s EDGE Lab to accelerate increasingly digital world unless it’s print; a clean and renewable energy the adoption of fog computing (www open and secure. The future of the hu- grid must actuate localized energy and .openfogconsortium.org)—is taking man experience depends on it. power control; and pervasive security steps to create a common reference is needed to detect and prevent poten- architecture covering hardware and tial threats. This vision is bold but crit- software platforms and highly so- ical to enabling smart living. phisticated capabilities. BRUCE MCMILLIN is a professor Cloud-only models face serious The distributed algorithms that run of computer science and associate challenges in latency, network band- on the devices that receive and act on lo- dean of engineering and computing width, geographic focus, reliability, cally sensed information are key to de- at Missouri University of Science and security. Fog computing reduces veloping fog computing. To meet diverse and Technology. Contact him at @ these challenges by providing a computing and communications needs, mst.edu. system- level horizontal architecture the fog must be semantically aware, un- to distribute computing, storage, con- derstanding the physics of power, trans- TAO ZHANG is a Cisco Distinguished trol, and networking resources and portation, and environment along with Engineer, a cofounder of the services from the cloud to connected psychosocial dynamics. OpenFog Consortium, and CIO of devices (“things”). Think of fog com- the IEEE Communications Society. puting as the cloud on the ground: it SECURITY AND PRIVACY Contact him at [email protected]. enables latency-sensitive computing Stringent security is critical for fog

The Internet of Things Will Upend Our Industry

verything is becoming a computer. Your alone by governments because we’ve been E microwave is a computer that makes things providing a steady stream of amazing techno- hot. Your refrigerator is a computer that keeps logical innovations that largely didn’t affect things cold. Your smartphone is a portable com- the world in a direct physical manner. That’s puter that makes phone calls. Your car is a dis- about to change in a big way. Even though it’s tributed system with more than 100 computers the same OS and maybe even the same vul- plus four wheels and an engine. More alarm- nerability, there’s a fundamental difference ingly, a nuclear power plant is a computer that between losing your data when a spreadsheet produces energy. This is happening at all levels crashes and losing your life when a car crashes. of our lives and all over the world. Once governments start regulating, they ©Julian Dodd As everything turns into a computer, tend to have a limited toolbox. They can reg- computer security becomes everything secu- ulate before the fact (product and category Bruce Schneier rity. This will upend the IT security industry, rules, licensing, and testing requirements) Harvard University because our knowledge and experience with and after the fact (fines and liabilities). They computer security will be much more broadly can mandate disclosure (product labeling, applicable, and the restrictions and regula- testing, and forensics agencies). They can also tions from the physical world will be applied shape the environment by funding research to the computer world. The beachhead for all and education or using their procurement to of this is the Internet of Things (IoT), which I drive improvement. What will work, and how, liken to a world-sized robot—one that can kill is unclear—and fraught with fear and anxiety. people and destroy property. But it’s a conversation we must have. When speaking to more general audi- What will it take? Automobile ransom- ences, I list five truisms from our world that ware at speed? A distributed denial-of-service are coming to theirs. One: most software is attack that takes out a power plant? Murder via poorly written and insecure—as an industry, a medical device? IoT sensors used to identify we’ve prioritized speed and price over qual- and arrest a minority group? None of these ity, and security has suffered as a result. Two, scenarios is science fiction, and many more of three, and four: the extensibility, complex- these real-world risks loom in the near future. ity, and interconnectedness of computerized Our choice is no longer one of government networked systems make them very hard to regulation versus no government regulation. secure. Five: the vulnerabilities and failure Our choice is smarter government regulation modes of computers are different from those versus stupider government regulation. As an of the more manual systems they replace. industry, we have to start thinking about the We know about how computer attacks pros and cons of the different approaches so scale, class breaks, and the dangers of software we’re not surprised when something is forced monoculture. But this comes as a complete sur- on us. We’re no longer in our separate IT world. prise to others: from auto and medical-device We’re part of the real world, and that world is manufacturers to the companies making part of us. We need to start acting that way. low-cost IoT devices like DVRs, webcams, and toys. The cybersecurity skills shortage of today Bruce Schneier is a security technologist, a This article originally is only going to get worse as these devices’ inse- Fellow at the Berkman Klein Center for appeared in curities start causing real-world damage. Internet and Society at Harvard University, IEEE Security & Privacy, Fear of that damage will cause the next big chief technology officer of IBM Resilient, vol. 49, no. 9, 2016. change to our industry: government regula- and Special Advisor to IBM Security. Con- tion. The IT industry has mostly been left tact him via www.schneier.com.

16 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE 108 March/April 2017 Copublished by the IEEE Computer and Reliability Societies 1540-7993/17/$33.00 © 2017 IEEE L eadershiSectionp Computing title Editors: James J. Hack, [email protected] | Michael E. Papka, [email protected]

Cancer’s Big Data Problem

Justin H.S. Breaux | Argonne National Laboratory

ata is pouring into the hands of cancer researchers, computational models to facilitate breakthroughs in the fight thanks to improvements in imaging, models, and against cancer on the molecular, patient, and population understanding of genetics. Today, the data from a levels. D single patient’s tumor in a clinical trial can add up The pilot builds on US President Barack Obama’s Pre- to one Tbyte, the equivalent of 130,000 books. cision Medicine Initiative and Vice President Joe Biden’s But we don’t yet have the tools to efficiently process the recent “Cancer Moonshot” to transition cancer therapy mountain of genetic data to make more precise predictions away from a “one size fits all” approach. Instead, the goal for therapy. And it’s needed: treating cancer remains a com- is to move toward individualized diagnosis and treatment plex moving target. We can’t yet say precisely how a specific that accommodates a patient’s unique body chemistry and tumor will react to any given drug, and as a patient is treated, genetics. cancer cells can continue to evolve, making the initial therapy “Cancer researchers are very good at generating all less effective. types of data, from genomic data, proteomic data, and im- Toward this goal, the US Department of Energy (DOE) aging data,” said Warren Kibbe, director of the Center for is partnering with the National Cancer Institute in an “all Biomedical Informatics and Information Technology at the government” approach to fighting cancer. Part of this part- National Cancer Institute. “What we’re not really good at nership is a three-year pilot project called the Joint Design yet is integrating all that information into a consistent model of Advanced Computing Solutions for Cancer (JDACSC), and making predictions on how a tumor will respond to a which will use DOE supercomputing to build sophisticated given treatment.”

2469-7087/17/$33.00 © 2017 IEEE Published by the IEEE Computer Society June 2017 17 March/April 2017 Copublished by the IEEE CS and the AIP 1521-9615/17/$33.00 © 2017 IEEE Computing in Science & Engineering 79 Leadership Computing

To date, machine learning studies have produced computational models that estimate drug response for a singular data point—say, a certain mutation.

CANDLE level pilot, Argonne leads the patient-level pilot, Key to this collaboration is a computational frame- Oak Ridge undertakes the population-level pilot, work called the CANcer Distributed Learning En- and Los Alamos looks at uncertainty quantifica- vironment (CANDLE). Over the years, many proj- tion across all three pilots. ects have amassed a huge volume of cancer data, including tumor genomes, patient data, and experi- Molecular Level ments on potential drugs. CANDLE is designed to Thirty percent of all cancers exhibit mutations in use machine learning algorithms to find patterns the Ras family, a collection of proteins that help in large datasets. Machine learning is a type of ar- trigger cellular machinery to make new cells or kill tificial intelligence that focuses on developing pro- old ones. grams that can teach themselves to grow and change The molecular-level pilot work being led by a when presented with new data. These patterns offer team at Lawrence Livermore will use the CANDLE insights that could ultimately result in improved architecture to predict how these proteins behave treatment or guidance on new experiments. on the top of cell membranes. The team will then To date, machine learning studies have pro- apply this knowledge to what researchers describe duced computational models that estimate drug as the “Ras pathway problem,” where glitches cause response for a singular data point—say, a certain genes to remain stuck in the “on” position, leading mutation. Researchers working with CANDLE, to cancerous tumors. however, envision a higher degree of complexity Researchers want to produce highly complex that integrates many types of information, such simulations that describe how a protein moves and as drug interactions and specifics about a patient’s binds to specific locations on the cell membrane. genealogy, as well as the tumor’s molecular charac- They hope such insights can be applied to the mil- teristics and how its protein expression varies over lions of Ras pathways and dramatically enhance time. our understanding of how they work by predicting “The research community has collected thou- the likelihood that a signal will take a certain path. sands of experiments with hundreds of thousands of data points characterizing tumors and their Patient Level response to the drugs,” said Rick Stevens, an as- Cancer encompasses hundreds of diseases, each sociate laboratory director at the DOE’s Argonne with thousands of possible causes. Thus, bringing National Laboratory and professor of computer precision to therapy selection for a specific patient science at the University of Chicago. “By working is the goal of the Argonne-led patient-level pilot. with the national laboratories, the National Can- With the CANDLE platform, researchers at cer Institute can now finally scale and quantify the the Argonne Leadership Computing Facility, a cancer problem.” DOE Office of Science User facility, will develop Using this computational architecture, partici- predictive models that guide drug treatment choices pating labs—Argonne, Lawrence Livermore, Los for tumors based on a much wider assortment of Alamos, and Oak Ridge national laboratories— data than is currently used. will focus on three problems singled out by the Na- To do this, they’ll merge one type of compu- tional Cancer Institute as the biggest bottlenecks tational model that uses data to predict phenom- to advancing cancer research. The three pilot proj- ena with another model that uses data to explain ects are aimed at understanding key protein inter- them. The hope is that by merging these two meth- actions, predicting drug response, and automating ods, they’ll be able to migrate lessons learned from patient information extraction to inform treatment computer simulations to the research laboratory, strategies. where researchers test mice to verify the computer’s Each pilot looks at cancer from a different prediction of how a tumor will respond to a given scale. Lawrence Livermore drives the molecular- therapy.

18 Computing Edge June 2017 80 March/April 2017 Leadership Computing

To date, machine learning studies have produced Researchers will also try to find mechanisms brand new tools to enable understanding of these computational models that estimate drug response for a for how a particular tumor evades a therapy or de- basic processes,” said Kibbe. “And if we can real- velops resistance. ly understand the interplay of mutations, normal singular data point—say, a certain mutation. “Conceptually, that’s how we’re thinking the biological processes, and cancer, we have a much future of cancer therapies is going to move. Right better chance of being able to interfere with—and now, we don’t understand the biological implica- end—cancer.” tions of resistance well enough for any particular CANDLE level pilot, Argonne leads the patient-level pilot, therapy to do a good job at predicting combinato- Key to this collaboration is a computational frame- Oak Ridge undertakes the population-level pilot, rial therapies,” said Kibbe. “We think that simula- s the pilots progress from building and merg- work called the CANcer Distributed Learning En- and Los Alamos looks at uncertainty quantifica- tion will allow us to do a much better job of pre- A ing computational models to testing them vironment (CANDLE). Over the years, many proj- tion across all three pilots. dicting which combination of therapies would be in the laboratory, Stevens admits that you’d have ects have amassed a huge volume of cancer data, most effective for a specific patient.” to be a little fearless to go after a problem like including tumor genomes, patient data, and experi- Molecular Level this. ments on potential drugs. CANDLE is designed to Thirty percent of all cancers exhibit mutations in Population Level “In my nearly 20 years of working in compu- use machine learning algorithms to find patterns the Ras family, a collection of proteins that help At any one point in time, 3 to 5 percent of patients tational biology, I can say that this is a really hard in large datasets. Machine learning is a type of ar- trigger cellular machinery to make new cells or kill with cancer participate in a cancer clinical trial, problem and it’s not clear if we know how to do tificial intelligence that focuses on developing pro- old ones. but cataloguing all this data is still a manual task. this,” said Stevens. “But what the Cancer Moon- grams that can teach themselves to grow and change The molecular-level pilot work being led by a Oak Ridge will help the National Cancer In- shot gives us all is the ability to show the world when presented with new data. These patterns offer team at Lawrence Livermore will use the CANDLE stitute scale its ability to monitor cancer patients how DOE labs can work in collaboration with the insights that could ultimately result in improved architecture to predict how these proteins behave across the country by automating the process of National Cancer Institute in a way that hasn’t been treatment or guidance on new experiments. on the top of cell membranes. The team will then entering and extracting information. By apply- done before.” To date, machine learning studies have pro- apply this knowledge to what researchers describe ing natural language processing and machine “With that level of collaboration, it starts to look duced computational models that estimate drug as the “Ras pathway problem,” where glitches cause learning algorithms to these millions of clinical like less of a far-off moonshot,” he said, “and more response for a singular data point—say, a certain genes to remain stuck in the “on” position, leading reports, computers will be able to derive meaning a problem that we have a real shot at addressing.” mutation. Researchers working with CANDLE, to cancerous tumors. from the notes that doctors and nurses write in however, envision a higher degree of complexity Researchers want to produce highly complex their reports. Acknowledgments that integrates many types of information, such simulations that describe how a protein moves and Once completed, this system will automatically Support for the initiative is provided by the US De- as drug interactions and specifics about a patient’s binds to specific locations on the cell membrane. analyze and extract information so that research- partment of Energy’s Office of Science, the National genealogy, as well as the tumor’s molecular charac- They hope such insights can be applied to the mil- ers can monitor country-wide outcomes, which Institutes of Health, and the National Nuclear Secu- teristics and how its protein expression varies over lions of Ras pathways and dramatically enhance can then inform treatment strategies for patients rity Administration. The Argonne Leadership Com- time. our understanding of how they work by predicting of different lifestyles, environments, and cancer puting Facility is a DOE Office of Science User Facil- “The research community has collected thou- the likelihood that a signal will take a certain path. types. ity supported under Contract DE-AC02-06CH11357. sands of experiments with hundreds of thousands Steps are being taken to de-identify data before Image courtesy of Argonne National Laboratory/Scott of data points characterizing tumors and their Patient Level population-level pilot data is shared with partici- Nychay. response to the drugs,” said Rick Stevens, an as- Cancer encompasses hundreds of diseases, each pating labs, Stevens said. sociate laboratory director at the DOE’s Argonne with thousands of possible causes. Thus, bringing Justin H.S. Breaux is a science writer and social media National Laboratory and professor of computer precision to therapy selection for a specific patient Next Steps specialist at Argonne National Laboratory. His interests science at the University of Chicago. “By working is the goal of the Argonne-led patient-level pilot. Over the next three years, both the National Can- include communicating difficult topics to unsuspecting with the national laboratories, the National Can- With the CANDLE platform, researchers at cer Institute and DOE have a monumental task, audiences, figuring out how to make a video go viral, cer Institute can now finally scale and quantify the the Argonne Leadership Computing Facility, a but they have a plan. and learning Spanish. Breaux received a BA in liberal cancer problem.” DOE Office of Science User facility, will develop The first year will focus on merging statistical arts and sciences with a major in philosophy from the Using this computational architecture, partici- predictive models that guide drug treatment choices models and building machine learning methods University of Illinois at Chicago and a BA in televi- pating labs—Argonne, Lawrence Livermore, Los for tumors based on a much wider assortment of that make the best of their ability to explain and sion from Columbia College Chicago. Contact him at Alamos, and Oak Ridge national laboratories— data than is currently used. predict phenomena. In the second year, computer [email protected]. will focus on three problems singled out by the Na- To do this, they’ll merge one type of compu- scientists will have to computationally estimate tional Cancer Institute as the biggest bottlenecks tational model that uses data to predict phenom- how confident they are in those predictions, to advancing cancer research. The three pilot proj- ena with another model that uses data to explain and in the final year, researchers will put all ects are aimed at understanding key protein inter- them. The hope is that by merging these two meth- these pieces together and integrate experimental actions, predicting drug response, and automating ods, they’ll be able to migrate lessons learned from design. patient information extraction to inform treatment computer simulations to the research laboratory, “I really think we are at a unique place right This article originallyRead your appeared subscriptions in through the strategies. where researchers test mice to verify the computer’s now. There are some unbelievably great conver- Computing in SciencemyCS publications & Engineering, portal at http:// Each pilot looks at cancer from a different prediction of how a tumor will respond to a given sations happening across government right now vol. 19, no. 2, 2017.mycs.computer.org. scale. Lawrence Livermore drives the molecular- therapy. about how we work together and integrate these

www.computer.org/computingedge 19 80 March/April 2017 www.computer.org/cise 81 EIC Message

From Artificial Intelligence to Augmented Intelligence

Yong Rui ince the official beginning of artificial intel- proves that autonomous cars have become less Lenovo S ligence back in 1956, AI has experienced jerky over the past few years. several ups and downs. Today, it’s at another Also, although there’s no doubt that autono- peak, fueled by better algorithms, more data, mous driving can significantly assist human and stronger computing power. AI is no longer drivers—especially when a driver is tired—one a topic discussed only by researchers; the of the panelists expressed concerns regarding masses have started to take notice, especially how autonomous cars will handle unforeseen since Google’s AlphaGo program beat Lee situations (that is, those with no training data). Sedol, a human champion of the game Go.1 Interestingly, the roads will likely be safer once The AI victory last March prompted people to all cars are autonomous, as opposed to having consider who is more intelligent, AI or humans. both human-driven and driverless cars, because Even before the victory, I was part of a panel the differing driving behaviors could cause con- at a TV dialogue program, where we discussed fusion and lead to accidents. the achievements of AI. Other panelists included representatives from both academia Human vs. Face Recognition Software and industry. The program, which featured The face is undoubtedly a familiar and impor- competitions between AI and humans, was tant object to humans, so AI’s vision skills are structured to be fun and engaging. Here, I significant to its progress. In the next competi- review the competitions and consider what the tion, live on stage, the participants were asked outcomes say about the state of AI. to state whether or not a certain person was in a predefined database with dozens of other peo- Stenographer vs. Dictation Software ple. The AI did quite well, outperforming the A stenographer competed against the speech human when the dataset pool was small, but dictation software live on site. In a low-noise neither the human nor AI did well when the environment, AI did quite well, with 80 percent dataset was large. accuracy versus the stenographer’s 75 percent accuracy. Although this is great progress, there are still several places where AI needs to I certainly performs better in memory and improve. A computing. For example, while it’s diffi- First of all, AI’s accuracy drops significantly cult for humans to remember 100 digits after when there is an accent. Second, when there is the decimal point of pi, this is trivial for AI. noise, such as at a party, AI experiences signifi- Also, AI computes faster than humans, cant challenges. Finally, AI still has a long way although it consumes more energy. For any task to go in understanding natural language at a with specific rules, such as chess or Go, AI will semantic level on top of speech recognition. always be better than humans—if not today, then in the near future. Human Driver vs. Autonomous Car This memory and computing capability is At the beginning of the next competition, a few mostly an extension of the human’s left brain. small objects were put on top of two cars, one The human’s right brain, which is more artis- car with a human driver and the other autono- tic and imaginative and can handle many mous, to see which car had the smoother ride. uncertainties, is quite different from today’s At the end of the test, the objects remained on AI. In fact, AI and humans have their own both cars, ending the competition in a tie. This strengths and weaknesses. So instead of saying

20 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE 4 1070-986X/17/$33.00 c 2017 IEEE Published by the IEEE Computer Society EIC Message AI still has a long way to go in understanding

natural language at a Recognizing Excellence in High-Performance Computing From Artificial Intelligence to semantic level on top of Nominations are Solicited for the Augmented Intelligence speech recognition. SEYMOUR CRAY SIDNEY FERNBACH Yong Rui ince the official beginning of artificial intel- proves that autonomous cars have become less “humans vs. AI,” it’s more interesting to say ligence back in 1956, AI has experienced jerky over the past few years. “humans AI,” because AI can make humans & KEN KENNEDY AWARDS Lenovo S þ several ups and downs. Today, it’s at another Also, although there’s no doubt that autono- stronger. I, along with many others, call this peak, fueled by better algorithms, more data, mous driving can significantly assist human augmented intelligence. This augmented intelli- and stronger computing power. AI is no longer drivers—especially when a driver is tired—one gence will be pervasive in multimedia. AI will a topic discussed only by researchers; the of the panelists expressed concerns regarding be stronger in some cases of image, speech, masses have started to take notice, especially how autonomous cars will handle unforeseen and video recognition—when the tasks are since Google’s AlphaGo program beat Lee situations (that is, those with no training data). well defined—but humans will be better in Deadline: 1 July 2017 1 Sedol, a human champion of the game Go. Interestingly, the roads will likely be safer once other tasks for years to come. MM All nomination details available at awards.computer.org The AI victory last March prompted people to all cars are autonomous, as opposed to having consider who is more intelligent, AI or humans. both human-driven and driverless cars, because Reference Even before the victory, I was part of a panel the differing driving behaviors could cause con- SEYMOUR CRAY COMPUTER at a TV dialogue program, where we discussed fusion and lead to accidents. 1. C. Moyer, “How Google’s AlphaGo Beat a Go ENGINEERING AWARD the achievements of AI. Other panelists World Champion,” The Atlantic, 28 Mar. 2016; The Seymour Cray Award is awarded to recognize innova- www.theatlantic.com/technology/archive/2016/ tive contributions to high-performance computing systems included representatives from both academia Human vs. Face Recognition Software that best exemplify the creative spirit demonstrated by Sey- and industry. The program, which featured The face is undoubtedly a familiar and impor- 03/the-invisible-opponent/475611. mour Cray. The award consists of a crystal memento and honorarium of US$10,000. competitions between AI and humans, was tant object to humans, so AI’s vision skills are structured to be fun and engaging. Here, I significant to its progress. In the next competi- Yong Rui is CTO of Lenovo. Contact him at review the competitions and consider what the tion, live on stage, the participants were asked [email protected]. outcomes say about the state of AI. to state whether or not a certain person was in a predefined database with dozens of other peo- Follow us on Twitter @ieeemultimedia. SIDNEY FERNBACH MEMORIAL AWARD The award, which consists of a certificate and a US$2,000 Stenographer vs. Dictation Software ple. The AI did quite well, outperforming the honorarium, is presented annually to an individual for A stenographer competed against the speech human when the dataset pool was small, but Read your subscriptions through “an outstanding contribution in the application of high- CS performance computers using innovative approaches.” dictation software live on site. In a low-noise neither the human nor AI did well when the This article originallythe my appeared publications in portal at http://mycs.computer.org. environment, AI did quite well, with 80 percent dataset was large. IEEE MultiMedia, vol. 24, no. 1, 2017. accuracy versus the stenographer’s 75 percent accuracy. Although this is great progress, there are still several places where AI needs to I certainly performs better in memory and improve. A computing. For example, while it’s diffi- First of all, AI’s accuracy drops significantly cult for humans to remember 100 digits after ACM/IEEE-CS KEN KENNEDY AWARD A certificate and US$5,000 honorarium are awarded jointly when there is an accent. Second, when there is the decimal point of pi, this is trivial for AI. by the ACM and the IEEE Computer Society for outstand- noise, such as at a party, AI experiences signifi- Also, AI computes faster than humans, ing contributions to programmability or productivity in cant challenges. Finally, AI still has a long way although it consumes more energy. For any task high-performance computing together with significant community service or mentoring contributions. to go in understanding natural language at a with specific rules, such as chess or Go, AI will semantic level on top of speech recognition. always be better than humans—if not today, then in the near future. Human Driver vs. Autonomous Car This memory and computing capability is At the beginning of the next competition, a few mostly an extension of the human’s left brain. small objects were put on top of two cars, one The human’s right brain, which is more artis- car with a human driver and the other autono- tic and imaginative and can handle many mous, to see which car had the smoother ride. uncertainties, is quite different from today’s At the end of the test, the objects remained on AI. In fact, AI and humans have their own both cars, ending the competition in a tie. This strengths and weaknesses. So instead of saying

www.computer.org/computingedge 21 4 1070-986X/17/$33.00 c 2017 IEEE Published by the IEEE Computer Society 5 Pervasive Health Editors: Anind K. Dey n Carnegie Mellon University n [email protected] Jesus Favela n CICESE n [email protected] Stephen Intille n Northeastern University n [email protected]

The Future of Pervasive Health

Kay Connelly, Indiana University Oscar Mayora, Create-Net Jesus Favela, CICESE Maia Jacobs, Georgia Tech Aleksandar Matic, Telefonica Alpha Chris Nugent, Ulster University Stefan Wagner, Aarhus University

n 16 May 2016, a group of nearly clinical professionals to assess the solu- extended, real-world deployments. O 15 researchers met at the Future of tions based on their clinical relevance to Given the breadth of work in this field, Pervasive Health Workshop, part of the treating specific diseases. we identified four important areas for International Pervasive Health Confer- While the technology and research research to allow fast prototype devel- ence, to discuss the future of the field. focus has matured, there remain many opment with both mature and novel Pervasive health as a field started in the challenges and opportunities. During technologies. late 90s with the promise of integrat- the Future of Pervasive Health Work- ing innovative health and wellbeing shop, we identified five general themes Personalization and Adaptivity services into everyday life using per- related to pervasive health solutions Pervasive health solutions must sup- vasive computing and communication that must be addressed: port personalization and adaptivity to technologies. After several years of ensure they meet the needs of heteroge- research and development, the field has • technological challenges and neous patient populations. The defini- evolved to a point in which the adop- opportunities, tion of new standards and frameworks tion of ubiquitous computing devices • adoption and adherence, will facilitate quickly incorporating and software is actually starting to • open data, these aspects into systems. become a reality, with patients system- • methods and ethical issues, and atically using such devices and software • education. Appropriation of Technologies in coordination with their caretakers There exist technologies from other and clinical professionals. This list doesn’t pretend to be compre- domains (such as the communication The evolution of this field coincides hensive, yet there was wide consensus or entertainment domains) that could with rapid technology advancements among workshop participants that be appropriated to the health space and with researchers’ growing interest addressing these issues is paramount (persuasive interaction, exergames, and in defining regulatory frameworks and for advancing the field. so on). There is a need to better under- integrating ICTs into clinical practice, stand how to assess such technologies which would increase opportunities for TEChnologICal ChallEngES and their appropriateness for health. embedding innovative and high-quality and oPPorTunITIES pervasive health solutions into our daily Despite the technology and field matur- Real-World Value lives.1 Moreover, pervasive health solu- ing to the point of being able to deploy As pervasive health research matures to tions have reached a maturity level such technologies with real patient popula- the point of having an impact on cur- that they’re not only prototypes built tions for significant periods of time, rent healthcare practices, researchers up from early technologies and tested pervasive health researchers will con- need to consider how to promote such for usability in a laboratory; they’re tinue to work on cutting-edge tech- a transition. It’s important to have now deployed with real patients and nologies that are often too brittle for standard metrics to assess the value of

June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE 16 22PERVASIVE computing Published by the IEEE CS n 1536-1268/17/$33.00 © 2017 IEEE Pervasive Health Editors: Anind K. Dey n Carnegie Mellon University n [email protected] Jesus Favela n CICESE n [email protected] a technology in real-world settings and must explicitly tackle this challenge. From another perspective, most of Stephen Intille n Northeastern University n [email protected] provide such information to the appro- Proposed strategies include using con- the existing healthcare technologies priate stakeholders, so they can make text-aware sensors to detect and tag have been designed targeting specific informed decisions about whether to non-adherent behavior during self- diseases and target groups but with- invest in this technology. measurement activities, using behav- out including features for personaliza- ior- and activity-estimation algorithms tion, which might be one of the key The Future of Pervasive Shared Research Testbeds that infer information from incom- reasons for the limited adherence. Finally, much early research occurs plete data, and encouraging adherence Although multiple scientific commu- Health within a single lab or project, with through various strategies. nities have been proposing different researchers having to create everything The ability to infer adoption is as concepts of personalization, in prac- Kay Connelly, Indiana University from scratch. While this is necessary, important as creating more reliable tice, there’s still a lack of truly indi- Oscar Mayora, Create-Net there’s a need to have shared research sensors, medical devices, or health- vidualized technologies that would Jesus Favela, CICESE testbeds and infrastructures to maxi- care infrastructures for self-care set- personalize interaction, adapt better Maia Jacobs, Georgia Tech mize the investment in infrastructure, tings. In addition to exploiting tech- to user’s routines, or find the right Aleksandar Matic, Telefonica Alpha instead of having each research lab nological advancements for providing moment for engaging a user. Chris Nugent, Ulster University invest in their own personal testbeds. better healthcare services, a parallel Stefan Wagner, Aarhus University Although we can borrow lessons focus must be placed on developing oPEn daTa learned from other technical domains, mechanisms for determining the data Mining large datasets and combining we still must investigate case studies and instances that correspond to lower user them with information from other het- protocols for how to develop, maintain, compliance with the recommended erogeneous sources could help advance n 16 May 2016, a group of nearly clinical professionals to assess the solu- extended, real-world deployments. and share such research testbeds with developments in public health by creat- O 15 researchers met at the Future of tions based on their clinical relevance to Given the breadth of work in this field, respect to human health. This applies to ing new applications in pervasive and treating specific diseases. we identified four important areas for software and hardware infrastructure Pervasive Health Workshop, part of the Limited or selected use of connected health and new strategic International Pervasive Health Confer- While the technology and research research to allow fast prototype devel- as well as study protocols. policies. In this regard, central to the ence, to discuss the future of the field. focus has matured, there remain many opment with both mature and novel biomedical sensors progression of computational tech- Pervasive health as a field started in the challenges and opportunities. During technologies. adoPTIon and adhErEnCE or fitness trackers could niques (such as for activity or behavior late 90s with the promise of integrat- the Future of Pervasive Health Work- As pervasive computing technolo- recognition) in pervasive health is the lead to biased results and ing innovative health and wellbeing shop, we identified five general themes Personalization and Adaptivity gies pave the way for transforming notion of data sharing and the wide- services into everyday life using per- related to pervasive health solutions Pervasive health solutions must sup- traditional episodic healthcare into a inaccurate inferences. spread availability of openly available vasive computing and communication that must be addressed: port personalization and adaptivity to continuous, user-centered, and pre- datasets. Indeed, the need for mak- technologies. After several years of ensure they meet the needs of heteroge- ventive healthcare model, the notion ing health-related data openly avail- research and development, the field has • technological challenges and neous patient populations. The defini- of patient adherence needs to be technology. In this regard, context tag- able has been strongly endorsed by evolved to a point in which the adop- opportunities, tion of new standards and frameworks extended to consider an appropriate ging and tracking user activities can the Health 8 group of global health tion of ubiquitous computing devices • adoption and adherence, will facilitate quickly incorporating use of the technologies (which range help in understanding the reliability of organizations. and software is actually starting to • open data, these aspects into systems. from self-measurement devices, such the collected data, reducing the noise in Models that are being developed to become a reality, with patients system- • methods and ethical issues, and as blood-pressure monitors, to wear- the input data, and informing the ser- characterize various forms of health and atically using such devices and software • education. Appropriation of Technologies able trackers and mobile health apps). vice logic and models. their correlation with behavioral patterns in coordination with their caretakers There exist technologies from other Limited or selected use of biomedical Previous studies have shown that within pervasive health have a funda- and clinical professionals. This list doesn’t pretend to be compre- domains (such as the communication sensors or fitness trackers could lead factors such as demographics, culture, mental requirement to be developed and The evolution of this field coincides hensive, yet there was wide consensus or entertainment domains) that could to biased results and inaccurate infer- lifestyle, and personality represent tested based on data gleaned from diverse with rapid technology advancements among workshop participants that be appropriated to the health space ences, yielding unpredictable effects of strong predictors,2 which can be auto- populations covering a range of ethnic, and with researchers’ growing interest addressing these issues is paramount (persuasive interaction, exergames, and the used technology. Because inappro- matically captured to build individual gender, social, and cultural backgrounds. in defining regulatory frameworks and for advancing the field. so on). There is a need to better under- priate use of health technologies can adherence models. Moreover, further From a technical perspective, we must integrating ICTs into clinical practice, stand how to assess such technologies result in erroneous input information, research is needed to better understand consider the characteristics of the sensor which would increase opportunities for TEChnologICal ChallEngES and their appropriateness for health. this can further affect the quality of the the reasons for non-adherence in spe- nodes being used, their locations within embedding innovative and high-quality and oPPorTunITIES service, which can lower the perceived cific use cases. This will help us better the environment, and the constraints that pervasive health solutions into our daily Despite the technology and field matur- Real-World Value utility and adoption of the technol- address user needs and remove poten- different physical environments impose. lives.1 Moreover, pervasive health solu- ing to the point of being able to deploy As pervasive health research matures to ogy and lead to a downward spiral of tial obstacles to higher compliance— Approaches that consider the true mul- tions have reached a maturity level such technologies with real patient popula- the point of having an impact on cur- adherence, reliable data collection, and for example, by defining novel strate- tifaceted complexities of the data will that they’re not only prototypes built tions for significant periods of time, rent healthcare practices, researchers adoption. gies for encouraging adoption, such inevitably move one step closer toward up from early technologies and tested pervasive health researchers will con- need to consider how to promote such However, it’s unrealistic to expect as designing less cumbersome devices, the development of solutions that work for usability in a laboratory; they’re tinue to work on cutting-edge tech- a transition. It’s important to have perfect adherence from all users, so including gamification and persuasion, in diverse environments and can produce now deployed with real patients and nologies that are often too brittle for standard metrics to assess the value of future pervasive health applications and so on. more generalizable findings.

Making research datasets available tant ethical considerations in pervasive for studies involving protected health beyond the original research team health research. information, researchers must rec- where they have been generated, in a In the first case study, research- ognize their responsibility in ensur- timely and responsible manner, sub- ers at Indiana University are working that participants understand how ject to appropriate safeguards and ing with people with rare diseases their personal data will be used and standards, will offer several benefits. to understand their needs and lived the risks associated with such data From a data collection perspective, experiences and to design tools to sharing, especially in the context of however, generating and collecting assist them.3 In the second case study, rapidly changing pervasive health tech- such data results is expensive and time- researchers at the Georgia Institute nologies. Future work ought to look at consuming. To date, the seemingly of Technology are examining how developing best practices for these con- uncoordinated efforts of the research mobile tools can help individuals sent procedures. community have further added to the diagnosed with breast cancer manage Increasingly, pervasive health stud- complexities, where many datasets are the responsibilities and challenges ies include controlled trials. Although being unnecessarily duplicated with that arise during cancer treatment.4 such studies provide stronger evidence, little or no effort being made to coor- Using these case studies to initiate researchers ought to consider the dinate, consolidate, or share research discussion, researchers at the Perva- potential detriment to participants who protocols, leading to the following sive Health Conference deliberated don’t receive the intervention. When questions: on the ethical questions that arise at an intervention might have important each phase of a research study’s life health impacts, researchers might need • How can we ensure that standards cycle. Although we don’t offer solu- to consider alternative methods that for data management are developed, tions here, awareness of these issues would allow participants in the control promoted, and entrenched so that group to receive the beneficial interven- research data can be shared routinely tion, such as running the intervention and re-used effectively? During recruitment, a central with the participants in the control • How can data be collected in an group after the experimental study is ethical manner, in keeping with best ethical concern is reducing finished. Conversely, researchers must practices? added stress placed on also consider what responsibility they • What frameworks should be imple- both participants and have for any negative health behaviors mented for data sharing to support that are fostered due to an imperfect the creation of an online repository? healthcare providers. intervention. • How can challenges around data Finally, when considering the end of labeling and ground truth annota- research studies, the primary ethical tion be overcome to ensure the high- might help researchers develop more concern we discussed was the return of quality validation of the data being ethically mindful studies. technical devices. Computing technolo- considered? The ethical issues discussed at the gies often become ingrained in people’s • How can data be properly homoge- conference centered around four points daily lives, and returning devices can nized to conduct studies that include in a research study’s life cycle: recruit- thus have negative emotional and health different datasets? ment, consent, data collection, and end consequences. Researchers should con- • How can best practices be transferred of study. During recruitment, a cen- sider what opportunities and barriers both to and from other domains? tral ethical concern is reducing added exist for allowing study participants to stress placed on both participants keep devices upon termination of the We must address these issues in mak- and healthcare providers, particularly research study (for example, devices ing open data repositories for pervasive when working with vulnerable popu- themselves might not be enough; main- health. lations. Researchers at the workshop tenance, technical support, and so on highlighted the need for collabora- might still be required). METhodS and EThICal tion between healthcare providers and ISSuES researchers to develop recruitment EduCaTIon When addressing the opportunities for strategies that place little burden on In the past decade, several efforts have innovations in pervasive health, consid- existing healthcare services. been carried out to formalize pervasive ering the ethical issues associated with Pervasive health research often uti- health as a discipline to be integrated these services is critical. Within the lizes participants’ personal health into the educational paths. In this rap- conference, two research case studies data as a tool for promoting behavior idly evolving context, where technol- motivated the discussion around impor- change. During consent procedures ogy, regulations, and clinical practice

Computing Edge June 2017 18 PERVASIVE24 computing www.computer.org/pervasive Pervasive HealtH Pervasive HealtH Pervasive HealtH

Making research datasets available tant ethical considerations in pervasive for studies involving protected health are reaching a more clear intersection, teaching modules, exercises, code, and when to encourage open source, open- beyond the original research team health research. information, researchers must rec- the curricula in pervasive health must be tutorials would be possible, including data, and international open collabora- where they have been generated, in a In the first case study, research- ognize their responsibility in ensur- revised. access to open source projects and open tion (such as protocol sharing and co- timely and responsible manner, sub- ers at Indiana University are working that participants understand how As a starting point, creating common datasets. Another relevant milestone in clinical study designs). ject to appropriate safeguards and ing with people with rare diseases their personal data will be used and pervasive healthcare terminology will the direction of consolidating this field standards, will offer several benefits. to understand their needs and lived the risks associated with such data allow researchers and students across would be the creation of an open access Technology Methods From a data collection perspective, experiences and to design tools to sharing, especially in the context of the field to collaborate and share results Journal of Pervasive Healthcare. Statistical modeling and reasoning, however, generating and collecting assist them.3 In the second case study, rapidly changing pervasive health tech- and methods more readily, letting us The curriculum covered in the pro- clinical decision support systems, sensor such data results is expensive and time- researchers at the Georgia Institute nologies. Future work ought to look at reproduce experiments in different posed e-book could contain—but systems and architectures, and design- consuming. To date, the seemingly of Technology are examining how developing best practices for these con- countries and settings. This will also wouldn’t be limited to—the sample ing for context awareness are also criti- uncoordinated efforts of the research mobile tools can help individuals sent procedures. lead to better collaboration between topics discussed here. cal methods that must be covered. community have further added to the diagnosed with breast cancer manage Increasingly, pervasive health stud- universities, industry, and user organi- complexities, where many datasets are the responsibilities and challenges ies include controlled trials. Although zations and will help us avoid and miti- Motivation for Study State-of-the-Art Pervasive being unnecessarily duplicated with that arise during cancer treatment.4 such studies provide stronger evidence, gate common practical problems faced Students must learn about the forces Enabling Technologies little or no effort being made to coor- Using these case studies to initiate researchers ought to consider the by the community. shaping healthcare and wellbeing in Students should learn about ambient dinate, consolidate, or share research discussion, researchers at the Perva- potential detriment to participants who Recognizing the fast-moving nature the coming years, including population sensors, as well as sensors that enable protocols, leading to the following sive Health Conference deliberated don’t receive the intervention. When of our field, we don’t recommend lock- aging and the rise in chronic diseases. context awareness. They should also questions: on the ethical questions that arise at an intervention might have important ing down the number of topics and There’s also the need for academic gain an understanding of smart spaces, each phase of a research study’s life health impacts, researchers might need technologies to use in training new pro- research in a field that’s already fast- smart homes, smart hospitals, and • How can we ensure that standards cycle. Although we don’t offer solu- to consider alternative methods that fessionals and researchers. Rather, we’d moving based on industrial innovation other intelligent environments, wear- for data management are developed, tions here, awareness of these issues would allow participants in the control like to identify and describe a common and marketing efforts. ables, and mobile devices. Other topics promoted, and entrenched so that group to receive the beneficial interven- backbone of knowledge in the shape of interest include bio sensors, distrib- research data can be shared routinely tion, such as running the intervention of one or more courses on pervasive Background uted computing, and cloud computing. and re-used effectively? During recruitment, a central with the participants in the control healthcare that we would encourage Students must gain an understanding • How can data be collected in an group after the experimental study is universities to provide to their students. of ubiquitous and pervasive computing Security Principles and Practices ethical manner, in keeping with best ethical concern is reducing finished. Conversely, researchers must Also, best practices, described in stan- concepts, methods, and history. They International and national regulatory practices? added stress placed on also consider what responsibility they dardized formats, should be included— should also be proficient about the state issues, best practices, and technologies • What frameworks should be imple- both participants and have for any negative health behaviors such as how to build and evaluate clini- of the art in research methods, technol- for secure pervasive healthcare are all mented for data sharing to support that are fostered due to an imperfect cal decision support systems, determine ogies, systems, and interventions. important topics to address. the creation of an online repository? healthcare providers. intervention. which sensors to deploy and how, and • How can challenges around data Finally, when considering the end of even understand how to use open Related Fields labeling and ground truth annota- research studies, the primary ethical source and open datasets. Moreover, Several fields are at times indistinguish- ere, we’ve identified a set of chal- tion be overcome to ensure the high- might help researchers develop more concern we discussed was the return of the multidisciplinary nature of the field able from pervasive health, including H lenges and opportunities as a good quality validation of the data being ethically mindful studies. technical devices. Computing technolo- will require promoting interdisciplin- telehealth, telemedicine, telecare, medi- starting point for evolving the pervasive considered? The ethical issues discussed at the gies often become ingrained in people’s ary courses with curricula related not cal informatics, mHealth, uHealth, and health research area. However, the list • How can data be properly homoge- conference centered around four points daily lives, and returning devices can only to HCI and signal processing but ambient assisted living. Such fields are of topics presented here is not exhaus- nized to conduct studies that include in a research study’s life cycle: recruit- thus have negative emotional and health also mechanical engineering, nursing, highly related and, oftentimes, these are tive, and future discussions should occur different datasets? ment, consent, data collection, and end consequences. Researchers should con- and medicine, among others. the fields recognizable by our clinical within the pervasive health community • How can best practices be transferred of study. During recruitment, a cen- sider what opportunities and barriers Furthermore, we propose develop- collaborators and other stakeholders, to define a thorough roadmap of actions both to and from other domains? tral ethical concern is reducing added exist for allowing study participants to ing a high-quality course e-book that including funding agencies. required for evolving the field. A good stress placed on both participants keep devices upon termination of the would be continuously updated and opportunity to continue these discussions We must address these issues in mak- and healthcare providers, particularly research study (for example, devices enhanced. The only available course Research and Design Methods would be organizing dedicated working open data repositories for pervasive when working with vulnerable popu- themselves might not be enough; main- book in the field dates back to 2006,5 It’s important to cover inter- and mul- shops during relevant events, such as the health. lations. Researchers at the workshop tenance, technical support, and so on and although it’s high-quality and tidisciplinary research topics and best Pervasive Health Conference, to maxi- highlighted the need for collabora- might still be required). homogenous, the field has evolved practices, field work and participatory mize participation and the dissemination METhodS and EThICal tion between healthcare providers and considerably since then. Rather than design methods, and interview and of ideas for the future of this field. ISSuES researchers to develop recruitment EduCaTIon merely revising this book, we suggest usability methods. Students should When addressing the opportunities for strategies that place little burden on In the past decade, several efforts have creating a community-based reposi- also learn about ubicomp-style proof- REFEREncES innovations in pervasive health, consid- existing healthcare services. been carried out to formalize pervasive tory, where the entire community of-concept prototyping methods, clini- ering the ethical issues associated with Pervasive health research often uti- health as a discipline to be integrated can provide input and share content cal proof-of-concept prototyping, and 1. O. Mayora, P. Lukowicz, and M. these services is critical. Within the lizes participants’ personal health into the educational paths. In this rap- about the different aspects of perva- the road to evidence-based medicine Marschollek, “Evolving Pervasive Health Research into Clinical Practice,” conference, two research case studies data as a tool for promoting behavior idly evolving context, where technol- sive health. In addition, the traditional methods, including how and when to Methods of Information in Medicine, motivated the discussion around impor- change. During consent procedures ogy, regulations, and clinical practice sharing of reading material and of perform randomized control trials and vol. 53, no. 5, 2014, pp. 380–381.

www.computer.org/computingedge 18 PERVASIVE computing www.computer.org/pervasive JANUAry–MArCh 2017 PERVASIVE computing25 19 Pervasive HealtH Pervasive HealtH This article originally appeared in IEEE Pervasive Computing, vol. 16, no. 1, 2017.

2. S. Wagner, “Towards Future Reliable Pervasive Healthcare with Adherence Strategy Engineering,” Proc. 10th EAI Int’l Conf. Pervasive Computing Technologies for Healthcare, 2016; doi: 10.4108/eai.16-5-2016.2263752. 3. H. Macleod et al., “Ethical Consider- ations in Pervasive Health Research,” Proc. 10th EAI Int’l Conf. Pervasive Computing Technologies for Health- care, 2016; doi: 10.4108/eai.16-5- 2016.2263813. 4. M. Jacobs, J. Clawson, and E.D. Mynatt, “A Cancer Journey Frame- work: Guiding the Design of Holistic Health Technology,” Proc. 10th EAI Int’l Conf. Pervasive Computing Tech- nologies for Healthcare, 2016.

5. A. Mihailidis and J. Bardram, eds., Per- vasive Computing in Healthcare, CRC Press, 2006.

Kay Connelly is an associate professor at Indiana University. Contact her at Call for Articles [email protected]. IEEE Software seeks practical, readable articles that will appeal to experts and nonexperts alike. The magazine aims oscar Mayora is head of the Ubiquitous health Group to deliver reliable, useful, leading-edge at Create-Net. Contact him information to software developers, at [email protected]. engineers, and managers to help them stay on top of rapid technology change. Topics include requirements, design, construction, tools, project management, Jesus Favela is a full profes- process improvement, maintenance, testing, sor at CICESE. Contact him at [email protected]. education and training, quality, standards, and more. Submissions must be original and no more than 4,700 words, including 250 words for each table and gure.

Maia Jacobs is a PhD candidate at Georgia Tech. Author guidelines: Contact her at mjacobs30@ www.computer.org/software/author gatech.edu. Further details: [email protected] www.computer.org/software

aleksandar Matic is a lead Chris nugent is full Stefan Wagner is an researcher at Telefonica Alpha. professor and director of associate professor at Aar- Contact him at aleksandar. the Computer Science hus University. Contact [email protected]. research Institute at Ulster him at [email protected]. University. Contact him at [email protected].

Computing Edge June 2017 20 PERVASIVE26 computing www.computer.org/pervasive Focus on Your Job Search

IEEE Computer Society Jobs helps you easily find a new job in IT, software development, computer engineering, research, programming, architecture, cloud computing, consulting, databases, and many other computer-related areas. New feature: Find jobs recommending or requiring the IEEE CS CSDA or CSDP certifications!

Visit www.computer.org/jobs to search technical job openings, plus internships, from employers worldwide.

http://www.computer.org/jobs

The IEEE Computer Society is a partner in the AIP Career Network, a collection of online job sites for scientists, engineers, and computing professionals. Other partners include Physics Today, the American Association of Physicists in Medicine (AAPM), American Association of Physics Teachers (AAPT), American Physical Society (APS), AVS Science and Technology, and the Society of Physics Students (SPS) and Sigma Pi Sigma. Data analytics EDitOR: seth Earley, Earley information science, [email protected]

The Evolving Role of the CDO

Seth Earley, Earley Information Science

igital transformations are a variety of venues for knowledge IT organization is supposed to data transformations— sharing have emerged to help define manage, fix, remediate, and make they require that enter- how this role needs to function in consumable—but, unfortunately, prises look at processes the information-driven enterprise. IT is not up to this task. The per- fromD an end-to-end, holistic Conferences, associations, and or- ception that IT should be solely perspective and remove physical ganizations centered around the responsible for data is changing, constraints from information com- community of people interested given that the overarching issues munications. Because our on- in data at the executive level are of transformation now require re- line world and the entire digital helping to shape this new C-suite examination of core business mod- experience is based on ones and member. els, core value propositions, and zeros, the only way to achieve These developments are part the fundamental assumptions on digital transformation is by get- of the emerging realization by which the business is based. ting control of the avalanche of senior executives that data is at The digital transformation of data that is generated by supplier, the soul of their enterprises. Or- the enterprise changes this per- customer, and employee interac- ganizations are in a never-ending spective. For the first time, orga- tions. Today’s data environment cycle of pursuing business value nizations must begin looking at is orders of magnitude more com- from their data, but despite large end-to-end processes and the en- plex, diverse, and voluminous amounts of resources and atten- tire customer lifecycle to effective- than that of prior computing eras, tion, many data initiatives have ly respond to customer needs. IT and the growth will continue at an resulted in disappointment. cannot be responsible for chang- exponential rate. ing the business value proposition Given this backdrop, enterprises Reexamining Core Models or fundamentals—this requires are realizing that new leadership and Value Propositions an executive mandate. Because is needed to take on responsibil- Part of the problem lies in the per- data is now inextricably at the ity for data and information value. ception of who is responsible for le- heart of transformation, and digi- The chief data officer (CDO) role veraging data in business processes. tal disruption will have an enor- has arisen to fill this need. A rela- Data has been an inconvenient mous impact on businesses and tively new role in the enterprise, means to an end—a mechanism institutions of all sizes, data is now the CDO has taken on a number to get work done. Data issues slow a C-level responsibility. of forms. Many organizations have businesses down and get in the way. Data must now be considered experimented with the role, and These issues are something that the the most valuable asset of the

28 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE 64 IT Pro January/February 2017 Published by the IEEE Computer Society 1520-9202/17/$33.00 © 2017 IEEE Data analytics EDitOR: seth Earley, Earley information science, [email protected]

organization. It is the source of derstands this and raises that point different? How will having a CDO competitive advantage and mar- of view above parochial interests. change how the business operates? ketplace differentiation. Business Enterprise capabilities and agil- How will it impact decision mak- models are created from data struc- ity around data are needed to ing, processes, policies, and the tures and flows. The customer ex- compete with born-digital and technologies that are used to run perience is entirely composed of other disruptive market entrants. the business day to day? The CDO or supported by data. But in most Being able to rapidly create new needs to be an empowered posi- large enterprises, the data land- products and offerings depends tion with the support of the rest The Evolving Role scape is so complex and fragment- on being able to quickly and cost- of the leadership team. The CDO ed that these models, structures, effectively assemble the correct can change the status quo in sev- and experiences can be adapted data from various internal and ex- eral ways: of the CDO only through time-intensive, cost- ternal sources, and then craft that ly, and difficult-to-execute initia- data into a user experience that • Increase visibility and awareness in tives. Experimentation, which is a provides value or solves a specific the C-suite. The CDO will have fundamental requirement in the problem. Rapid experimentation the ability to communicate and digital ecosystem, is too slow and and evolution are part of the faster collaborate with leadership. Seth Earley, Earley Information Science cost-prohibitive to be viable. clock speeds that every industry is What will a CDO communicate experiencing. According to Tim that a CIO will not? Is this an- A Culture of Proactive Fitzgerald, vice president of digi- other aspect of the CIO’s job, or Stewardship tal transformation at Avnet, the is it distinctly different from that CDOs have their work cut out for rate of change that digital trans- function? The answer is that them. They need to create a culture formations are causing businesses CDOs can raise the visibility of of proactive stewardship, get the is unprecedented (as per his talk process issues and focus atten- igital transformations are a variety of venues for knowledge IT organization is supposed to buy-in of users, line-of-business at the Worldwide Business Re- tion on innovation with data that data transformations— sharing have emerged to help define manage, fix, remediate, and make managers, and leadership, and search B2B Connect Conference; the CIO does not have band- they require that enter- how this role needs to function in consumable—but, unfortunately, eliminate the “acts of heroics” that https://twitter.com/sethearley/ width for or awareness of. prises look at processes the information-driven enterprise. IT is not up to this task. The per- many organizations rely on to get status/777622008397869056). • Focus on cultural change. Many fromD an end-to-end, holistic Conferences, associations, and or- ception that IT should be solely products to market, campaigns This next series of articles in the CDOs are evangelists and spend perspective and remove physical ganizations centered around the responsible for data is changing, launched, and programs success- Data Analytics department, based their time educating executives constraints from information com- community of people interested given that the overarching issues fully completed. In addition, they mostly on interviews conducted about the importance of data munications. Because our on- in data at the executive level are of transformation now require re- need to establish data ownership, with CDOs who attended the MIT and what can be done to inno- line world and the entire digital helping to shape this new C-suite examination of core business mod- remediate upstream processes, re- Chief Data Officers Symposium vate with data assets. This effort experience is based on ones and member. els, core value propositions, and solve conflicts, and change the way (www.mitcdoiq.org), will share the could be directed toward pro- zeros, the only way to achieve These developments are part the fundamental assumptions on systems and processes produce experiences of several business ducing a specific outcome, or digital transformation is by get- of the emerging realization by which the business is based. and consume data. Finally, they and technology leaders in deal- toward increasing awareness of ting control of the avalanche of senior executives that data is at The digital transformation of need to have a cross-functional ing with data challenges related cross-functional challenges and data that is generated by supplier, the soul of their enterprises. Or- the enterprise changes this per- and cross-business-unit under- to digital transformation. Some of processes that impact the ability customer, and employee interac- ganizations are in a never-ending spective. For the first time, orga- standing of how information flows these leaders are CDOs, but input to leverage data most effectively. tions. Today’s data environment cycle of pursuing business value nizations must begin looking at through the enterprise and know was also provided by CEOs, senior • Focus on process change. Some is orders of magnitude more com- from their data, but despite large end-to-end processes and the en- how to measure efficiency, quality, vice presidents, and senior man- CDOs roll up their sleeves and plex, diverse, and voluminous amounts of resources and atten- tire customer lifecycle to effective- and effectiveness. agers. They offer their practical work with managers at the level of than that of prior computing eras, tion, many data initiatives have ly respond to customer needs. IT The traditional responsibility for experience and lessons learned in specific initiatives and programs and the growth will continue at an resulted in disappointment. cannot be responsible for chang- data has been with the CIO and data initiatives. These are practical that are geared toward a new ca- exponential rate. ing the business value proposition within business units that were approaches for understanding the pability or specific outcome. Given this backdrop, enterprises Reexamining Core Models or fundamentals—this requires producing and consuming data. role and responsibility of the CDO • Focus on data quality issues. Data are realizing that new leadership and Value Propositions an executive mandate. Because The focus has been on serving and how this role is providing the quality in a particular domain— is needed to take on responsibil- Part of the problem lies in the per- data is now inextricably at the customers and employees rather necessary foundation for digital for example, customer data— ity for data and information value. ception of who is responsible for le- heart of transformation, and digi- than looking at data holistically. transformations. requires significant effort to The chief data officer (CDO) role veraging data in business processes. tal disruption will have an enor- Data from these functions might address due to the number of has arisen to fill this need. A rela- Data has been an inconvenient mous impact on businesses and be consumed elsewhere, but this Why Is a New Role stakeholders, processes, systems, tively new role in the enterprise, means to an end—a mechanism institutions of all sizes, data is now viewpoint was not at the root of Needed? and applications that are impact- the CDO has taken on a number to get work done. Data issues slow a C-level responsibility. process and policy decisions. The When an organization is established. Many times, these are part of of forms. Many organizations have businesses down and get in the way. Data must now be considered costs and benefits of data are not ing a new role, several questions master data initiatives, which can experimented with the role, and These issues are something that the the most valuable asset of the evenly distributed. The CDO un- need to be answered: What will be take years to implement, so this

should be seen as a long-term not succeed if you don’t have qual- sources, and they understand the responsibility. ity data to begin with.” When data constraints of the data and quality • Focus on governance. Data and in- issues are not addressed at the needs of downstream processes. formation governance is a thread source, downstream users will try They routinely discuss quality is- that runs through all of these to fix them. This is problematic sues and develop plans to address areas, and, conceivably, a CDO because problems. Stewardship with focus could stand up a governance and discipline, guided by a data program to tackle all of these • the downstream users are usu- strategy and augmented with the challenges. Surprisingly, (con- ally high-cost analysts, actuar- proper tools, is an essential com- sidering the close tie-in of the ies, or data scientists; ponent. CDOs work in different realization of data value to ef- • when these users fix the data, functional areas with managers fective governance), some CDOs they create a new source that who are trying to achieve business do not get directly involved in others in the organization might objectives, such as greater market governance, acting more as am- not know about or be able to ac- penetration and increased profit- bassadors and liaisons rather cess; and ability. They help those managers than taking on this operational • when others cannot access the to understand how the quality of aspect of the role. Lack of ef- “fixed” data, they might use data impacts their business objec- fective governance (which in- different assumptions or ap- tives. Making this connection is an cludes data strategy, data quality, proaches that will make it dif- important part of the CDO’s role. and data ownership) is a sig- ficult to reconcile the resulting nificant impediment to digital multiple copies of the data. The CDO Communicates transformations. Value to the CEO Gilligan goes on to observe that Many corporate goals fail to be According to IDC, within the there are numerous sources of data achieved because top-level sup- next two years, two-thirds of and multiple streams—both inter- port is lacking. The CEO must Global 2000 enterprises will have nal and external—from partners be an evangelist for the message digital transformation at the center and vendors that have different re- that data quality is the first step in of their corporate strategy. Data is strictions and usages. There needs the information ecosystem. This at the core of digital business, and to be a coordinated process for re- message starts with the CDO, who data without integrity won’t be able porting on quality and developing can connect abstract concepts to to support digital transformation remediation plans to address qual- tangible business outcomes. The initiatives.1 Addressing the root ity issues. Putting a remediation CDO understands that the pro- causes of data issues requires au- plan into place at the source is not cess starts with the onboarding of thority at the highest levels of the always straightforward. In some data—whether internal or exter- enterprise. Organizations need to cases, the source does not have a nal. For example, customer and empower a senior executive to have mandate or authority. In other cas- marketplace data goes through an the resources to fix legacy issues es, they might not have funding to ecosystem of processes and appli- (which can be significant) and be clean the data. Governance of the cations and eventually becomes able to effect process and cultural data source, with authority and re- part of a marketing strategy. Data change. That role is the CDO. sourcing, is the answer. quality issues will cause that mar- Data governance is less glamor- keting strategy to fail, however, Establishing the Need in ous than some other areas, such as because poor quality data leads to an Organization visualization, according to Steve faulty baseline assumptions. According to Jim Gilligan, former Orrin, chief technologist of Intel Improving business outcomes CEO of Blue Cross Life Insurance Federal. “But you need the gover- is the driver for many CDO ini- of Canada, “A lot of challenges are nance in order to get those inter- tiatives. Mark Nance, CDO of still the traditional ones, such as esting visualizations, especially American Fidelity, notes that get- data management, such as making when analytics are being done ting greater value out of business sure you do data onboarding in a across different types of data.” data was the catalyst for initiating quality and efficient fashion.” The its CDO role three years ago. “Ini- increasing use of analytics through Data Stewards Collaborate tially, we were going to focus on more powerful tools may expose About Quality governance, but business value problems. “Of course,” adds Gil- Data stewards need to be part of came to be seen as the core moti- ligan, “your analytics strategy will the process—they are close to data vator for the role.”

30 Computing Edge June 2017 66 IT Pro January/February 2017 Data analytics should be seen as a long-term not succeed if you don’t have qual- sources, and they understand the This kind of business-critical develop the capabilities needed to The CDO and Cybersecurity responsibility. ity data to begin with.” When data constraints of the data and quality message must be communicated to battle these disruptors by bring- Cybersecurity is a part of risk and • Focus on governance. Data and in- issues are not addressed at the needs of downstream processes. the board, where key decisions are ing scale, long-term relationships, data management, and these are formation governance is a thread source, downstream users will try They routinely discuss quality is- being made. The CDO can also and market presence to the battle. board-level issues. Boards have to that runs through all of these to fix them. This is problematic sues and develop plans to address help the CEO translate the mes- The CDO arms the organization become more conversant in cy- areas, and, conceivably, a CDO because problems. Stewardship with focus sage into issues that might have a with the digital agility to do battle bersecurity because digital trans- could stand up a governance and discipline, guided by a data higher priority, such as organiza- with these disruptors. formation requires foundational program to tackle all of these • the downstream users are usu- strategy and augmented with the tional risk. In the case of risks and Digital agility gives traditional cybersecurity, which is also depen- challenges. Surprisingly, (con- ally high-cost analysts, actuar- proper tools, is an essential com- risk mitigation, the CDO will work institutions the opportunity to dent on good governance of data. sidering the close tie-in of the ies, or data scientists; ponent. CDOs work in different with a chief risk officer (CRO) to interface with their customers in Digital disruption will revolu- realization of data value to ef- • when these users fix the data, functional areas with managers quantify the impact of data quality. a completely different way. The tionize the whole relationship that fective governance), some CDOs they create a new source that who are trying to achieve business By identifying the issues with the issue for both incumbents and companies have with customers; do not get directly involved in others in the organization might objectives, such as greater market greatest urgency, the CDO can le- innovators is ensuring that they it will change how business-to- governance, acting more as am- not know about or be able to ac- penetration and increased profit- verage in-flight initiatives to benefit move from analog to digital in a business companies are aware of bassadors and liaisons rather cess; and ability. They help those managers multiple departments. By improv- way that maintains quality and consumer needs. One of those than taking on this operational • when others cannot access the to understand how the quality of ing data quality for risk mitiga- augments and streamlines the paramount needs is trust, which aspect of the role. Lack of ef- “fixed” data, they might use data impacts their business objec- tion, the marketing organization relationship. A digital solution al- is based on safeguarding and pro- fective governance (which in- different assumptions or ap- tives. Making this connection is an can benefit as well. However, the lows for experimentation and in- tecting data and assets. Safeguard- cludes data strategy, data quality, proaches that will make it dif- important part of the CDO’s role. importance of support from the novation at a faster clock speed ing customers’ data and privacy and data ownership) is a sig- ficult to reconcile the resulting CEO cannot be overstated. Der- than ever before. But digital plat- means understanding what they nificant impediment to digital multiple copies of the data. The CDO Communicates ek Strauss, CDO of Ameritrade, forms require that data quality need and not looking at data just transformations. Value to the CEO observed that “a CDO needs ex- processes are in place. This is the for business needs. The CDO’s Gilligan goes on to observe that Many corporate goals fail to be ecutive sponsorship to provide air basis of the value proposition—a overarching role in working with According to IDC, within the there are numerous sources of data achieved because top-level sup- cover.”2 The CEO not only needs new or enhanced way of interact- data managers in all departments next two years, two-thirds of and multiple streams—both inter- port is lacking. The CEO must to understand the importance of ing with the customer. helps maintain an enterprise-wide Global 2000 enterprises will have nal and external—from partners be an evangelist for the message data to the business, but must also Even with the best technology, vision of monitoring data security, digital transformation at the center and vendors that have different re- that data quality is the first step in drive, from the top, the kind of there is a dependency on data, and including vulnerabilities resulting of their corporate strategy. Data is strictions and usages. There needs the information ecosystem. This cultural change that is needed for data has a dependency on process, from the trend toward increasing at the core of digital business, and to be a coordinated process for re- message starts with the CDO, who transformation. and process has a dependency partner and customer access to data without integrity won’t be able porting on quality and developing can connect abstract concepts to on governance. Although some corporate data. to support digital transformation remediation plans to address qual- tangible business outcomes. The The CDO and Digital CDOs choose to evangelize, they initiatives.1 Addressing the root ity issues. Putting a remediation CDO understands that the pro- Disruption typically do so in less mature or- The CDO Role Varies by causes of data issues requires au- plan into place at the source is not cess starts with the onboarding of Pure play, “born digital” financial ganizations in which business ex- Organization thority at the highest levels of the always straightforward. In some data—whether internal or exter- and insurance technology start- ecutives need to understand why Different types of CDOs are need- enterprise. Organizations need to cases, the source does not have a nal. For example, customer and ups (known as “FinTech” and “In- this is so important. Without that ed for different types of industries empower a senior executive to have mandate or authority. In other cas- marketplace data goes through an sureTech”) are transforming and understanding, the work of op- and maturity levels. For an organi- the resources to fix legacy issues es, they might not have funding to ecosystem of processes and appli- disrupting their respective sectors. erationalization cannot gain the zation early in the learning and ma- (which can be significant) and be clean the data. Governance of the cations and eventually becomes The value proposition of these needed attention and support and turity curve, the job is more about able to effect process and cultural data source, with authority and re- part of a marketing strategy. Data startups is in client-centricity. En- therefore will not gain traction. In education and awareness than get- change. That role is the CDO. sourcing, is the answer. quality issues will cause that mar- gagement is much more effective cases where people are nodding ting involved in day-to-day issues. Data governance is less glamor- keting strategy to fail, however, than with traditional institutions their heads and saying, “Yes, we Others set up an infrastructure of Establishing the Need in ous than some other areas, such as because poor quality data leads to because these businesses are de- understand, now how do we do decision-making bodies and rules an Organization visualization, according to Steve faulty baseline assumptions. signed from the ground up to take it?,” the CDO role shifts to one and then move on, while others According to Jim Gilligan, former Orrin, chief technologist of Intel Improving business outcomes full advantage of serving the cus- of leading, or at least sponsoring, continue to lead those decision- CEO of Blue Cross Life Insurance Federal. “But you need the gover- is the driver for many CDO ini- tomer through digital means. governance. According to Gartner, making groups on an ongoing ba- of Canada, “A lot of challenges are nance in order to get those inter- tiatives. Mark Nance, CDO of This disruption is hard for tra- the CDO is typically responsible sis. Depending on the maturity of still the traditional ones, such as esting visualizations, especially American Fidelity, notes that get- ditional organizations to battle for “the firm’s enterprise-wide data the organization and the demands data management, such as making when analytics are being done ting greater value out of business because legacy systems and lum- and information strategy, gover- of its industry’s state of transforma- sure you do data onboarding in a across different types of data.” data was the catalyst for initiating bering processes prevent them nance, control, policy develop- tion (for example, drastic change is quality and efficient fashion.” The its CDO role three years ago. “Ini- from delivering the value propo- ment, and effective exploitation.”3 impacting insurance and financial increasing use of analytics through Data Stewards Collaborate tially, we were going to focus on sition of client-centricity through This makes sense once leadership services more quickly than the in- more powerful tools may expose About Quality governance, but business value digital data and end-to-end ho- is on board. Before that point, the dustrial products vertical), as well problems. “Of course,” adds Gil- Data stewards need to be part of came to be seen as the core moti- listic thinking around serving CDO is more cheerleader than as the state of its infrastructure ligan, “your analytics strategy will the process—they are close to data vator for the role.” the customer. The CDO helps to catalyst for implementation. (some transformations require

www.computer.org/computingedge 31 66 IT Pro January/February 2017 computer.org/ITPro 67 this role is likely to be on a growth PURPOSE: The IEEE Computer Society is the world’s largest association of computing trajectory for some time. professionals and is the leading provider of technical information in the field. MEMBERSHIP: Members receive the monthly magazine Computer, discounts, and References opportunities to serve (all activities are led by volunteer members). Membership is 1. “IDC Reveals Worldwide Digital open to all IEEE members, affiliate society members, and others interested in the Transformation Predictions; Kicks computer field. Off IDC FutureScape Web Con- OMBUDSMAN: Email [email protected]. ference,” IDC press release, 4 COMPUTER SOCIETY WEBSITE: www.computer.org Nov. 2015; www.idc.com/getdoc. jsp?containerId=prUS40553515. Next Board Meeting: 30 January–3 February 2017, Anaheim, CA, USA 2. L.T. Moss and S. Adelman, “The Role of Chief Data Offi- EXECUTIVE COMMITTEE cer in the 21st Century,” Data In- President: Jean-Luc Gaudiot sight and Social Bull., vol. 13, no. 2; President-Elect: Hironori Kasahara; Past President: Roger U. Fujii; Secretary: Forrest Shull; First VP, Treasurer: David Lomet; Second VP, Publications: Gregory http://experts.cutter.com/acton/ T. Byrd; VP, Member & Geographic Activities: Cecilia Metra; VP, Professional & attachment/19169/f-006c/1/-/-/-/-/ Educational Activities: Andy T. Chen; VP, Standards Activities: Jon Rosdahl; V P, Role%20of%20the%20Chief%20 Technical & Conference Activities: Hausi A. Müller; 2017–2018 IEEE Director & Data%20Officer.pdf. Delegate Division VIII: Dejan S. Milojičić; 2016–2017 IEEE Director & Delegate 3. Understanding the Chief Data Officer Division V: Harold Javid; 2017 IEEE Director-Elect & Delegate Division V-Elect: Role, Gartner, 18 Feb. 2015; www. John W. Walz gartner.com/smarterwithgartner/ this role is likely to be on a growth understanding-the-chief-data BOARD OF GOVERNORS PURPOSE: The IEEE Computer Society is the world’sTerm largest Expiring association 2017: Alfredoof computing Benso, Sy-Yen Kuo, Ming C. Lin, Fabrizio Lombardi, trajectory for some time. -officer-role/. professionals and is the leading provider of technicalHausi information A. Müller, in Dimitrios the field. Serpanos, Forrest J. Shull 4. R. Bean, “Chief Data Officers Blaze MEMBERSHIP: Members receive the monthly magazineTerm Expiring Computer 2018:, discounts, Ann DeMarle, and Fred Douglis, Vladimir Getov, Bruce M. References Uncharted Corporate Frontier,” opportunities to serve (all activities are led by volunteerMcMillin, members). Cecilia Metra,Membership Kunio isUchiyama, Stefano Zanero 1. “IDC Reveals Worldwide Digital Wall Street J., 21 Apr. 2014; http:// open to all IEEE members, affiliate society members,Term and Expiring others interested2019: Saurabh in the Bagchi, Leila De Floriani, David S. Ebert, Jill I. Gostin, Transformation Predictions; Kicks computerblogs.wsj.com/cio/2014/04/21/chief field. William Gropp, Sumi Helal, Avi Mendelson Off IDC FutureScape Web Con- OMBUDSMAN:-data-officers-blaze-uncharted Email [email protected]. ference,” IDC press release, 4 -corporate-frontier/. EXECUTIVE STAFF COMPUTER SOCIETY WEBSITE: www.computer.org Nov. 2015; www.idc.com/getdoc. 5. T. Chakravorty, “CDO vs. CIO: Executive Director: Angela R. Burgess; Director, Governance & Associate Executive Director: Anne Marie Kelly; Director, Finance & Accounting: Sunny Hwang; jsp?containerId=prUS40553515. NextWho Board Does Meeting: What,” 30 January–3 CIO Februaryfrom 2017, Anaheim, CA, USA Director, Information Technology & Services: Sumit Kacker; Director, Membership 2. L.T. Moss and S. Adelman, IDG, 9 July 2014; www.cio.in/ Development: Eric Berkowitz; Director, Products & Services: Evan M. Butterfield; “The Role of Chief Data Offi- EXECUTIVEopinions/cdo-vs-cio-who-does COMMITTEE Director, Sales & Marketing: Chris Jensen cer in the 21st Century,” Data In- President:-what. Jean-Luc Gaudiot sight and Social Bull., vol. 13, no. 2; 6.President-Elect: Top Performers Hironori Appoint Kasahara;Chief Data Past Of- President:COMPUTER Roger U. Fujii; SOCIETY Secretary: OFFICES Forrest Shull; First VP, Treasurer: David Lomet; Second VP, Publications: Gregory http://experts.cutter.com/acton/ ficers, Forrester, 20 Aug. 2015; www. Washington, D.C.: 2001 L St., Ste. 700, Washington, D.C. 20036-4928 T. Byrd; VP, Member & Geographic Activities: Cecilia Metra; VP, Professional & attachment/19169/f-006c/1/-/-/-/-/ forrester.com/report/Top+Perform Phone: +1 202 371 0101 • Fax: +1 202 728 9614 • Email: [email protected] Educational Activities: Andy T. Chen; VP, Standards Activities: Jon Rosdahl; V P, Data analytics Role%20of%20the%20Chief%20 ers+Appoint+Chief+Data+Officers/ Los Alamitos: 10662 Los Vaqueros Circle, Los Alamitos, CA 90720 Technical & Conference Activities: Hausi A. Müller;Phone: 2017–2018 +1 714 IEEE821 8380 Director • Email: & [email protected] Data%20Officer.pdf. -/E-RES123064. Delegate Division VIII: Dejan S. Milojičić; 2016–2017Membership IEEE Director & Publication & Delegate Orders 3. Understanding the Chief Data Officer Division V: Harold Javid; 2017 IEEE Director-ElectPhone: & Delegate +1 800 Division 272 6657 V-Elect • Fax:: +1 714 821 4641 • Email: [email protected] major technology upgrades, such data that is much more user Tamal Chakravorty, director of IT this role is likely to be on a growth Role, Gartner, 18 Feb. 2015; www. Seth Earley is CEO of Earley Information PURPOSE: The IEEE Computer Society is the world’sJohn W.largest Walz association of computing Asia/Pacific: Watanabe Building, 1-4-2 Minami-Aoyama, Minato-ku, Tokyo 107- as costly enterprise resource plan- friendly and accessible to business and Test for Ericsson Global Ser- thistrajectory role is for likely some to betime. on a growth gartner.com/smarterwithgartner/ Science (www.earley.com). He’s an expert PURPOSE:professionals The and IEEE is the Computer leading Societyprovider is ofthe technical world’sBOARD largestinformation OF associationGOVERNORS in the field.of computing 0062, Japan • Phone: +81 3 3408 3118 • Fax: +81 3 3408 3553 • Email: tokyo.ofc@ ning replacements), the organiza- users than the previous solution. vices, the CDO needs to “know trajectory for some time. understanding-the-chief-data in knowledge processes, enterprise data ar- computer.org professionalsMEMBERSHIP: and Members is the leading receive provider the monthly of technical magazineTerm informationExpiring Computer 2017: in, discounts,the Alfredo field. Benso, and Sy-Yen Kuo, Ming C. Lin, Fabrizio Lombardi, References -officer-role/. chitecture, and customer experience man- tion will need very different types Thomas believes that this approach how to find the required informa- MEMBERSHIP:opportunities to Membersserve (all receiveactivities the are monthly led by magazinevolunteer members).Computer, discounts,Membership and is 5 Hausi A. Müller, Dimitrios Serpanos, Forrest J. ShullIEEE BOARD OF DIRECTORS of CDOs and CDO agendas. to data, maturity in governance, tion to fuel business growth.” References 1. “IDC Reveals Worldwide Digital 4. openR. Bean, to all “Chief IEEE members, Data Officers affiliate Blaze society members,agement andstrategies. others interestedHis interests in the include opportunities to serve (all activities are led by volunteerTerm Expiring members). 2018: Membership Ann DeMarle, is Fred Douglis,President Vladimir &Getov, CEO: Bruce Karen M. Bartleson; President-Elect: James Jefferies; Past President: However, the core responsibility and ability to effect cultural 1. Transformation“IDC Reveals Worldwide Predictions; Digital Kicks Uncharted Corporate Frontier,” customer experience analytics, knowledge opencomputer to all field. IEEE members, affiliate society members,McMillin, and Cecilia others Metra,interested Kunio in theUchiyama, StefanoBarry Zanero L. Shoop; Secretary: William Walsh; Treasurer: John W. Walz; Director & OffTransformation IDC FutureScape Predictions; Web KicksCon- Wall Street J., 21 Apr. 2014; http:// management, structured and unstructured for data, data quality, and achiev- change provides a competitive computerOMBUDSMAN: field. Email [email protected] Expiring 2019: Saurabh Bagchi, Leila De Floriani,President, David IEEE-USA: S. Ebert, KarenJill I. Gostin, Pedersen; Director & President, Standards Association: ing greater business value from that advantage among peers. (See a fu- he interviews I conducted ference,”Off IDC FutureScape IDC press release,Web Con 4- OMBUDSMAN:blogs.wsj.com/cio/2014/04/21/chief Email [email protected]. data systems and strategy, and machine Forrest Don Wright; Director & VP, Educational Activities: S.K. Ramesh; Director COMPUTER SOCIETY WEBSITE: www.computer.orgWilliam Gropp, Sumi Helal, Avi Mendelson data is common to all the manifes- ture column for the full interview with leaders knowledgeable Nov.ference,” 2015; IDC www.idc.com/getdoc. press release, 4 -data-officers-blaze-uncharted learning. Contact him at [email protected]. & VP, Membership and Geographic Activities: Mary Ellen Randall; Director & VP, COMPUTER SOCIETY WEBSITE: www.computer.org tations of the CDO role. with Brandon Thomas.) about the CDO role point jsp?containerId=prUS40553515.Nov. 2015; www.idc.com/getdoc. Next-corporate-frontier/. Board Meeting: 30 January–3 February 2017,EXECUTIVE Anaheim, CA, STAFF USA Publication Services and Products: Samir El-Ghazaly; Director & VP, Technical T Executive Director: Angela R. Burgess; Director, Activities:Governance Marina & Associate Ruggieri; Executive Director & Delegate Division V: Harold Javid; Director to several observations: 2. L.T.jsp?containerId=prUS40553515. Moss and S. Adelman, 5. NextT. Chakravorty, Board Meeting: “CDO 30 January–3 vs. CIO: February 2017, Anaheim, CA,Read USA your subscriptions Director: Anne Marie Kelly; Director, Finance & Accounting:& Delegate Division Sunny Hwang; VIII: Dejan S. Milojičić Zions Bancorporation’s Changing Trends for CDOs 2. “TheL.T. MossRole ofand Chief S. DataAdelman, Offi- EXECUTIVEWho Does COMMITTEEWhat,” CIO from through the myCS publi- Director, Information Technology & Services: Sumit Kacker; Director, Membership Journey According to Yang Lee of North- • The role is new. “The role of cer“The in theRole 21st of Century,”Chief Data Data Offi In- EXECUTIVEPresident:IDG, 9 JulyJean-Luc COMMITTEE 2014; Gaudiot www.cio.in/ cations portal at http:// revised 2 Dec. 2016 Development: Ericmycs.computer.org. Berkowitz; Director, Products & Services: Evan M. Butterfield; Brandon Thomas, CDO of Zions eastern University, co-director CDO is in its early stages, much cersight in and the Social 21st Bull.Century,”, vol. 13, Data no. In 2;- President:President-Elect:opinions/cdo-vs-cio-who-does Jean-Luc Hironori Gaudiot Kasahara; Past President: Roger U. Fujii; Secretary: Forrest Shull; First VP, Treasurer: David Lomet;Director, Second VP,Sales Publications: & Marketing: Gregory Chris Jensen Bancorporation, described the path for MIT’s Chief Data Officer Re- like the role of CMO was 20 years sighthttp://experts.cutter.com/acton/ and Social Bull., vol. 13, no. 2; President-Elect:-what. Hironori Kasahara; Past President: Roger U. Fujii; Secretary: ForrestT. Byrd; Shull; VP, Member First VP, & Treasurer:Geographic David Activities: Lomet; Cecilia Second Metra; VP, Publications: VP, Professional Gregory & of his organization in getting con- search Program, “Traditionally, ago,” according to Rich Wendell, attachment/19169/f-006c/1/-/-/-/-/http://experts.cutter.com/acton/ 6. Top Performers Appoint Chief Data Of- COMPUTER SOCIETY OFFICES T.Educational Byrd; VP, MemberActivities: & AndyGeographic T. Chen; Activities: VP, Standards Cecilia Activities: Metra; VP, Jon Professional Rosdahl; V & P, trol of data and processes. A cata- data practices were led by middle CEO and CDO of Tellic, a firm Role%20of%20the%20Chief%20attachment/19169/f-006c/1/-/-/-/-/ ficers, Forrester, 20 Aug. 2015; www. Washington, D.C.: 2001 L St., Ste. 700, Washington, D.C. 20036-4928 computer.org/ITPro 69 EducationalTechnical & Activities:Conference Andy Activities: T. Chen; Hausi VP, A.Standards Müller; 2017–2018Activities: IEEE Jon Rosdahl; Director V& P, lyst for launching his company’s managers, lacking executive col- specializing in pharma data sci- Data%20Officer.pdf.Role%20of%20the%20Chief%20 forrester.com/report/Top+Perform Phone: +1 202 371 0101 • Fax: +1 202 728 9614 • Email: [email protected] TechnicalDelegate Division & Conference VIII: Dejan Activities: S. Miloji Hausičić; A.2016–2017 Müller; 2017–2018IEEE Director IEEE & Director Delegate & journey was the increased em- laboration. The emerging CDO ences as a service. 3. UnderstandingData%20Officer.pdf. the Chief Data Officer ers+Appoint+Chief+Data+Officers/ Los Alamitos: 10662 Los Vaqueros Circle, Los Alamitos, CA 90720 DivisionDelegate V:Division Harold VIII: Javid; Dejan 2017 S.IEEE Miloji Director-Electčić; 2016–2017Phone: & IEEEDelegate +1 714 Director 821 Division 8380 & Delegate •V-Elect Email: : [email protected] • The role is likely to remain fluid 3. RoleUnderstanding, Gartner, the18 ChiefFeb. 2015;Data Officerwww. -/E-RES123064. JohnDivision W. WalzV: Harold Javid; 2017 IEEE Director-ElectMembership & Delegate & PublicationDivision V-Elect Orders: for a number of years. Rolegartner.com/smarterwithgartner/, Gartner, 18 Feb. 2015; www. John W. Walz Phone: +1 800 272 6657 • Fax: +1 714 821 4641 • Email: [email protected] • How the role is expressed de- gartner.com/smarterwithgartner/understanding-the-chief-data SethBOARD Earley isOF CEO GOVERNORS of Earley Information Term Expiring 2017: Alfredo Benso, Sy-Yen Kuo,Asia/Pacific: Ming C. Lin, Watanabe Fabrizio Lombardi,Building, 1-4-2 Minami-Aoyama, Minato-ku, Tokyo 107- The core responsibility for data, data quality, and pends on the company’s matu- understanding-the-chief-data-officer-role/. ScienceBOARD (www.earley.com). OF GOVERNORS He’s an expert TermHausi ExpiringA. Müller, 2017: Dimitrios Alfredo Serpanos, Benso, Sy-YenForrest Kuo, J. Shull0062, Ming Japan C. Lin, • Phone: Fabrizio +81 Lombardi, 3 3408 3118 • Fax: +81 3 3408 3553 • Email: tokyo.ofc@ rity level with respect to data 4. -officer-role/.R. Bean, “Chief Data Officers Blaze in knowledge processes, enterprise data ar- HausiTerm ExpiringA. Müller, 2018: Dimitrios Ann DeMarle,Serpanos, Fred Forrest Douglis, J. Shullcomputer.org Vladimir Getov, Bruce M. achieving greater business value from that data is management, as well as other 4. R.Uncharted Bean, “Chief Corporate Data Officers Frontier,” Blaze chitecture, and customer experience man- TermMcMillin, Expiring Cecilia 2018: Metra, Ann Kunio DeMarle, Uchiyama, Fred Douglis, Stefano VladimirZanero Getov, Bruce M. factors such as size and industry UnchartedWall Street J.,Corporate 21 Apr. 2014; Frontier,” http:// agement strategies. His interests include IEEE BOARD OF DIRECTORS common to all the manifestations of the CDO role. McMillin,Term Expiring Cecilia 2019: Metra, Saurabh Kunio Bagchi, Uchiyama, Leila Stefano De PresidentFloriani, Zanero David & CEO: S. Ebert,Karen JillBartleson; I. Gostin, President-Elect: James Jefferies; Past President: sector. (Highly regulated indus- Wallblogs.wsj.com/cio/2014/04/21/chief Street J., 21 Apr. 2014; http:// customer experience analytics, knowledge TermWilliam Expiring Gropp, 2019:Sumi SaurabhHelal, Avi Bagchi, Mendelson Leila De BarryFloriani, L. Shoop;David S.Secretary: Ebert, Jill William I. Gostin, Walsh; Treasurer: John W. Walz; Director & tries might lean more toward blogs.wsj.com/cio/2014/04/21/chief-data-officers-blaze-uncharted management,William Gropp, structured Sumi andHelal, unstructured Avi Mendelson President, IEEE-USA: Karen Pedersen; Director & President, Standards Association: EXECUTIVE STAFF a governance model, whereas -data-officers-blaze-uncharted-corporate-frontier/. data systems and strategy, and machine Forrest Don Wright; Director & VP, Educational Activities: S.K. Ramesh; Director Executive Director: Angela R. Burgess; Director, Governance & Associate Executive phasis on data sources for “stress practices are executive-led, ac- those in digital startups might 5. -corporate-frontier/.T. Chakravorty, “CDO vs. CIO: learning.EXECUTIVE Contact him STAFF at [email protected]. & VP, Membership and Geographic Activities: Mary Ellen Randall; Director & VP, Director: Anne Marie Kelly; Director, Finance & Accounting: Sunny Hwang; tests” that financial institutions countable, and sustainable.”4 His focus on business value.) 5. T.Who Chakravorty, Does What,” “CDO CIO vs. fromCIO: Executive Director: Angela R. Burgess; Director,Publication Governance Services & Associate and Products: Executive Samir El-Ghazaly; Director & VP, Technical Director:Director, InformationAnne Marie Kelly;Technology Director, & Services: Finance &Sumit Accounting: Kacker; Director,Sunny Hwang; Membership are now required to conduct based research shows that • It is important to align with and WhoIDG, 9Does July What,”2014; www.cio.in/CIO from Read your subscriptions Activities: Marina Ruggieri; Director & Delegate Division V: Harold Javid; Director Director,Development: Information Eric Berkowitz; Technology Director, & Services: Products Sumit& Delegate& Services: Kacker; Division Director,Evan M.VIII: Butterfield;Membership Dejan S. Miloji čić on the Dodd-Frank Act. The first delineate clear responsibilities IDG,opinions/cdo-vs-cio-who-does 9 July 2014; www.cio.in/ through the myCS publi- Development:Director, Sales Eric & Marketing: Berkowitz; Chris Director, Jensen Products & Services: Evan M. Butterfield; year of developing data governance • 65 percent of CDO positions among related roles such as CIO opinions/cdo-vs-cio-who-does-what. cations portal at http:// revised 2 Dec. 2016 Director, Salesmycs.computer.org. & Marketing: Chris Jensen was foundational, setting standards have been established just with- and CTO. 6. -what.Top Performers Appoint Chief Data Of- COMPUTER SOCIETY OFFICES and policies. Zions appointed a re- in the past three years; • Data is the foundation of digi- 6. Topficers Performers, Forrester, Appoint 20 Aug. Chief 2015; Data www. Of- COMPUTERWashington, D.C.: SOCIETY 2001 L OFFICESSt., Ste. 700, Washington, D.C. 20036-4928 presentative for each of the 10 sub- • 64 percent of CDOs report to tal-born companies and essen- ficersforrester.com/report/Top+Perform, Forrester, 20 Aug. 2015; www. Washington,Phone: +1 202 D.C.: 371 20010101 L• Fax:St., Ste. +1 700,202 728Washington, 9614 • Email: D.C. [email protected] ject areas on the data governance a business executive—CEO, tial for the transformation of forrester.com/report/Top+Performers+Appoint+Chief+Data+Officers/ LosPhone: Alamitos: +1 202 10662 371 0101 Los • Vaqueros Fax: +1 202Circle, 728 Los 9614 Alamitos, • Email: CA [email protected] 90720 computer.org/ITPro 69 Phone: +1 714 821 8380 • Email: [email protected] council. “Interest and participa- chief operating officer (COO), traditional firms. The role of ers+Appoint+Chief+Data+Officers/-/E-RES123064. Los Alamitos: 10662 Los Vaqueros Circle, Los Alamitos, CA 90720 Phone:Membership +1 714 & 821Publication 8380 • Email: Orders [email protected] tion from the executive level was CRO, or chief marketing officer CDO will be pivotal to ensure -/E-RES123064. MembershipPhone: +1 800 & 272Publication 6657 • Fax: Orders +1 714 821 4641 • Email: [email protected] high,” Thomas reported. “Custo- (CMO); and the agility needed for future Seth Earley is CEO of Earley Information Phone:Asia/Pacific: +1 800 Watanabe 272 6657 Building,• Fax: +1 1-4-2 714 821Minami-Aoyama, 4641 • Email: Minato-ku,[email protected] Tokyo 107- mers want companies to be re- • 36 percent of CDOs report to a growth and success. SethScience Earley (www.earley.com). is CEO of Earley He’s Information an expert Asia/Pacific:0062, Japan • WatanabePhone: +81 Building, 3 3408 31181-4-2 • Minami-Aoyama, Fax: +81 3 3408 Minato-ku,3553 • Email: Tokyo tokyo.ofc@ 107- Sciencein knowledge (www.earley.com). processes, enterprise He’s an data expert ar- sponsive and have their informa- technology executive—CIO or 0062,computer.org Japan • Phone: +81 3 3408 3118 • Fax: +81 3 3408 3553 • Email: tokyo.ofc@ tion available quickly.” CTO. Forrester research has found chitecture,in knowledge and processes, customer enterprise experience dataman ar- computer.org In the course of developing its that 54 percent of top performing chitecture,agement strategies. and customer His experience interests includeman- IEEE BOARD OF DIRECTORS governance program, the team Ultimately, the impact of the role organizations (those with revenue agement customer strategies. experience Hisanalytics, interests knowledge include IEEEPresident BOARD & CEO: OF Karen DIRECTORS Bartleson; President-Elect: James Jefferies; Past President: Barry L. Shoop; Secretary: William Walsh; Treasurer: John W. Walz; Director & also looked at its architecture and will be measured by how well the growth greater than 10 percent) customermanagement, experience structuredanalytics, and unstructured knowledge President & CEO: Karen Bartleson; President-Elect: James Jefferies; Past President: BarryPresident, L. Shoop; IEEE-USA: Secretary: Karen William Pedersen; Walsh; Director Treasurer: & President, John W. Standards Walz; Director Association: & evaluated such issues as whether CDO has transformed the organi- have appointed CDOs versus 33 management,data systems structuredand strategy, and andunstructured machine President,Forrest Don IEEE-USA: Wright; Director Karen Pedersen; & VP, Educational Director & Activities: President, S.K. Standards Ramesh; Association: Director it was keeping too much data in zation through the use of data. percent of low performers (those datalearning. systems Contact and him strategy, at [email protected]. and machine Forrest& VP, Membership Don Wright; and Director Geographic & VP, EducationalActivities: Mary Activities: Ellen Randall; S.K. Ramesh; Director Director & VP, the data warehouse. Thomas em- CIOs have traditionally been with less than 4 percent revenue learning. Contact him at [email protected]. &Publication VP, Membership Services and and Geographic Products: SamirActivities: El-Ghazaly; Mary Ellen Director Randall; & VP, Director Technical & VP, phasized that “new technologies charged with developing and man- growth).6 Clearly, the CDO is a Read your subscriptions PublicationActivities: Marina Services Ruggieri; and Products: Director Samir & Delegate El-Ghazaly; Division Director V: Harold & VP, Javid; Technical Director were required in order to establish aging the enterprise’s information role that can provide significant Readthrough your the myCSsubscriptions publi- Activities:& Delegate MarinaDivision Ruggieri; VIII: Dejan Director S. Miloji & Delegatečić Division V: Harold Javid; Director data governance.” Among the new infrastructure. This is likely to value to the enterprise. Given to- throughcations portalthe myCS at http:// publi- &revised Delegate 2 Dec. Division 2016 VIII: Dejan S. Milojičić tools was a dictionary for tagging remain the case, but according to day’s data-driven environment, mycs.computer.org.cations portal at http:// revised 2 Dec. 2016 mycs.computer.org.

32 Computing Edge June 2017 68 IT Pro January/February 2017 computer.org/ITPro 69 computer.org/ITPro 69

Untitled-15 33 5/15/17 7:30 PM this role is likely to be on a growth PURPOSE: The IEEE Computer Society is the world’s largest association of computing trajectory for some time. professionals and is the leading provider of technical information in the field. MEMBERSHIP: Members receive the monthly magazine Computer, discounts, and References opportunities to serve (all activities are led by volunteer members). Membership is 1. “IDC Reveals Worldwide Digital open to all IEEE members, affiliate society members, and others interested in the Transformation Predictions; Kicks computer field. Off IDC FutureScape Web Con- OMBUDSMAN: Email [email protected]. ference,” IDC press release, 4 COMPUTER SOCIETY WEBSITE: www.computer.org Nov. 2015; www.idc.com/getdoc. jsp?containerId=prUS40553515. Next Board Meeting: 30 January–3 February 2017, Anaheim, CA, USA 2. L.T. Moss and S. Adelman, “The Role of Chief Data Offi- EXECUTIVE COMMITTEE cer in the 21st Century,” Data In- President: Jean-Luc Gaudiot sight and Social Bull., vol. 13, no. 2; President-Elect: Hironori Kasahara; Past President: Roger U. Fujii; Secretary: Forrest Shull; First VP, Treasurer: David Lomet; Second VP, Publications: Gregory http://experts.cutter.com/acton/ T. Byrd; VP, Member & Geographic Activities: Cecilia Metra; VP, Professional & attachment/19169/f-006c/1/-/-/-/-/ Educational Activities: Andy T. Chen; VP, Standards Activities: Jon Rosdahl; V P, Role%20of%20the%20Chief%20 Technical & Conference Activities: Hausi A. Müller; 2017–2018 IEEE Director & Data%20Officer.pdf. Delegate Division VIII: Dejan S. Milojičić; 2016–2017 IEEE Director & Delegate 3. Understanding the Chief Data Officer Division V: Harold Javid; 2017 IEEE Director-Elect & Delegate Division V-Elect: Role, Gartner, 18 Feb. 2015; www. John W. Walz gartner.com/smarterwithgartner/ this role is likely to be on a growth understanding-the-chief-data BOARD OF GOVERNORS PURPOSE: The IEEE Computer Society is the world’sTerm largest Expiring association 2017: Alfredoof computing Benso, Sy-Yen Kuo, Ming C. Lin, Fabrizio Lombardi, trajectory for some time. -officer-role/. professionals and is the leading provider of technicalHausi information A. Müller, in Dimitrios the field. Serpanos, Forrest J. Shull 4. R. Bean, “Chief Data Officers Blaze MEMBERSHIP: Members receive the monthly magazineTerm Expiring Computer 2018:, discounts, Ann DeMarle, and Fred Douglis, Vladimir Getov, Bruce M. References Uncharted Corporate Frontier,” opportunities to serve (all activities are led by volunteerMcMillin, members). Cecilia Metra,Membership Kunio isUchiyama, Stefano Zanero 1. “IDC Reveals Worldwide Digital Wall Street J., 21 Apr. 2014; http:// open to all IEEE members, affiliate society members,Term and Expiring others interested2019: Saurabh in the Bagchi, Leila De Floriani, David S. Ebert, Jill I. Gostin, Transformation Predictions; Kicks computerblogs.wsj.com/cio/2014/04/21/chief field. William Gropp, Sumi Helal, Avi Mendelson Off IDC FutureScape Web Con- OMBUDSMAN:-data-officers-blaze-uncharted Email [email protected]. ference,” IDC press release, 4 -corporate-frontier/. EXECUTIVE STAFF COMPUTER SOCIETY WEBSITE: www.computer.org Nov. 2015; www.idc.com/getdoc. 5. T. Chakravorty, “CDO vs. CIO: Executive Director: Angela R. Burgess; Director, Governance & Associate Executive Director: Anne Marie Kelly; Director, Finance & Accounting: Sunny Hwang; jsp?containerId=prUS40553515. NextWho Board Does Meeting: What,” 30 January–3 CIO Februaryfrom 2017, Anaheim, CA, USA Director, Information Technology & Services: Sumit Kacker; Director, Membership 2. L.T. Moss and S. Adelman, IDG, 9 July 2014; www.cio.in/ Development: Eric Berkowitz; Director, Products & Services: Evan M. Butterfield; “The Role of Chief Data Offi- EXECUTIVEopinions/cdo-vs-cio-who-does COMMITTEE Director, Sales & Marketing: Chris Jensen cer in the 21st Century,” Data In- President:-what. Jean-Luc Gaudiot sight and Social Bull., vol. 13, no. 2; 6.President-Elect: Top Performers Hironori Appoint Kasahara;Chief Data Past Of- President:COMPUTER Roger U. Fujii; SOCIETY Secretary: OFFICES Forrest Shull; First VP, Treasurer: David Lomet; Second VP, Publications: Gregory http://experts.cutter.com/acton/ ficers, Forrester, 20 Aug. 2015; www. Washington, D.C.: 2001 L St., Ste. 700, Washington, D.C. 20036-4928 T. Byrd; VP, Member & Geographic Activities: Cecilia Metra; VP, Professional & attachment/19169/f-006c/1/-/-/-/-/ forrester.com/report/Top+Perform Phone: +1 202 371 0101 • Fax: +1 202 728 9614 • Email: [email protected] Educational Activities: Andy T. Chen; VP, Standards Activities: Jon Rosdahl; V P, Data analytics Role%20of%20the%20Chief%20 ers+Appoint+Chief+Data+Officers/ Los Alamitos: 10662 Los Vaqueros Circle, Los Alamitos, CA 90720 Technical & Conference Activities: Hausi A. Müller;Phone: 2017–2018 +1 714 IEEE821 8380 Director • Email: & [email protected] Data%20Officer.pdf. -/E-RES123064. Delegate Division VIII: Dejan S. Milojičić; 2016–2017Membership IEEE Director & Publication & Delegate Orders 3. Understanding the Chief Data Officer Division V: Harold Javid; 2017 IEEE Director-ElectPhone: & Delegate +1 800 Division 272 6657 V-Elect • Fax:: +1 714 821 4641 • Email: [email protected] major technology upgrades, such data that is much more user Tamal Chakravorty, director of IT this role is likely to be on a growth Role, Gartner, 18 Feb. 2015; www. Seth Earley is CEO of Earley Information PURPOSE: The IEEE Computer Society is the world’sJohn W.largest Walz association of computing Asia/Pacific: Watanabe Building, 1-4-2 Minami-Aoyama, Minato-ku, Tokyo 107- as costly enterprise resource plan- friendly and accessible to business and Test for Ericsson Global Ser- thistrajectory role is for likely some to betime. on a growth gartner.com/smarterwithgartner/ Science (www.earley.com). He’s an expert professionalsPURPOSE: The and IEEE is the Computer leading Societyprovider is ofthe technical world’sBOARD largestinformation OF associationGOVERNORS in the field.of computing 0062, Japan • Phone: +81 3 3408 3118 • Fax: +81 3 3408 3553 • Email: tokyo.ofc@ ning replacements), the organiza- users than the previous solution. vices, the CDO needs to “know trajectory for some time. understanding-the-chief-data in knowledge processes, enterprise data ar- computer.org professionalsMEMBERSHIP: and Members is the leading receive provider the monthly of technical magazineTerm informationExpiring Computer 2017: in, discounts,the Alfredo field. Benso, and Sy-Yen Kuo, Ming C. Lin, Fabrizio Lombardi, References -officer-role/. chitecture, and customer experience man- tion will need very different types Thomas believes that this approach how to find the required informa- MEMBERSHIP:opportunities to Membersserve (all receiveactivities the are monthly led by magazinevolunteer members).Computer, discounts,Membership and is 5 Hausi A. Müller, Dimitrios Serpanos, Forrest J. ShullIEEE BOARD OF DIRECTORS of CDOs and CDO agendas. to data, maturity in governance, tion to fuel business growth.” References 1. “IDC Reveals Worldwide Digital 4. openR. Bean, to all “Chief IEEE members, Data Officers affiliate Blaze society members,agement andstrategies. others interestedHis interests in the include opportunities to serve (all activities are led by volunteerTerm Expiring members). 2018: Membership Ann DeMarle, is Fred Douglis,President Vladimir &Getov, CEO: Bruce Karen M. Bartleson; President-Elect: James Jefferies; Past President: However, the core responsibility and ability to effect cultural 1. Transformation“IDC Reveals Worldwide Predictions; Digital Kicks Uncharted Corporate Frontier,” customer experience analytics, knowledge opencomputer to all field. IEEE members, affiliate society members,McMillin, and Cecilia others Metra,interested Kunio in theUchiyama, StefanoBarry Zanero L. Shoop; Secretary: William Walsh; Treasurer: John W. Walz; Director & OffTransformation IDC FutureScape Predictions; Web KicksCon- Wall Street J., 21 Apr. 2014; http:// management, structured and unstructured for data, data quality, and achiev- change provides a competitive computerOMBUDSMAN: field. Email [email protected] Expiring 2019: Saurabh Bagchi, Leila De Floriani,President, David IEEE-USA: S. Ebert, KarenJill I. Gostin, Pedersen; Director & President, Standards Association: ing greater business value from that advantage among peers. (See a fu- he interviews I conducted ference,”Off IDC FutureScape IDC press release,Web Con 4- OMBUDSMAN:blogs.wsj.com/cio/2014/04/21/chief Email [email protected]. data systems and strategy, and machine Forrest Don Wright; Director & VP, Educational Activities: S.K. Ramesh; Director COMPUTER SOCIETY WEBSITE: www.computer.orgWilliam Gropp, Sumi Helal, Avi Mendelson data is common to all the manifes- ture column for the full interview with leaders knowledgeable Nov.ference,” 2015; IDC www.idc.com/getdoc. press release, 4 -data-officers-blaze-uncharted learning. Contact him at [email protected]. & VP, Membership and Geographic Activities: Mary Ellen Randall; Director & VP, COMPUTER SOCIETY WEBSITE: www.computer.org tations of the CDO role. with Brandon Thomas.) about the CDO role point jsp?containerId=prUS40553515.Nov. 2015; www.idc.com/getdoc. Next-corporate-frontier/. Board Meeting: 30 January–3 February 2017,EXECUTIVE Anaheim, CA, STAFF USA Publication Services and Products: Samir El-Ghazaly; Director & VP, Technical T Executive Director: Angela R. Burgess; Director, Activities:Governance Marina & Associate Ruggieri; Executive Director & Delegate Division V: Harold Javid; Director to several observations: 2. L.T.jsp?containerId=prUS40553515. Moss and S. Adelman, 5. NextT. Chakravorty, Board Meeting: “CDO 30 January–3 vs. CIO: February 2017, Anaheim, CA,Read USA your subscriptions Director: Anne Marie Kelly; Director, Finance & Accounting:& Delegate Division Sunny Hwang; VIII: Dejan S. Milojičić Zions Bancorporation’s Changing Trends for CDOs 2. “TheL.T. MossRole ofand Chief S. DataAdelman, Offi- EXECUTIVEWho Does COMMITTEEWhat,” CIO from through the myCS publi- Director, Information Technology & Services: Sumit Kacker; Director, Membership Journey According to Yang Lee of North- • The role is new. “The role of cer“The in theRole 21st of Century,”Chief Data Data Offi In- EXECUTIVEPresident:IDG, 9 JulyJean-Luc COMMITTEE 2014; Gaudiot www.cio.in/ cations portal at http:// revised 2 Dec. 2016 Development: Ericmycs.computer.org. Berkowitz; Director, Products & Services: Evan M. Butterfield; Brandon Thomas, CDO of Zions eastern University, co-director CDO is in its early stages, much sightcer in and the Social 21st Bull.Century,”, vol. 13, Data no. In 2;- President:President-Elect:opinions/cdo-vs-cio-who-does Jean-Luc Hironori Gaudiot Kasahara; Past President: Roger U. Fujii; Secretary: Forrest Shull; First VP, Treasurer: David Lomet;Director, Second VP,Sales Publications: & Marketing: Gregory Chris Jensen Bancorporation, described the path for MIT’s Chief Data Officer Re- like the role of CMO was 20 years sighthttp://experts.cutter.com/acton/ and Social Bull., vol. 13, no. 2; President-Elect:-what. Hironori Kasahara; Past President: Roger U. Fujii; Secretary: ForrestT. Byrd; Shull; VP, Member First VP, & Treasurer:Geographic David Activities: Lomet; Cecilia Second Metra; VP, Publications: VP, Professional Gregory & of his organization in getting con- search Program, “Traditionally, ago,” according to Rich Wendell, attachment/19169/f-006c/1/-/-/-/-/http://experts.cutter.com/acton/ 6. Top Performers Appoint Chief Data Of- COMPUTER SOCIETY OFFICES T.Educational Byrd; VP, MemberActivities: & AndyGeographic T. Chen; Activities: VP, Standards Cecilia Activities: Metra; VP, Jon Professional Rosdahl; V & P, trol of data and processes. A cata- data practices were led by middle CEO and CDO of Tellic, a firm Role%20of%20the%20Chief%20attachment/19169/f-006c/1/-/-/-/-/ ficers, Forrester, 20 Aug. 2015; www. Washington, D.C.: 2001 L St., Ste. 700, Washington, D.C. 20036-4928 computer.org/ITPro 69 EducationalTechnical & Activities:Conference Andy Activities: T. Chen; Hausi VP, A.Standards Müller; 2017–2018Activities: IEEE Jon Rosdahl; Director V& P, lyst for launching his company’s managers, lacking executive col- specializing in pharma data sci- Data%20Officer.pdf.Role%20of%20the%20Chief%20 forrester.com/report/Top+Perform Phone:This +1article 202 371originally 0101 • Fax: appeared +1 202 728in 9614 • Email: [email protected] TechnicalDelegate Division & Conference VIII: Dejan Activities: S. Miloji Hausičić; A.2016–2017 Müller; 2017–2018IEEE Director IEEE & Director Delegate & journey was the increased em- laboration. The emerging CDO ences as a service. 3. UnderstandingData%20Officer.pdf. the Chief Data Officer ers+Appoint+Chief+Data+Officers/ LosIT Alamitos: Professional, 10662 Losvol. Vaqueros 19, no. 1,Circle, 2017. Los Alamitos, CA 90720 DivisionDelegate V:Division Harold VIII: Javid; Dejan 2017 S.IEEE Miloji Director-Electčić; 2016–2017Phone: & IEEEDelegate +1 714 Director 821 Division 8380 & Delegate •V-Elect Email: : [email protected] • The role is likely to remain fluid 3. RoleUnderstanding, Gartner, the18 ChiefFeb. 2015;Data Officerwww. -/E-RES123064. JohnDivision W. WalzV: Harold Javid; 2017 IEEE Director-ElectMembership & Delegate & PublicationDivision V-Elect Orders: for a number of years. Rolegartner.com/smarterwithgartner/, Gartner, 18 Feb. 2015; www. John W. Walz Phone: +1 800 272 6657 • Fax: +1 714 821 4641 • Email: [email protected] • How the role is expressed de- gartner.com/smarterwithgartner/understanding-the-chief-data SethBOARD Earley isOF CEO GOVERNORS of Earley Information Term Expiring 2017: Alfredo Benso, Sy-Yen Kuo,Asia/Pacific: Ming C. Lin, Watanabe Fabrizio Lombardi,Building, 1-4-2 Minami-Aoyama, Minato-ku, Tokyo 107- The core responsibility for data, data quality, and pends on the company’s matu- -officer-role/.understanding-the-chief-data ScienceBOARD (www.earley.com). OF GOVERNORS He’s an expert TermHausi ExpiringA. Müller, 2017: Dimitrios Alfredo Serpanos, Benso, Sy-YenForrest Kuo, J. Shull0062, Ming Japan C. Lin, • Phone: Fabrizio +81 Lombardi, 3 3408 3118 • Fax: +81 3 3408 3553 • Email: tokyo.ofc@ rity level with respect to data 4. -officer-role/.R. Bean, “Chief Data Officers Blaze in knowledge processes, enterprise data ar- HausiTerm ExpiringA. Müller, 2018: Dimitrios Ann DeMarle,Serpanos, Fred Forrest Douglis, J. Shullcomputer.org Vladimir Getov, Bruce M. achieving greater business value from that data is management, as well as other 4. R.Uncharted Bean, “Chief Corporate Data Officers Frontier,” Blaze chitecture, and customer experience man- TermMcMillin, Expiring Cecilia 2018: Metra, Ann Kunio DeMarle, Uchiyama, Fred Douglis, Stefano VladimirZanero Getov, Bruce M. factors such as size and industry UnchartedWall Street J.,Corporate 21 Apr. 2014; Frontier,” http:// agement strategies. His interests include IEEE BOARD OF DIRECTORS common to all the manifestations of the CDO role. McMillin,Term Expiring Cecilia 2019: Metra, Saurabh Kunio Bagchi, Uchiyama, Leila Stefano De PresidentFloriani, Zanero David & CEO: S. Ebert,Karen JillBartleson; I. Gostin, President-Elect: James Jefferies; Past President: Wallblogs.wsj.com/cio/2014/04/21/chief Street J., 21 Apr. 2014; http:// customer experience analytics,EXECUTIVE knowledge STAFF sector. (Highly regulated indus- TermWilliam Expiring Gropp, 2019:Sumi SaurabhHelal, Avi Bagchi, Mendelson Leila De Floriani, David S. Ebert, Jill I. Gostin, Executive Director: Angela R.Barry Burgess; L. Shoop; Director, Secretary: Governance William & Associate Walsh; Treasurer: John W. Walz; Director & tries might lean more toward -data-officers-blaze-unchartedblogs.wsj.com/cio/2014/04/21/chief management,William Gropp, structured Sumi andHelal, unstructured Avi Mendelson President, IEEE-USA: Karen Pedersen; Director & President, Standards Association: EXECUTIVE STAFFExecutive Director: Anne Marie Kelly; Director, Finance & Accounting: a governance model, whereas -data-officers-blaze-uncharted-corporate-frontier/. data systems and strategy, and machine Forrest Don Wright; Director & VP, Educational Activities: S.K. Ramesh; Director PURPOSE: The IEEE Computer Society is the world’s largestEXECUTIVEExecutive association Director: STAFFSunny Angela Hwang; R. Burgess; Director, Director, Information Governance Technology & Associate & Services: Executive Sumit phasis on data sources for “stress practices are executive-led, ac- those in digital startups might of computing5. -corporate-frontier/.T. Chakravorty, professionals and“CDO is the vs.leading CIO: provider learning. of technical Contact him atKacker; [email protected]. Director, Membership& Development:VP, Membership Eric andBerkowitz; Geographic Director, Activities: Mary Ellen Randall; Director & VP, 4 ExecutiveDirector: Anne Director: Marie Angela Kelly; R.Director, Burgess; Finance Director, & Accounting:Governance &Sunny Associate Hwang; Executive tests” that financial institutions countable, and sustainable.” His focus on business value.) information 5. T.Who Chakravorty, in theDoes field. What,” “CDO CIO vs. fromCIO: Products & Services: Evan M.Publication Butterfield; Director,Services Salesand Products: & Marketing: Samir El-Ghazaly; Director & VP, Technical Director:Director, InformationAnne Marie Kelly;Technology Director, & Services: Finance &Sumit Accounting: Kacker; Director,Sunny Hwang; Membership are now required to conduct based research shows that • It is important to align with and MEMBERSHIP:WhoIDG, 9Does Members July What,”2014; receive www.cio.in/theCIO monthly from magazine Computer, Read yourChris subscriptions Jensen Activities: Marina Ruggieri; Director & Delegate Division V: Harold Javid; Director Director,Development: Information Eric Berkowitz; Technology Director, & Services: Products Sumit& Delegate& Services: Kacker; Division Director,Evan M.VIII: Butterfield;Membership Dejan S. Miloji čić on the Dodd-Frank Act. The first delineate clear responsibilities discounts,IDG,opinions/cdo-vs-cio-who-does and 9opportunities July 2014; to servewww.cio.in/ (all activities are led by volunteerthrough the myCS publi- members). Membership is open to all IEEE members, affiliateDevelopment:Director, society Sales Eric & Marketing: Berkowitz;COMPUTER Chris Director, SOCIETYJensen Products OFFICES & Services: Evan M. Butterfield; year of developing data governance • 65 percent of CDO positions among related roles such as CIO opinions/cdo-vs-cio-who-does-what. cations portal at http:// revised 2 Dec. 2016 members, and others interested in the computer field. Director, Sales & Marketing:Washington, Chris D.C.: Jensen 2001 L St., Ste. 700, Washington, D.C. 20036-4928 COMPUTERmycs.computer.org. SOCIETY OFFICES was foundational, setting standards have been established just with- and CTO. OMBUDSMAN: 6. -what.Top Performers Email [email protected]. Appoint Chief Data Of- Phone: +1 202 371 0101 • Fax: +1 202 728 9614 • Email: [email protected] and policies. Zions appointed a re- in the past three years; • Data is the foundation of digi- COMPUTER 6. Topficers Performers ,SOCIETY Forrester, WEBSITE:Appoint 20 Aug. Chief www.computer.org2015; Data www. Of- COMPUTERWashington, D.C.: SOCIETY 2001Los Alamitos: L OFFICESSt., Ste. 10662 700, Washington,Los Vaqueros D.C.Circle, 20036-4928 Los Alamitos, CA 90720 Washington,Phone: +1 202 D.C.: 371 20010101Phone: L• Fax:St.,+1 Ste.714 +1 821 700,202 8380 728Washington, •9614 Email: • Email: [email protected] D.C. [email protected] presentative for each of the 10 sub- • 64 percent of CDOs report to tal-born companies and essen- Next Boardficersforrester.com/report/Top+Perform, Forrester,Meeting: 12–1720 Aug. June 2015; 2017, www. Phoenix, AZ, USA ject areas on the data governance a business executive—CEO, tial for the transformation of forrester.com/report/Top+Performers+Appoint+Chief+Data+Officers/ Phone:Los Alamitos: +1 202 10662 371 0101MEMBERSHIP Los • Vaqueros Fax: +1 & 202Circle, PUBLICATION 728 Los 9614 Alamitos, • Email: ORDERS CA [email protected] 90720 computer.org/ITPro 69 Phone: +1 714 821 8380 • Email: [email protected] council. “Interest and participa- chief operating officer (COO), traditional firms. The role of EXECUTIVEers+Appoint+Chief+Data+Officers/-/E-RES123064. COMMITTEE Los Alamitos: 10662Phone: Los Vaqueros +1 800 272 Circle, 6657 Los• Fax: Alamitos, +1 714 821 CA 464190720 • Email: [email protected] President: Jean-Luc Gaudiot; President-Elect: Hironori Kasahara;Phone:Membership +1 Past 714 & 821Publication 8380Asia/Pacific: • Email: Orders Watanabe [email protected] Building, 1-4-2 Minami-Aoyama, Minato-ku, Tokyo tion from the executive level was CRO, or chief marketing officer CDO will be pivotal to ensure -/E-RES123064. President: Roger U. Fujii; Secretary: Forrest Shull; First VP,MembershipPhone: Treasurer: +1 800 & 272Publication 107-0062,6657 • Fax: Orders Japan +1 •714 Phone: 821 4641+81 3• 3408Email: 3118 [email protected] • Fax: +81 3 3408 3553 • high,” Thomas reported. “Custo- (CMO); and the agility needed for future Seth Earley is CEO of Earley Information David Lomet; Second VP, Publications: Gregory T. Byrd; Phone:Asia/Pacific:VP, Member +1 800 & Watanabe 272 Email:6657 Building,• [email protected]: +1 1-4-2 714 821Minami-Aoyama, 4641 • Email: Minato-ku,[email protected] Tokyo 107- mers want companies to be re- • 36 percent of CDOs report to a growth and success. GeographicSethScience Earley (www.earley.com). Activities: is CEO Cecilia of Earley Metra;He’s Information an VP, expert Professional & Educational Asia/Pacific:0062, Japan • WatanabePhone:IEEE +81 Building, BOARD3 3408 31181-4-2 OF • Minami-Aoyama, DIRECTORSFax: +81 3 3408 Minato-ku,3553 • Email: Tokyo tokyo.ofc@ 107- Activities:Sciencein knowledge (www.earley.com).Andy processes,T. Chen; VP, enterprise Standards He’s an data Activities:expert ar- Jon Rosdahl; V P, sponsive and have their informa- technology executive—CIO or 0062,computer.org Japan • Phone: +81 3 3408 3118 • Fax: +81 3 3408 3553 • Email: tokyo.ofc@ Technical & Conference Activities: Hausi A. Müller; 2017–2018 IEEE President & CEO: Karen Bartleson; President-Elect: James Jefferies; Past tion available quickly.” CTO. Forrester research has found inchitecture, knowledge and processes, customer enterprise experience dataman ar- computer.org President: Barry L. Shoop; Secretary: William Walsh; Treasurer: John Director & Delegate Division VIII: Dejan S. Milojičić; 2016–2017IEEE BOARD IEEE OF DIRECTORS In the course of developing its that 54 percent of top performing chitecture,agement strategies. and customer His experience interests includeman- W. Walz; Director & President, IEEE-USA: Karen Pedersen; Director & Director & Delegate Division V: Harold Javid; 2017 IEEE PresidentDirector-Elect & CEO: & Karen Bartleson; President-Elect: James Jefferies; Past President: governance program, the team Ultimately, the impact of the role organizations (those with revenue customeragement strategies. experience Hisanalytics, interests knowledge include IEEE BOARD OF DIRECTORSPresident, Standards Association: Forrest Don Wright; Director & VP, Delegate Division V-Elect: John W. Walz Barry L. Shoop; Secretary: William Walsh; Treasurer: John W. Walz; Director & also looked at its architecture and will be measured by how well the growth greater than 10 percent) customermanagement, experience structuredanalytics, and unstructured knowledge President & CEO: KarenEducational Bartleson; Activities: President-Elect: S.K. Ramesh; James Director Jefferies; & VP, Past Membership President: and BOARD OF GOVERNORS BarryPresident, L. Shoop; IEEE-USA: Secretary:Geographic Karen William Pedersen; Activities: Walsh; Director Mary Treasurer: Ellen & President, Randall; John W. DirectorStandards Walz; &Director VP,Association: Publication & evaluated such issues as whether CDO has transformed the organi- have appointed CDOs versus 33 datamanagement, systems structuredand strategy, and andunstructured machine Term Expiring 2017: Alfredo Benso, Sy-Yen Kuo, Ming C.President,Forrest Lin, Fabrizio Don IEEE-USA: Wright; Services Director Karen andPedersen; & VP,Products: Educational Director Samir & El-Ghazaly;Activities: President, S.K.Director Standards Ramesh; & VP, Association: Director Technical it was keeping too much data in zation through the use of data. percent of low performers (those Lombardi,datalearning. systems Hausi Contact A.and Müller, him strategy, at Dimitrios [email protected]. and Serpanos,machine Forrest J.Forrest& Shull VP, Membership Don Wright; Activities:and Director Geographic &Marina VP, EducationalActivities: Ruggieri; Director Mary Activities: Ellen & Delegate Randall; S.K. Ramesh; Division Director Director V: &Harold VP, Javid; the data warehouse. Thomas em- CIOs have traditionally been with less than 4 percent revenue Termlearning. Expiring Contact 2018: himAnn atDeMarle, [email protected]. Fred Douglis, Vladimir&Publication VP,Getov, Membership Bruce Services Directorand and Geographic Products: & Delegate SamirActivities: Division El-Ghazaly; MaryVIII: Dejan Ellen Director S.Randall; Miloji & VP,č iDirectorć Technical & VP, phasized that “new technologies charged with developing and man- growth).6 Clearly, the CDO is a M. McMillin, Cecilia ReadMetra, your Kunio subscriptions Uchiyama, Stefano ZaneroPublicationActivities: Marina Services Ruggieri; and Products: Director Samir & Delegate El-Ghazaly; Division Director V: Harold & VP, Javid; Technical Director were required in order to establish aging the enterprise’s information role that can provide significant Term Expiring 2019:throughRead Saurabh your the Bagchi, myCSsubscriptions Leila publi- De Floriani, David&Activities: Delegate S. Ebert, MarinaDivision Jill Ruggieri; VIII: Dejan Director S. Miloji & Delegatečić Division V: Harold Javid; Director data governance.” Among the new infrastructure. This is likely to value to the enterprise. Given to- I. Gostin, William Gropp,throughcations Sumi portalthe Helal, myCS at Avihttp:// publi- Mendelson &revised Delegate 2 Dec. Division 2016 VIII: Dejan S. Milojičić revised 26 Jan. 2017 tools was a dictionary for tagging remain the case, but according to day’s data-driven environment, mycs.computer.org.cations portal at http:// revised 2 Dec. 2016 mycs.computer.org.

www.computer.org/computingedge 33 68 IT Pro January/February 2017 computer.org/ITPro 69 computer.org/ITPro 69

Untitled-15 33 5/15/17 7:30 PM CLOUD TIDBITS

terest in PaaS. Products such as AWS Lambda offering many features that would normally be PaaS Death associated with a PaaS provider, Microsoft and Google have serverless offerings as well.

While PaaS offerings vary greatly, most provide Watch? facilities for application design, deployment, testing, and self-provisioned hosting. More advanced services may exist in the offering as well, such as team collaboration, database integration, middleware services, Web service integration, storage, PAAS, OR PLATFORM-AS-A-SERVICE, IS state management, and version management ser- SOMETHING THAT’S NOT USED AS MUCH AS vices. However, these patterns are not at all consis- INITIALLY PREDICTED WITHIN ENTERPRISES. tent from PaaS provider to PaaS provider, and this There are many reasons why. However, the core is- leads to some major confusion in the emerging PaaS sues seem to be: space, as well as amongst enterprises that are looking to implement PaaS. • Strong tools within Infrastructure as a Service At the heart of the problem is the fact that PaaS (IaaS) providers such as Amazon Web Services is today’s most ill-defined area of cloud computing. (AWS) and Microsoft Azure seem to be driving The approaches, features, and definitions vary widely, interest in those platforms. While they may have with many PaaS providers offering a specific focus. PaaS as part of their platform, most developers This may include support for specific programming opt for the core IaaS services, such as storage languages, such as Salesforce.com’s Heroku, support and compute platforms. for Ruby, Node.js, Python, or Java, or, perhaps tight • Developers do not like to be placed within a integration with major databases, such as Oracle’s sandbox. Many PaaS providers have restrictions, Cloud Platform. Others focus on a specific standard, such as tools, databases, and programming lan- such as Cloud Foundry. Most take their own propri- guages, that developers typically don’t like. ety and closed approach to PaaS, no matter if they say • IaaS seems to be a better fit than PaaS for De- they are following a standard or not. vOps organizations, considering that they are Also, a dark side of PaaS emerged in 2016. Some able to replicate the platforms of IaaS providers, consider PaaS too complicated and too limiting for such as the ones that are on premises as well. most development efforts, and developers. Indeed, • The rise of serverless computing within all ma- most developers working on IaaS cloud today, and jor IaaS provider has led to a dimensioning in- most of the interest in hiring has been with IaaS clouds. Thus, there is not market incentive to invest in PaaS or PaaS training given the IaaS alternatives. Most PaaS offerings put the developer into a sandbox, with only the features and functions that the PaaS provider furnishes to build and deploy applications. While this makes development an easy and controlled process, many developers need to gain access to the resources and tools required to DAVID S. support specific features, such as remote and native LINTHICUM APIs, as well as middleware and database services. This limitation may be a big problem for some Cloud Technology Partners developers, and not an issue with others. It does [email protected] cause some concern. Many enterprises consider PaaS their new platform for application develop-

d1tid.indd 6 5/15/17 5:29 PM CLOUD TIDBITS

terest in PaaS. Products such as AWS Lambda ment, testing, deployment, and operations. It needs offering many features that would normally be to do everything, and it might not. This disap- PaaS Death associated with a PaaS provider, Microsoft and pointing set of facts lead to the diminished interest Google have serverless offerings as well. in PaaS in the last few years, not as much because PaaS did not live up to expectations, but because While PaaS offerings vary greatly, most provide other alternatives became more productive Watch? facilities for application design, deployment, test- On the positive side, PaaS does provide the abil- ing, and self-provisioned hosting. More advanced ity to automate much of the development and de- services may exist in the offering as well, such as ployment activities, as well as provide the developers team collaboration, database integration, middle- with the ability to offer self- and auto-provisioning ware services, Web service integration, storage, capabilities. This means that application develop- PAAS, OR PLATFORM-AS-A-SERVICE, IS state management, and version management ser- ers can focus on the applications, and not have to SOMETHING THAT’S NOT USED AS MUCH AS vices. However, these patterns are not at all consis- deal with the purchase of hardware, software, and INITIALLY PREDICTED WITHIN ENTERPRISES. tent from PaaS provider to PaaS provider, and this development tools. While this seems like an obvious There are many reasons why. However, the core is- leads to some major confusion in the emerging PaaS benefi t of cloud computing, self-provisioning is the sues seem to be: space, as well as amongst enterprises that are look- core reason that developers choose to leverage PaaS. FIGURE 1. According to Technology Business Research (TBR), Platform- ing to implement PaaS. However, there again there are sound alterna- as-a-service (PaaS) growth will likely not provide signiﬁ cant traction, • Strong tools within Infrastructure as a Service At the heart of the problem is the fact that PaaS tives on the horizon. The advent of serverless com- when compared to Software as a Service (SaaS) and Infrastructure as a (IaaS) providers such as Amazon Web Services is today’s most ill-defined area of cloud computing. puting from the major IaaS providers has solved the Service (IaaS), relative to market size (Source: tbri.com). (AWS) and Microsoft Azure seem to be driving The approaches, features, and definitions vary widely, self-provisioning problem for IaaS. This has and interest in those platforms. While they may have with many PaaS providers offering a specific focus. will lead to PaaS moving down in the market. One PaaS as part of their platform, most developers This may include support for specific programming could argue that the IaaS player are just becoming will be most interesting is to see how they morph opt for the core IaaS services, such as storage languages, such as Salesforce.com’s Heroku, support PaaS themselves. their PaaS technology as the market shifts. In the and compute platforms. for Ruby, Node.js, Python, or Java, or, perhaps tight larger view, both Microsoft and Google follow AWS • Developers do not like to be placed within a integration with major databases, such as Oracle’s The Market Speaks and the shift in PaaS should be no exception. sandbox. Many PaaS providers have restrictions, Cloud Platform. Others focus on a specific standard, As you can see by Figure 1, the growth of PaaS was such as tools, databases, and programming lan- such as Cloud Foundry. Most take their own propri- predicted to be only a fraction of Software as a ser- Deﬁ ning the PaaS Opportunity guages, that developers typically don’t like. ety and closed approach to PaaS, no matter if they say vice (SaaS), and IaaS growth, just moving from $5 PaaS providers all approach PaaS differently, but • IaaS seems to be a better fit than PaaS for De- they are following a standard or not. billion to $11 billion from 2013 to 2018, according they all provide common sets of services. It’s impor- vOps organizations, considering that they are Also, a dark side of PaaS emerged in 2016. Some to TBR (www.tbri.com). While that’s in proportion tant to understand these services before comparing able to replicate the platforms of IaaS providers, consider PaaS too complicated and too limiting for to the growth of the market, it’s well below the inter- PaaS providers for your own specifi c use cases. such as the ones that are on premises as well. most development efforts, and developers. Indeed, est level for most major cloud providers and Global Generally speaking, PaaS providers, while pro- • The rise of serverless computing within all ma- most developers working on IaaS cloud today, and 2000 enterprises. Most analysts were predicting a viding platforms for development and deployment of jor IaaS provider has led to a dimensioning in- most of the interest in hiring has been with IaaS 1/3 – 1/3 – 1/3 marketing position between PaaS, applications, are not to be confused with traditional clouds. Thus, there is not market incentive to invest IaaS, and SaaS. That has yet to come true, and like- development environments and toolsets. The focus in PaaS or PaaS training given the IaaS alternatives. ly won’t come true. when deciding to leverage PaaS should be focused Most PaaS offerings put the developer into a The core competitor to PaaS is IaaS, and that on standardizing around a specifi c platform to pro- sandbox, with only the features and functions that technology seems to be more relevant to enterprises. mote productivity by using common development the PaaS provider furnishes to build and deploy ap- Considering that most PaaS providers are invest- and deployment services, and the ability to stream- plications. While this makes development an easy ment-driven, the plug could be pulled on the smaller line development, testing, staging, and deployment, and controlled process, many developers need to players, which means only a few providers, such as in support of DevOps. Therefore, we’re talking gain access to the resources and tools required to Google, Microsoft, and AWS, will continue to pro- about a change in processes (sometimes culture), as DAVID S. support specific features, such as remote and native vide PaaS. However, their focus is clearly on their well as a change in technology, else the true value of LINTHICUM APIs, as well as middleware and database services. IaaS services, and they won’t focus on PaaS services PaaS is likely not to emerge. This limitation may be a big problem for some if those services don’t serve their purpose. I’ve defi ned PaaS as a set of layered services (see Cloud Technology Partners developers, and not an issue with others. It does Thus, the next few years will be very telling. Figure 2), including database and infrastructure ser- [email protected] cause some concern. Many enterprises consider While AWS clearly is not focusing on their PaaS, vices at the lowest layer. Moving up you’ll fi nd devel- PaaS their new platform for application develop- Microsoft and Google perhaps still are. So, what opment and deployment services, as well as services

d1tid.indd 6 5/15/17 5:29 PM d1tid.indd 7 5/15/17 5:29 PM CLOUD TIDBITS

Runtime services Monitoring Use-based accounting Provisioning Scaling

Mobile Session management Mobile Tenant management

Continuous delivery continuous integration

Code Test Stage Deploy Manage

Deployment services

Security Testing Conﬁgure management Mobile Staging

Development services Governance

Languages Code management Service catalog Mobile

Analytics OLTP Storage Compute

FIGURE 2: While all PaaS providers are a bit different, there are some common patterns that have emerged (Source: Cloud Technology Partners).

in support of DevOps, including continuous delivery players to claim additional database support, and and integration. Finally, up to runtime services that thus keep up with the big guys. deal with how applications are externalized to humans In looking at database services, you should con- through user interfaces, or to other systems through sider the number of databases offered, including tradi- APIs. Of course, security and governance services are tional, NoSQL, online transaction processing (OLTP), systemic to most PaaS offerings, with varying degrees and analytics focused. Also, how the database are in- of functionality, depending on the PaaS provider. tegrated into other levels, including development and The key message here is that enterprises should deployment services. Those that are tightly integrated not avoid investing in PaaS, but they should be care- are easier to use, as well as easier to change over time ful how they use it. A few years ago, it was common to as requirements change. However, the tradeoff in le- find enterprises that had made a major bet with PaaS veraging the database services contained in the PaaS platforms, and have since switched over to IaaS. What is a clear dependency upon that PaaS provider. scared them off was the idea of moving to a PaaS pro- Infrastructure services are, in essence, every- vider, including changing most of their applications, thing you want to see in an IaaS provider, but in and then have that provider not invest anymore in the a PaaS provider. Again, the larger players such as core PaaS services, or, worse, go out of business. Microsoft, Google, and AWS have their own IaaS Enterprises consider all their options, in terms services, and those serve as infrastructure services of how PaaS compares to IaaS, and what value exists as well for their PaaS offerings. Smaller players, in which platform. When the dust settles, we’ll find typically have the minimum amount of storage and that, for all practical purposes, PaaS is leaving the compute services to support application runtime op- market by 2018. You heard it here first. erations, and some provide deployment services to Database services typically offer access to data- external IaaS providers. Those with both PaaS and bases that are native to the PaaS provider, such as IaaS services seem to rise to the top, and this is a those offered by Microsoft, Google, or AWS. Or, in clear trend moving forward. some cases, they augment a limited number of data- Development services typically include what you base options. The PaaS provider may access a data- would expect, including support for a bundle of pro- base service that actually exists within a third party gramming languages that are quickly expanding within Data-as-a-Service provider. This allows the smaller most PaaS offerings. Also, code management services

36 Computing Edge June 2017 8 IEEE CLOUD COMPUTING WWW.COMPUTER.ORG/CLOUDCOMPUTING

d1tid.indd 8 5/15/17 5:29 PM CLOUD TIDBITS

This article originally appeared in IEEE Cloud Computing, vol. 4, no. 1, 2017. Runtime services Monitoring Use-based accounting Provisioning Scaling that are linked to configuration management, and a keep up with the features and functions that they Mobile Session management Mobile Tenant management services catalog. Offerings from some PaaS provid- need. The larger players will do what larger players ers, including Force.com, include deep support for do, and the features and functions of their PaaS of- Continuous delivery continuous integration mobile application development, and provide the fering will become “systemic” to the IaaS platform. Code Test Stage Deploy Manage ability to consider mobile deployment from develop- These include Google, Microsoft, and AWS. ment to deployment, and to runtime operations. I suspect the smaller player will be absorbed in Deployment services Deployment services includes testing, configu- the larger ones. There are a few PaaS providers that ration management, application staging, and, once can’t keep up with the larger IaaS providers, and

Security Testing Conﬁgure management Mobile Staging again, mobile deployment services. It’s at this layer selling will be their only option. This is bad news that the application code goes through basic testing for those that have built their applications on that Governance Development services procedures, and is linked with a configuration man- PaaS platform, and will have to find other homes. Languages Code management Service catalog Mobile agement system, staged, and placed into production. That will be difficult considering that the code and While we show the DevOps processes above the data are not easily ported. deployment layer, this is typically where the action Some PaaS providers will be purchased by larger Analytics OLTP Storage Compute occurs, in terms of continuous delivery and integra- SaaS player as a tool to augment their SaaS offer- tion. Thus, the services at the deployment layers of ing, much like SAP provide ABAP. SaaS providers FIGURE 2: While all PaaS providers are a bit different, there are some common patterns that have emerged PaaS providers are largely automated, perhaps even are always getting requests to changing their offer- (Source: Cloud Technology Partners). leveraging well-known configuration management ings to accommodate the way that enterprises want tools such as Chef or Puppet. to use the software. This indeed could be the right DevOps services, which are discussed below in pivot for most of the PaaS providers out in the mar- in support of DevOps, including continuous delivery players to claim additional database support, and detail as a disruptive vector, provide the automated ket that may be looking for other things to do. and integration. Finally, up to runtime services that thus keep up with the big guys. processes to achieve “continuous delivery” and “con- So, this is really about the concept of PaaS. deal with how applications are externalized to humans In looking at database services, you should continuous integration,” creating a dynamic link be- While the government defined PaaS in partnership through user interfaces, or to other systems through sider the number of databases offered, including tradi- tween those who are building the applications, and with other industry players back in 2008, the likeli- APIs. Of course, security and governance services are tional, NoSQL, online transaction processing (OLTP), those who are deploying and operating the applica- hood is that PaaS will be one of those components systemic to most PaaS offerings, with varying degrees and analytics focused. Also, how the database are in- tions. The key feature of this layer is automation of of cloud computing that fades into the background. of functionality, depending on the PaaS provider. tegrated into other levels, including development and these services, that allows for continual software re- Thus, it’s not dead as per unavailability in the mar- The key message here is that enterprises should deployment services. Those that are tightly integrated leases for largely cloud-based applications. ket, but dead in terms of the value that PaaS can not avoid investing in PaaS, but they should be care- are easier to use, as well as easier to change over time Runtime services are everything you need to op- bring to those who are building applications within ful how they use it. A few years ago, it was common to as requirements change. However, the tradeoff in le- erate an application in production, including moni- cloud-based platforms. find enterprises that had made a major bet with PaaS veraging the database services contained in the PaaS toring, tenant management, session management, The larger question is should you invest in platforms, and have since switched over to IaaS. What is a clear dependency upon that PaaS provider. provisioning, and scaling. The objective here is to keep PaaS? Or, leverage a PaaS platform to build applica- scared them off was the idea of moving to a PaaS pro- Infrastructure services are, in essence, every- the application working, and be able to dynamically tions? The best advice that I would have is to look vider, including changing most of their applications, thing you want to see in an IaaS provider, but in adjust to an increasing or changing workload. More- at what PaaS is able to do versus that of IaaS plat- and then have that provider not invest anymore in the a PaaS provider. Again, the larger players such as over, runtime services provide basic manage services, forms. If you’re building net new applications, then core PaaS services, or, worse, go out of business. Microsoft, Google, and AWS have their own IaaS such as monitoring and logging, as well as use-based PaaS could still be a fit. However, for the majority Enterprises consider all their options, in terms services, and those serve as infrastructure services accounting, so those who need to provide charge-back of applications enterprises are seeking platform ana- of how PaaS compares to IaaS, and what value exists as well for their PaaS offerings. Smaller players, and show-back are tracking that data as well. Finally, logs for their traditional systems, for development, in which platform. When the dust settles, we’ll find typically have the minimum amount of storage and mobile runtime services are offered to manage mobile and that’s IaaS. As the IaaS tools and services get that, for all practical purposes, PaaS is leaving the compute services to support application runtime op- application operations out of the PaaS. better, PaaS will die a slow death, certainly won’t be market by 2018. You heard it here first. erations, and some provide deployment services to relevant in the market anymore. Database services typically offer access to data- external IaaS providers. Those with both PaaS and Death or Not? bases that are native to the PaaS provider, such as IaaS services seem to rise to the top, and this is a So will PaaS die the death of a 1,000 cuts? PaaS those offered by Microsoft, Google, or AWS. Or, in clear trend moving forward. technology will not likely have a death event, but DAVID S. LINTHICUM is senior vice president of some cases, they augment a limited number of data- Development services typically include what you will be absorbed into larger IaaS platforms even fur- Cloud Technology Partners. He’s also Gigaom’s re- base options. The PaaS provider may access a data- would expect, including support for a bundle of pro- ther than it is already. The likelihood is that Global search analyst and frequently writes for InfoWorld base service that actually exists within a third party gramming languages that are quickly expanding within 2000 enterprises will lose interest in the smaller on deep technology subjects. Contact him at david Data-as-a-Service provider. This allows the smaller most PaaS offerings. Also, code management services players, considering them a bit risky and unable to @davidlinthicum.com

www.computer.org/computingedge 37 8 IEEE CLOUD COMPUTING WWW.COMPUTER.ORG/CLOUDCOMPUTING JANUARY/FEBRUARY 2017 IEEE CLOUD COMPUTING 9

d1tid.indd 8 5/15/17 5:29 PM d1tid.indd 9 5/15/17 5:29 PM From the Editors: Editor’s Select Predictability or Early Warning: Using Social Media in Modern Emergency Response

Marco Avvenuti • University of Pisa, Italy

Stefano Cresci and Andrea Marchetti • National Research Council (CNR), Italy

Carlo Meletti • National Institute of Geophysics and Volcanology (INGV), Italy

Maurizio Tesconi • National Research Council (CNR), Italy

ariluz Congosto and her colleagues1 the propagation speed of communication media recently mentioned the possibility of and that of seismic waves. Instead, adhering to Mpredicting natural disasters using social a more genuine interpretation, the hypothesis of media data. They support their claims by add- predicting earthquakes using social media data ing references to two works that employed can be ruled out straightaway as the relevant Twitter data for the detection of earth- data are produced and shared as a result of the quakes.2,3 Within this context, we feel that the earthquake itself. As such, it’s not available references to these two other works2,3 might before the earthquake occurs. be overstated. For instance, in our work,2 we Another source of ambiguity could be a few discussed a system for the detection and dam- references in Sakaki and his colleagues’ work3 age assessment of earthquakes in Italy. Simi- mentioning research efforts aimed at inves- larly, in Takeshi Sakaki and his colleagues’ tigating the possibility for intermediate- and work,3 they propose a system for detecting short-term earthquake prediction. However, in earthquakes as well as detecting and esti- this work,3 there’s no specific reference to the mating the trajectory of tornadoes in Japan. results of such studies, and ultimately, neither In both works, no claims are made about the the effectiveness nor the applicability of such systems’ capabilities to predict earthquake predictive techniques are supported. occurrences, but more about their ability to Over the course of more than 100 years, detect and report. We believe this opens an the potential promise of earthquake predic- even more interesting discussion. tion has attracted researchers in seismology as A potential source of ambiguity, and a pos- well as amateurs and impostors. Many theories sible reason for this claim by Congosto and have been tested and systematically rejected.4 her colleagues could lie in the meaning given Recent works in seismology have demonstrated to “predict,” when the definition “say or esti- once more that current knowledge in this field mate that a specific thing will happen in the is unable to predict earthquake occurrences in future” is interpreted as the action of figuring any way,5 let alone using Twitter (or other social on (and notifying) the imminent arrival of the media) data. Answering the question raised by shake, following a significant earthquake that Robert Geller and his colleagues as to whether has occurred far away. However, this isn’t true earthquake prediction is “inherently impossible prediction; rather, this is an early warning sys- or just fiendishly difficult” will probably require tem enabled by the difference existing between other decades of research in seismology.4

4 Published by the IEEE Computer Society 1089-7801/16/$33.00 © 2016 IEEE IEEE INTERNET COMPUTING 38 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE From the Editors: Editor’s Select Predictability or Early Warning: Using Social Media in Modern Emergency Response

What social media data actu- of qualitative and quantitative dam- tion of the extent to which models Predictability or Early Warning: ally can contribute to is assisting age assessment,7,8 and to important developed for a type of disaster (such emergency responders in assessing advances in the fields of situational as an earthquake) can be carried and mitigating natural and manmade awareness and crisis mapping.9 over to other disasters (for example, Using Social Media in Modern disasters. The most promising applica- Among the most notable findings of floods or wildfires), and the design tion for crowdsourced and fast-paced such works are the evidence of mod- of hybrid solutions that combine the social media data is in the time span erate-to-strong correlations between strengths of social media with those Emergency Response between the disaster’s occurrence tweet-derived predictors and earth- of traditional emergency-response and the impact/damage assessments quake intensity. Other interesting technologies. performed by means of tradi- results are about the possibility of Marco Avvenuti • University of Pisa, Italy tional — and typically less respon- producing impromptu crisis maps sive — technologies. Indeed, this is solely from social media, in a real- he achievement of reliable Stefano Cresci and Andrea Marchetti • National Research Council (CNR), Italy supported by the growing interest time fashion, thus allowing unpre- T results in this critical field of civil protection agencies, such as dictable disasters to be tracked as involves designing a system based Carlo Meletti • National Institute of Geophysics and Volcanology (INGV), Italy the US Federal Emergency Manage- they unfold. However, quantitative on past disasters and then thor- ment Agency (FEMA), in social media and verifiable results in these fields oughly testing it against future ones, Maurizio Tesconi • National Research Council (CNR), Italy emergency communications.6 are still few and far between, and which requires a great deal of effort Furthermore, state-of-the-art despite these compelling findings, and time. Meanwhile, current soci- approaches in social media-based many questions still remain unan- etal and environmental challenges, emergency response have recently swered. Among the most pressing such as climate change, are bound to led to the first results in the fields research challenges are the evalua- cause more frequent and more severe ariluz Congosto and her colleagues1 the propagation speed of communication media recently mentioned the possibility of and that of seismic waves. Instead, adhering to Standards Senior Business Development Manager: Sandy Brown predicting natural disasters using social a more genuine interpretation, the hypothesis of Yong Cui • [email protected] Advertising Coordinator: Deborah Sims, M View from the Cloud [email protected] media data. They support their claims by add- predicting earthquakes using social media data George Pallis • [email protected] ing references to two works that employed can be ruled out straightaway as the relevant Additional Editorial Board Members Technical cosponsor: Twitter data for the detection of earth- data are produced and shared as a result of the Editor in Chief Elisa Bertino • [email protected] quakes.2,3 Within this context, we feel that the earthquake itself. As such, it’s not available M. Brian Blake • [email protected] Fabian Bustamante • [email protected] references to these two other works2,3 might before the earthquake occurs. Associate Editors in Chief Fred Douglis* • [email protected] Now there’s Barry Leiba • [email protected] Stephen Farrell • [email protected] be overstated. For instance, in our work,2 we Another source of ambiguity could be a few Read all your IEEE magazinesElena Ferrari • [email protected] Anirban Mahanti • [email protected] IEEE Internet Computing 3 George Pallis • [email protected] Robert E. Filman* • [email protected] even more to discussed a system for the detection and dam- references in Sakaki and his colleagues’ work and journals your WAY onMichael N. Huhns • [email protected] IEEE Computer Society Publications Office age assessment of earthquakes in Italy. Simi- mentioning research efforts aimed at inves- Columnists Arun Iyengar • [email protected] 10662 Los Vaqueros Circle larly, in Takeshi Sakaki and his colleagues’ tigating the possibility for intermediate- and Backspace Anne-Marie Kermarrec • [email protected] Alamitos, about CA 90720 USA your 3 Vinton G. Cerf • [email protected] Peter Mika • [email protected] work, they propose a system for detecting short-term earthquake prediction. However, in Digital Citizen Dejan Milojicic • [email protected] Editorial. Unless otherwise stated, bylined articles, membership...as well as product and service descriptions, reflect earthquakes as well as detecting and esti- this work,3 there’s no specific reference to the Kieron O’Hara • [email protected] Michael Rabinovich* • [email protected] From the Editors Amit Sheth • [email protected] the author’s or firm’s opinion. Inclusion in IEEE mating the trajectory of tornadoes in Japan. results of such studies, and ultimately, neither M. Brian Blake • [email protected] Weisong Shi • [email protected] Internet Computing does not necessarily constitute In both works, no claims are made about the the effectiveness nor the applicability of such Peering Maarten van Steen • [email protected] endorsement by IEEE or the IEEE Computer Society. Charles J. Petrie* • [email protected] * EIC emeritus All submissions are subject to editing for style, clarity, systems’ capabilities to predict earthquake predictive techniques are supported. and length. Practical Security CS Magazine Operations Committee occurrences, but more about their ability to Over the course of more than 100 years, Hilarie Orman • [email protected] Submissions. For detailed instructions, see the author Forest Shull (chair), M. Brian Blake, Maria Ebling, guidelines (www.computer.org/internet/author.htm) detect and report. We believe this opens an the potential promise of earthquake predic- DepartmentIntroducing Editors myCS, the digital magazineLieven Eeckhout, Miguel Encarnacao, or log onto IEEE Internet Computing’s author center even more interesting discussion. tion has attracted researchers in seismology as Beyondportal Wires from IEEE Computer Society. Nathan Ensmenger, Sumi Helal, San Murugesan, at ScholarOne (https://mc.manuscriptcentral.com/ Yong Rui, Ahmad-Reza Sadeghi, Diomidis Spinellis, A potential source of ambiguity, and a pos- well as amateurs and impostors. Many theories Yih-FarnGo Robinbeyond Chen • static,[email protected] hard-to-read PDFs cs-ieee). Articles are peer reviewed for technical Big Data Bites George K. Thiruvathukal, Mazin Yousif, Daniel Zeng merit. 4 with an easily accessible, customizable, sible reason for this claim by Congosto and have been tested and systematically rejected. Jimmy Lin • [email protected] Letters to the Editors. Email lead editor Brian Brannon, and adaptive experience. CS Publications Board her colleagues could lie in the meaning given Recent works in seismology have demonstrated Developing World David S. Ebert (VP for Publications), Alfredo Benso, [email protected] Kamal Bhattacharya • [email protected] On the Web. www.computer.org/internet/. to “predict,” when the definition “say or esti- once more that current knowledge in this field Irena Bojanova, Greg Byrd, Min Chen, Robert Dupuis, InternetThere’s Governance No Additional Cost!Niklas Elmqvist, Davide Falessi, William Ribarsky, Subscribe. Visit www.computer.org/subscribe/. mate that a specific thing will happen in the is unable to predict earthquake occurrences in Virgilio Almeida • [email protected] Forrest Shull, Melanie Tory Subscription Change of Address. Send requests to 5 Internet of Things, People, and Processes [email protected]. future” is interpreted as the action of figuring any way, let alone using Twitter (or other social Schahram Dustdar • [email protected] Staff Missing or Damaged Copies. Contact help@ on (and notifying) the imminent arrival of the media) data. Answering the question raised by Linked Data Editorial Management: Tammi Titsworth computer.org. shake, following a significant earthquake that Robert Geller and his colleagues as to whether Carole Goble • [email protected] Manager, Editorial Services Content Development: Brian To Order Article Reprints. Email internet@computer. Natural Web Interfaces Brannon, [email protected] org or fax +1 714 821 4010. has occurred far away. However, this isn’t true earthquake prediction is “inherently impossible Munindar P. Singh* • [email protected]►LEARN PublicationsMORE Coordinator: AT: [email protected] mycs.computer.orgIEEE prohibits discrimination, harassment, and prediction; rather, this is an early warning sys- or just fiendishly difficult” will probably require Spotlight Director, Products & Services: Evan Butterfield bullying. For more information, visit www.ieee.org/ Gustavo Rossi • [email protected] Senior Manager, Editorial Services: Robin Baldwin web/aboutus/whatis/policies/p9-26.html. tem enabled by the difference existing between other decades of research in seismology.4

IEEE myCS half Page Space Ad 2016_4-26-16.indd 1 4/28/16 3:03 PM 4 Published by the IEEE Computer Society 1089-7801/16/$33.00 © 2016 IEEE IEEE INTERNET COMPUTING NOvEMbER/dECEMbER 2016 5 www.computer.org/computingedge 39 From the Editors

Introducing “Editor’s Select” Stefano Cresci is a PhD student in the Depart- ment of Information Engineering at the Perhaps one of the greatest aspects of being the Editor in Chief of a magazine University of Pisa and a Research Fel- that’s central to my own research area is the ability to read early work on the low at the Institute of Informatics and most innovative topics. As EIC, I also get the opportunity to experience firsthand Telematics of the National Research the interplay between our researchers as they develop new approaches to real- Council (CNR), Italy. His research inter- world problems. ests include social media mining and Often, contributions from one article can provide significant insights into other knowledge discovery. Cresci has an MSc works, even when the approaches are different. Other times, authors get to chal- in computer engineering and an MSc in lenge the work of others in such a way that a greater understanding for both proj- Big Data analytics and social mining ects can be attained. from the University of Pisa. He is a stu- In this issue of IEEE Internet Computing, I’m using the “From the Editors” col- dent member of IEEE and member of the umn as a special “Editor’s Select” feature — a platform similar to letters to the IEEE Computer Society. Contact him at editor, but with adequate space for discussion. This “Editor’s Select” highlights [email protected]. insights from authors who read one of our earlier published articles and want to express a deeper insight. I hope that you enjoy their comments and appreciate Andrea Marchetti is a technologist at the the scholarly spirit of what they observed from reading the works of IEEE Internet Institute of Informatics and Telematics Computing. of the National Research Council (CNR), I would also like to take a moment to thank the guest editors for this issue, Kaus- Italy. His research interests include tubh Joshi and Theophilus benson. This special issue focuses on network function social media analysis, e-government virtualization (NFv), a burgeoning area of interest to many. services quality, and the Semantic — M. Brian Blake, Editor in Chief Web. Marchetti has a degree in computer science from the University of Pisa. Con- tact him at [email protected]. disasters. Thus, globally we must Annals of Geophysics, 2011, vol. 54, no. Carlo Meletti is a senior technological scien- increase scientific and technological 4; doi:10.4401/ag-5350. tist at the National Institute of Geophys- efforts in this direction, because it’s 6. Federal Emergency Management Agency ics and Volcanology (INGV), Italy. His destined to be of crucial importance (FEMA), 2014–2018 FEMA Strategic Plan, research interests include seismic hazard, for years to come. tech. report, FEMA, 2014; www.fema.gov/ seismic risk, and loss-reduction policy. media-library/assets/documents/96981. Meletti has an MSc in geological sci- References 7. Y. Kryvasheyeu et al., “Rapid Assessment of ences from the University of Pisa. Con- 1. M. Congosto, D. Fuentes-Lorenzo, and L. Disaster Damage Using Social Media Activ- tact him at [email protected]. Sánchez, “Microbloggers as Sensors for ity,” Science Advances, vol. 2, no. 3, 2016, p. Public Transport Breakdowns,” IEEE Internet e1500779; doi:10.1126/sciadv.1500779. Maurizio Tesconi is a computer science Computing, 2015, vol. 19, no. 6, pp. 18–25. 8. M. Avvenuti et al., “Nowcasting of Earth- researcher at IIT-CNR. His research inter- 2. M. Avvenuti et al., “EARS (Earthquake quake Consequences Using Big Social ests include social Web mining, social Alert and Report System): A Real Time Data,” IEEE Internet Computing, 2016, network analysis, and visual analytics Decision Support System for Earthquake preprint; doi:10.1109/MIC.2016.47. within the context of open source intel- Crisis Management,” Proc. 20th ACM 9. M. Avvenuti et al., “Impromptu Crisis Map- ligence. Tesconi has a PhD in information SIGKDD Int’l Conf. Knowledge Discovery ping to Prioritize Emergency Response,” engineering from the University of Pisa. and Data Mining, 2014, pp. 1749–1758. Computer, vol. 49, no. 5, 2016, pp. 28–37. He is a member of the permanent team 3. T. Sakaki, M. Okazaki, and Y. Matsuo, of the European Laboratory on Big Data “Tweet Analysis for Real-Time Event Detec- Marco Avvenuti is a professor of operating Analytics and Social Mining. Contact tion and Earthquake Reporting System systems in the Department of Infor- him at [email protected]. Development,” IEEE Trans. Knowledge and mation Engineering at the University Data Eng., vol. 25, no. 4, 2013, pp. 919–931. of Pisa. His research interests include 4. R.J. Geller et al., “Earthquakes Cannot Be human-centric sensing and social media Predicted,” Science, 2007, vol. 275, no. 5306, analysis. Avvenuti has a PhD in infor- Read your subscriptions p. 1616; doi:10.1126/science.275.5306.1616. mation engineering from the Univer- This article originally appeared in through the myCS publi- 5. T. Jordan et al., “Operational Earth- sity of Pisa. He’s a member of the IEEE IEEE Internet Computing, vol. 20, cations portal at http:// quake Forecasting: State of Knowledge Computer Society. Contact him at marco. no. 6, 2016. mycs.computer.org. and Guidelines for Implementation,” [email protected].

6 www.computer.org/internet/ IEEE INTERNET COMPUTING 40 Computing Edge June 2017 PREPARE TO CONNECT

The IEEE Computer Society is launching INTERFACE, a new communication tool to help members engage, collaborate and stay current on CS activities. Use INTERFACE to learn about member accomplishments and find out how your peers are changing the world with technology.

We’re putting our professional section and student branch chapters in the spotlight, sharing their recent activities and giving leaders a window into how chapters around the globe meet member expectations. Plus, INTERFACE will keep you informed on CS activities so you never miss a meeting, career development opportunity or important industry update.

Launching this spring. Watch your email for its debut. IT and FuTure employmenT edITor: George Strawn, [email protected]

IT and Future Unemployment: Part 2

George Strawn, US National Academies of Sciences, Engineering, and Medicine

n a previous IT and Future A Man and a Dog? that US manufacturing is actually Employment column,1 I dis- A recent New York Times article doing fairly well, and that it’s just cussed the specter of massive, highlighted a textile mill that had manufacturing employment that automation-caused unem- been closed, idling 2,000 workers has begun a downward automa- ployment.I Given that this specter and sending the work to Asia.2 tion spiral. could be decades away, it gets little Several years ago, that same mill Business today certainly has dif- attention in the press and, until re- reopened, providing work for only ferent constraints and opportuni- cently, has gotten almost no atten- 200 workers and turning out more ties from when I began my career tion in politics. On the other hand, textiles annually than the previ- at a major US corporation more current unemployment, apparently ous 2,000 employees had been than 50 years ago. At that point in caused by the recent recession and able to do. This trend gives rise to time, I was told that we had three offshoring of jobs, is a regular topic the unhappy joke that the textile pillars of business: excellent prod- for news and politicians. Politicians mill of the future will have two ucts and services for our custom- of all stripes regularly say, “My op- employees: a man and a dog. The ers, respect for our employees, and ponent’s policies would be job kill- man’s job is to feed the dog, and a fair return for our stockholders. ers; my policies will create jobs.” I’ll the dog’s job is to keep the man From reading newspaper accounts begin this column by suggesting away from the machinery. today, I deduce that many busi- several possible relationships be- Offshoring and automation nesses now have only one pillar: tween long-term, automation-caused can be two acts of the same play. maximum return for stockhold- unemployment and short-term, Act one is sending jobs overseas ers. Rhetoric aside, I suspect that offshoring-caused unemploy- to take advantage of cheap labor, capitalism has always placed prof- ment. Next, I’ll give my opinion and act two is bringing a few jobs itability above other aims, and about the connection between un- back to a more automated envi- whether you think that’s good or employment and the presidential ronment and claiming job cre- bad probably depends on your po- election of 2016. Finally, I’ll high- ation. In addition to very visible litical views. The relevant point for light two recent interviews, the offshoring, not-so-visible (and this article is that if business can first given by a prominent politi- not-yet-massive) automation is save money by adopting more au- cian on the left and the second already under way. Between 1975 tomation, it will. given by a prominent thinker on and 2010, US manufacturing out- the right. Both interviews express put more than doubled, while em- The Angst of 2016 concern about automation-caused ployment in the sector decreased This column is being written just unemployment and discuss a rem- by 31 percent.3 Statistics such after the presidential election of edy for it. as these would seem to indicate 2016. My (amateur) analysis of the

42 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE 70 IT Pro January/February 2017 Published by the IEEE Computer Society 1520-9202/17/$33.00 © 2017 IEEE IT and FuTure employmenT edITor: George Strawn, [email protected]

campaign is as follows. The status reason textile automation began or auditors, that might disappear. quo candidates of the parties were in England and not in India or ... I don’t know what you think challenged by “rogues” from both China—the traditional centers about universal basic income [ital- directions. From the left came of textiles—was because English ics added], but as we start to see Bernie Sanders, who lost the pri- wages were so high that pursuing people getting displaced, there’s mary, and from the right came automation was profitable. If one also this idea that we can look at Donald Trump, who won. As the tenant of Trumpism is making other models.” general election ground to its end, and keeping jobs, will any politi- The president responded, some commentators predicted cian admit to fears of long-term IT and Future that Trump might go away, but unemployment? Now, whether a universal income Trumpism would not. We now is the right model—is it gonna be know that a more accurate prog- A Prominent Politician accepted by a broad base of peo- Unemployment: nostication would have been that Expresses His Fears ple?—that’s a debate that we’ll neither Trump nor Trumpism will No less a politician than Barack be having over the next 10 or 20 go away. Next, I’ll say what I think Obama (who doesn’t have to win years. You’re also right that the jobs Part 2 Trumpism means (in part). any more elections!) recently gave that are going to be displaced by AI First, I believe that Trumpism an interview to Wired magazine, are not just low-skill service jobs; and Sanders-ism are expressions along with Joi Ito, in which he ex- they might be high-skill jobs but George Strawn, US National Academies of Sciences, Engineering, and Medicine of the same angst people feel about pressed his concerns about future ones that are repeatable and that the future. Part of that angst is unemployment caused by artifi- computers can do. What is indis- people’s fear of losing their jobs cial intelligence and robotics—the putable, though, is that as AI gets n a previous IT and Future A Man and a Dog? that US manufacturing is actually to offshoring or automation. The leading edge of today’s IT. further incorporated, and the soci- Employment column,1 I dis- A recent New York Times article doing fairly well, and that it’s just two differ in that Trumpism as- cussed the specter of massive, highlighted a textile mill that had manufacturing employment that serts that government is part of the automation-caused unem- been closed, idling 2,000 workers has begun a downward automa- problem, whereas Sanders-ism as- If the choice faced by US businesses is between ployment.I Given that this specter and sending the work to Asia.2 tion spiral. serts that government is part of the could be decades away, it gets little Several years ago, that same mill Business today certainly has dif- solution. Focusing on Trumpism, high-cost US labor and automation, more automation attention in the press and, until re- reopened, providing work for only ferent constraints and opportuni- it would appear to be a populism will happen sooner. cently, has gotten almost no atten- 200 workers and turning out more ties from when I began my career of the right, focusing on the fact tion in politics. On the other hand, textiles annually than the previ- at a major US corporation more that the public is not being served current unemployment, apparently ous 2,000 employees had been than 50 years ago. At that point in by the status quo and vowing to caused by the recent recession and able to do. This trend gives rise to time, I was told that we had three implement right-wing solutions to Obama stated that We’ve been ety potentially gets wealthier, the offshoring of jobs, is a regular topic the unhappy joke that the textile pillars of business: excellent prod- “make America great again.” seeing specialized AI in every aspect of link between production and dis- for news and politicians. Politicians mill of the future will have two ucts and services for our custom- Some of the solutions advo- our lives, from medicine and transpor- tribution, how much you work and of all stripes regularly say, “My op- employees: a man and a dog. The ers, respect for our employees, and cated by Trump during the cam- tation to how electricity is distributed, how much you make, gets further ponent’s policies would be job kill- man’s job is to feed the dog, and a fair return for our stockholders. paign were reduced immigration and it promises to create a vastly more and further attenuated. ers; my policies will create jobs.” I’ll the dog’s job is to keep the man From reading newspaper accounts (to control both terrorism and job productive and efficient economy. If begin this column by suggesting away from the machinery. today, I deduce that many busi- loss to immigrants), renegotiated properly harnessed, it can generate enor- It remains to be seen if more several possible relationships be- Offshoring and automation nesses now have only one pillar: trade deals (to favor US exports mous prosperity and opportunity. But politicians will join the discussion tween long-term, automation-caused can be two acts of the same play. maximum return for stockhold- and hence US jobs), and penalties it also has some downsides that we’re about automation and unemploy- unemployment and short-term, Act one is sending jobs overseas ers. Rhetoric aside, I suspect that for US corporations who ship jobs gonna have to figure out in terms of not ment. It also remains to be seen offshoring-caused unemploy- to take advantage of cheap labor, capitalism has always placed prof- overseas. All of these solutions eliminating jobs. ... Low-wage, low-skill if and when a universal basic in- ment. Next, I’ll give my opinion and act two is bringing a few jobs itability above other aims, and aim to make or keep more jobs for individuals become more and more re- come (UBI) policy gets serious about the connection between un- back to a more automated envi- whether you think that’s good or US citizens, supporting the view dundant, and their jobs may not be re- consideration in the US. employment and the presidential ronment and claiming job cre- bad probably depends on your po- that Trumpism focuses on our placed, but wages are suppressed. And if election of 2016. Finally, I’ll high- ation. In addition to very visible litical views. The relevant point for fear of losing our jobs. we are going to successfully manage this A Universal Basic Income? light two recent interviews, the offshoring, not-so-visible (and this article is that if business can However, these solutions might transition, we are going to have to have I touched on the UBI idea in my first given by a prominent politi- not-yet-massive) automation is save money by adopting more au- actually hasten the arrival of more a societal conversation about how we previous column on unemploy- cian on the left and the second already under way. Between 1975 tomation, it will. automation and more job loss. manage this. ... The social compact has ment,1 and I’ll expand on it here. given by a prominent thinker on and 2010, US manufacturing out- If the choice faced by US busi- to accommodate these new technologies, Interestingly, this policy was the right. Both interviews express put more than doubled, while em- The Angst of 2016 nesses is between high-cost US and our economic models have to accom- proposed many years ago by a concern about automation-caused ployment in the sector decreased This column is being written just labor and automation, more au- modate them.4 right-leaning economist, Milton unemployment and discuss a rem- by 31 percent.3 Statistics such after the presidential election of tomation will happen sooner. For Ito added, “There are [also] very Friedman, and it originally had edy for it. as these would seem to indicate 2016. My (amateur) analysis of the example, it has been said that the high-level jobs, things like lawyers nothing to do with unemployment.

More recently, a variation of trum see it as a solution to massive www.wired.com/2016/10/president this policy was detailed in a now unemployment makes me think -obama-mit-joi-ito-interview/. 10-year-old book written by there might be room for compro- 5. C. Murray, In Our Hands: A Plan to another right-leaning author, mise—for example, some social Replace the Welfare State, revised, up- Charles Murray.5 Both scholars programs might be replaced and dated ed., Aei Press, 2016. proposed the UBI to solve the some retained. The UBI certainly 6. A. Holt, “Universal Basic Income and “problem” of the welfare state— isn’t politically feasible in the US the New Corporatist Democrats,” the social services provided to at this time, nor is it necessary yet. New America, 11 Aug. 2016; www. citizens by the government (in the As Obama said, “That’s a debate newamerica.org/weekly/edition-131/ US, this means Medicare, Medic- that we’ll be having over the next universal-basic-income-and-new- aid, Social Security, food stamps, 10 or 20 years.” corporatist-democrats/. and so on). However, in a recent talk discussing the book upon References George Strawn is the director of the its republishing in 2016 (www. 1. G. Strawn, “IT and Future Unem- Board for Research Data and Informa- c-span.org/video/?415461-1/charles ployment,” IT Professional, vol. 18, tion at the US National Academies of -murray-discusses-hands), Murray no. 1, 2016, pp. 62–64. Sciences, Engineering, and Medicine. stated that he now considers 2. S. Clifford, “US Textile Plants Re- He is the former director of the National automation-caused unemploy- turn, With Floors Largely Empty Coordination Office for the Networking ment to be a serious threat and of People,” New York Times, 19 Sept. and Information Technology Research that the UBI could also be a so- 2016; http://nyti.ms/2iYLCWz. and Development Program (NITRD). lution to that potential problem. 3. V. de Rugy, “US Manufacturing: Out- Contact him at [email protected]. Also relevant to this topic is an ar- put vs. Jobs Since 1975,” Mercatus Cen- ticle published by the nonpartisan ter, George Mason Univ., 24 Jan. 2011; New American Foundation.6 www.mercatus.org/publication/us In Murray’s plan, every Ameri- -manufacturing-output-vs-jobs-1975. Read your subscriptions can 21 years old and older would 4. S. Dadich, “Barack Obama, Neural This article originallythrough the appearedmyCS publi in- receive $13,000 per year (in a Nets, Self-Driving Cars, and the Fu- IT Professional,cations vol. portal 19, no.at http:// 1, 2017. mycs.computer.org. monthly payment into a bank ac- ture of the World,” Wired, Nov. 2016; count). Of this amount, $3,000 would be directed to a “cata- strophic” heathcare program, and

$10,000 could be spent however July–September 2016 IEEE MultiMedia each recipient might choose. Of IT Professional (ISSN 1520-9202) is published bimonthly by the IEEE course, $10,000 per year is not a Computer Society. IEEE Headquarters, Threehttp://www.computer.org Park Avenue, 17th Floor,

livable wage, but a married couple New York, NY 10016-5997; IEEE Computer Society Publications Office, or a group of roommates, espe- 10662 Los Vaqueros Circle, PO Box 3014, Los Alamitos, 2016 july–september CA 90720-1314; cially if some are working even voice +714 821 8380; fax +714 821 4010; IEEE Computer Society Head-

quarters, 1828 L St. N❚ W, Suite 1202, Washington, DC 20036. Annual

low-paying jobs, could approach Quality Modeling subscription: $49 in addition to any IEEE Computer Society dues. Non-

middle class earnings. A person earning $60,000 or more would member rates are available on request. Back issues: $20 for members, only receive half of the normal $143 for nonmembers. amount. By Murray’s calculation, Postmaster: Send undelivered copies and address changes to IT Pro- we were already spending enough fessional, MembershipVolume 23 Number 3 Processing Dept., IEEE Service Center, 445 Hoes on government social programs Lane, Piscataway, NJ 08854-4141. Periodicals Postage Paid at New by 2009 to pay for this UBI, and York, NY, and at additionalmult-22-03-c1 Cover-1 mailing offices. CanadianJuly 12, 2016 4:40 PM GST #125634188. we are already spending more. CanadaIEEE Po MultiMediast Publications serves M theail Acommunitygreement ofN uscholars,mber 40013885. developers, R eturn practitioners, and students who are interested in multiple media undeliverable Canadian addresses to PO Box 122, Niagara Falls, ON L2E types and work in fields such as image and video processing, audio 6S8, Canada. Printed in the USA. analysis, text retrieval, and data fusion. hose on the left may see the Editorial: Unless otherwise stated, bylined articles, as well as product UBI as an additional so- and service descriptions,Read reflect the It author’s Today! or firm’s opinion. Inclusion in T cial program, added to and IT Professional does not necessarily constitute endorse- not replacing existing social pro- ment by the Iwww.computer.org/multimediaEEE or the Computer Society. All submis- grams. However, the fact that sions are subject to editing for style, clarity, and space. people on both sides of the spec-

44 Computing Edge June 2017 72 IT Pro January/February 2017 Editor: Rafael Prikladnicki Pontifica Universidade Catolica VOICE OF EVIDENCE do Rio Grande do Sul [email protected]

Preparing Tomorrow’s Software Engineers for Work in a Global Environment

Sarah Beecham, Tony Clear, John Barr, Mats Daniels, Michael Oudshoorn, and John Noll

GLOBAL SOFTWARE ENGINEERING complexities of working in multisite en- (GSE; building software in a global environments. They must enable students vironment with distributed, multisite de- to deliver high-quality, robust, secure velopment teams) is fast becoming com- code when working with team members mon. Thus, today’s software engineering they might never meet in person, who students will likely become tomorrow’s might be asleep when they need an an- global software engineers. Recognizing swer to an urgent question, who might this nearly 10 years ago, John White, be offended by an off-the-cuff comment then the Association for Computing Ma- because of cultural differences, or who chinery’s chief executive of cer, intro- might not understand the terminology duced the report of the ACM Job Mi- used because they don’t share the same gration Task Force on the Globalization rst language. In short, GSE exaggerates and Offshoring of Software as follows: existing colocated teamwork challenges and introduces some new ones. The eld of computing and informa- Teaching GSE by engaging in dis- tion technology has experienced a tributed, multiuniversity development dramatic shift in the past ve years to projects entails challenges, including a truly global industry. The forces that considerable instructor overhead, but have driven and shaped this change are our study revealed that many universi- still at play and will continue. … The ties are doing it all the same.2 Schools in educational systems that underpin our nearly 40 countries are collaborating to profession will need to change.1 prepare their students for multisite distributed development by, for example, The traditional computer science giving them rsthand experience and curriculum—in which students typi- teaching them the importance of soft cally work with friends in self-selected skills and teamwork. This article originally groups—prepares students for tech- A group of researchers and computer appeared in nical challenges associated with soft- science educators at the 2015 ACM IEEE Software, vol. 34, ware development but not for working Conference on Innovation and Tech- no. 1, 2017. in globally distributed teams. Educa- nology in Computer Science Education tors now must prepare students for the looked at various ways to support GSE

those who have tried it, both success- VALENTINE’S TAXONOMY fully and unsuccessfully. The sum- mary of our ndings, fully reported We applied a taxonomy for computer-science education publications1 to judge elsewhere,2 shows how GSE educa- the quality of the 82 studies in our literature review. This taxonomy has six tors plan and conduct their courses. categories: (See our Web extra at https://extras .computer.org/extra/mso2017 • Experimental papers present rigorously researched studies. 010009s1.pdf or our associated sys- • Marco Polo papers describe the authors’ experiences with a new course, tematic literature-review protocol.4) language, or curriculum. Generally, the 82 papers in our study • Philosophy papers try to generate debate on an issue. were based on educators’ direct ex- • Tools papers describe new tools or techniques. perience, as shown by our literature • Nifty papers present new and interesting ways to teach something. categorization, which was based on • John Henry papers relate the authors’ experiences with idiosyncratic or David Valentine’s system (see the extremely difficult courses. “Valentine’s Taxonomy” sidebar).

Table A shows the categorization of the studies we examined. The fact that Recommendations half were in the Experimental category demonstrates our analysis’s quality. Also, GSE education’s key challenges and about 80 percent of the studies re ected direct teaching experiences. our recommendations for overcom- ing them fall into seven major categories (see Figure 1): global distance, Table A. A categorization of 82 global- teamwork, soft issues, stakeholder software-engineering education papers, or role, infrastructure, curriculum using David Valentine’s taxonomy.1 and pedagogy, and development process. For each, we identi ed up to Category No. of courses Percentage of the total

TABLE A TABLE three main challenges and the most Experimental 42 51.2 frequently mentioned methods of ad- Marco Polo 24 29.3 dressing them (see Table 1 on p. 12). Some of the challenges arise from Philosophy 6 7.3 the nature of GSE; others are more Tools 9 11.0 education related. Conducting distributed, multi- Nifty 1 1.2 university courses isn’t for the faint- John Henry 0 0 hearted. Students might feel some of the pain, although studies indicated they generally enjoyed and pro ted Reference from the experience. However, the 1. D. Valentine, “CS Educational Research: A Meta-analysis of SIGCSE Technical Symposium overhead the programs create falls Proceedings,” Proc. 35th SIGCSE Technical Symp. Computer Science Education (SIGCSE squarely on the instructors. 04), 2004, pp. 255–259. Our analysis revealed that global distance was another key area to consider when offering GSE courses but that other types of distance were education. Based on the idea that Reviewing the Literature also important (see Figure 2). universities can simulate the contem- Conducting distributed courses in- porary workplace by partnering with volving multiple universities is com- Industry: The Key Stakeholder other schools on distributed class plex and time-consuming, but we Our report found that some univer- projects, we came up with a set of can bene t by reviewing the literature sities are actively pursuing GSE proj- challenges and recommendations.3 and consolidating the experience of ects in collaboration with compa-

4610 IEEE SOFTWAREComputing | Edge WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE June 2017 VOICE OF EVIDENCE VOICE OF EVIDENCE

those who have tried it, both success- nies, which would benefit in the long References Soft issues 10% VALENTINE’S TAXONOMY fully and unsuccessfully. The sum- run from university students who are 1. W. Aspray, F. Mayadas, and M. Vardi, Development process 4% mary of our ndings, fully reported better educated in real-world soft- Globalization and Offshoring of We applied a taxonomy for computer-science education publications1 to judge elsewhere,2 shows how GSE educa- ware development practices. How- Software: A Report of the ACM Job the quality of the 82 studies in our literature review. This taxonomy has six tors plan and conduct their courses. ever, GSE projects often suffer from Migration Task Force, ACM, 2006. categories: (See our Web extra at https://extras a dearth of willing industry collabo- 2. S. Beecham et al, “How Best to Teach Global distance .computer.org/extra/mso2017 rators, compromising many attempts Global Software Engineering? Educa- Infrastructure 32% 11% • Experimental papers present rigorously researched studies. 010009s1.pdf or our associated sys- to emulate real-world settings. tors are Divided,” IEEE Software, • Marco Polo papers describe the authors’ experiences with a new course, tematic literature-review protocol.4) This situation has several causes. vol. 34, no. 1, 2017, pp. 16. Stakeholder/role 13% language, or curriculum. Generally, the 82 papers in our study For example, companies might not 3. T. Clear et al., “Challenges and • papers try to generate debate on an issue. Recommendations for the Design Teamwork Philosophy were based on educators’ direct ex- want to participate because they fear 17% • Tools papers describe new tools or techniques. perience, as shown by our literature that students—who typically take and Conduct of Global Software • Nifty papers present new and interesting ways to teach something. categorization, which was based on multiple courses at the same time— Engineering Courses: A Systematic • papers relate the authors’ experiences with idiosyncratic or Review,” Proc. 20th Conf. Innova- John Henry David Valentine’s system (see the might not give the GSE project a high Curriculum/pedagogy 13% extremely difficult courses. “Valentine’s Taxonomy” sidebar). priority. Also, students generally don’t tion and Technology in Computer have the skills or awareness of impor- Science Education on Working FIGURE 1. Global software-engineering Table A shows the categorization of the studies we examined. The fact that Recommendations tant issues that professionals have. Group Reports (ITiCSE-WGR 15), (GSE) education challenges. This figure half were in the Experimental category demonstrates our analysis’s quality. Also, GSE education’s key challenges and Nonetheless, companies should 2015, pp. 1–39. shows the distribution of the seven about 80 percent of the studies re ected direct teaching experiences. our recommendations for overcom- also look at the advantages of par- 4. S. Beecham et al., Challenges and categories of challenges the authors ing them fall into seven major cate- ticipating in universities’ GSE proj- Recommendations for the Design found in the 775 observations they made gories (see Figure 1): global distance, ects. After all, these students could and Conduct of Global Software from the 82 GSE-education papers that Table A. A categorization of 82 global- teamwork, soft issues, stakeholder be their future employees. GSE edu- Engineering Courses: A System- they studied. software-engineering education papers, or role, infrastructure, curriculum cation is a difficult endeavor that atic Review Protocol, tech. report using David Valentine’s taxonomy.1 and pedagogy, and development pro- schools undertake with limited re- Lero_TR_2015_01, Lero—The Irish cess. For each, we identi ed up to sources. Companies could help by Software Research Centre, 2015. Category No. of courses Percentage of the total

TABLE A TABLE three main challenges and the most supplying resources such as tools, Linguistic Institutional Experimental 42 51.2 frequently mentioned methods of ad- expertise, or project involvement. SARAH BEECHAM is a Senior Research Fellow at Lero–The Irish Software Research Centre. Marco Polo 24 29.3 dressing them (see Table 1 on p. 12). Some of the challenges arise from Contact her at [email protected]. Cultural Philosophy 6 7.3 the nature of GSE; others are more arly feedback on our re- Tools 9 11.0 education related. search from several instruc- TONY CLEAR is an associate professor in the Conducting distributed, multi- tors indicates that it’s a valu- Auckland University of Technology’s School of Nifty 1 1.2 E GSE university courses isn’t for the faint- able reference source for educators Engineering, Computer and Mathematical Sci- John Henry 0 0 hearted. Students might feel some of seeking to offer GSE courses in part- ences. Contact him at [email protected]. the pain, although studies indicated nership with other universities and they generally enjoyed and pro ted companies. We’ve provided clear JOHN BARR is an associate professor in Ithaca Temporal Geographic Reference from the experience. However, the recommendations to overcome each College’s Department of Computer Science. 1. D. Valentine, “CS Educational Research: A Meta-analysis of SIGCSE Technical Symposium overhead the programs create falls challenge raised in the studies we ex- Contact him at [email protected]. Proceedings,” Proc. 35th SIGCSE Technical Symp. Computer Science Education (SIGCSE squarely on the instructors. amined. The key takeaways for edu- FIGURE 2. Distances in global- 04), 2004, pp. 255–259. Our analysis revealed that global cators are to prepare early, partner MATS DANIELS is a senior lecturer and software-engineering (GSE) education. distance was another key area to with people you know you can work director of undergraduate studies in Uppsala Uni- Various types of distances—physical, consider when offering GSE courses with, know your students’ experi- versity’s Department of Information Technology. time zone, cultural, language, and but that other types of distance were ence and expertise levels, and plan Contact him at [email protected]. institutional—between universities education. Based on the idea that Reviewing the Literature also important (see Figure 2). the course accordingly. participating in multisite GSE courses universities can simulate the contem- Conducting distributed courses in- MICHAEL OUDSHOORN is an associate pro- make teaching the topic difficult. porary workplace by partnering with volving multiple universities is com- Industry: The Key Stakeholder Acknowledgments fessor in Northwest Missouri State University’s other schools on distributed class plex and time-consuming, but we Our report found that some univer- Science Foundation Ireland grant 10/CE/ School of Computer Science and Informa- projects, we came up with a set of can bene t by reviewing the literature sities are actively pursuing GSE proj- I1855 to Lero—The Irish Software Research tion Systems. Contact him at oudshoorn@ JOHN NOLL is a senior lecturer at the University challenges and recommendations.3 and consolidating the experience of ects in collaboration with compa- Centre partly supported this research. nwmissouri.edu. of East London. Contact him at [email protected].

10 IEEE SOFTWARE | WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE www.computer.org/computingedge JANUARY/FEBRUARY 2017 | IEEE SOFTWARE 11 47 VOICE OF EVIDENCE

Global software engineering (GSE) education challenges and recommended solutions.

Area of concern Challenges Recommendations Global distance Limited workday overlap between distant • Schedule regular meetings in advance.

TABLE 1 TABLE sites inhibits communication and causes • Make teams communicate more often than they normally would. delays. • Use synchronous and asynchronous media. Participating institutions have different • Synchronize schedules and evaluation schemes, while remaining flexible. term schedules, expectations, and • Ensure roles are well defined at each institution. regulations. • Find a supportive, reliable, and patient collaborating partner. Students don’t know how to work • Include cultural-awareness topics in course content. with people from different cultural • Provide opportunities for students to learn about their cultural differences. backgrounds. Teamwork Students require great effort and • Create exercises to teach students how to use communication tools. commitment to work in global teams. • Monitor students’ communication levels to ensure they remain high. Local teams often forget about global • Start communication early in the process to form relationships. teammates or misunderstand their • Create a shared vision among local and global teams. intentions. • Create bonding exercises. Students and instructors have trouble • Map students to specific skill positions. creating teams with balanced skills and • Appoint team leaders and in-country champions. managing them effectively. • Keep teams small. • Use self-organizing teams. Soft issues Students lack motivation. • Use contests and games to boost motivation. • Be an exceptionally enthusiastic instructor. • Require student participation to maintain engagement. There is a lack of trust among teams. • Implement social-bonding activities and communication protocols from the start. • Encourage students to be honest about their technical abilities. Stakeholder or role Students lack distributed-project- • Clearly define project organization and requirements. management experience. • Require a designated group project manager. Instructor workload increases. • Create a class structure that clearly delineates instructor responsibilities. • Use a coordinator or coach to help teams. • Establish support systems for students and instructors. • If changes are required, scale down the project to reduce complexity. Companies that could be potential project • Use simulators in place of real customers. customers don’t have experience with • Use programming contests instead of customers. university-based efforts. • Ensure that students build close relationships with customers. Infrastructure Communication and collaboration • Use groupware and similar collaboration tools. technology poses challenges. • Use wikis and other knowledge-management tools. • Use videoconferencing. Tools and environments are • Use a common environment or toolset across sites. heterogeneous. There’s no shared software configuration • Use Git or a similar SCCS or version-control system. and control system (SCCS). Curriculum and Tensions exist between the course • Support student learning via course design and mentoring. pedagogy design, learning outcomes, and student • Explicitly teach soft skills. inclinations. • Encourage reflective attitudes. Globally dispersed projects can be • Design achievable and authentic learning experiences. difficult. • Tailor the assessment process. Development System or code integration failures occur. • Manage merge conflicts when integrating software. process • Use design by contract to specify module interfaces; require mandatory code review of interfaces before implementation. Requirements, design, and testing • Start design and implementation early. failures occur. • Partition designs for independent development. • Document design decisions and rationales. Other process failures happen. • Require regular status meetings and/or reports. • Force students to use a common programming language.

4812 IEEE SOFTWAREComputing | Edge WWW.COMPUTER.ORG/SOFTWARE | @IEEESOFTWARE June 2017 AI AND SUSTAINABILITY Editor: Douglas H. Fisher, Vanderbilt University, douglas.h.ﬁ [email protected]

Incorporating Sustainability into Computing Education

Douglas H. Fisher, Zimei Bian, and Selina Chen, Vanderbilt University

omputing is ubiquitous and integral to many sustainability focus, such as courses in computer fi elds, so it is not surprising that higher edu- organization, databases, and artifi cial intelligence. C We dive deeper into component-level and course- cation in the environmental sciences, for example, level integration and then speculate on the future of will typically include computing (for example, for sustainability-themed computer science education, Earth-system simulations) in its coursework. What to include curriculum-level integration, which is of integrating sustainability, also pervasive in concerned with larger-than-course-curricular con- scope, into computer science education? structs, such as concentrations, minors, and majors. Until sustainability is in the bedrock of computing education, few students, professionals, or re- Component-Level Integration searchers are conscious of the possibilities of the There are many ways of infusing fi ne-grained sus- computing/sustainability nexus to transform soci- tainability elements into a computing course, to ety. This article surveys the integration of environ- include homework exercises, lecture and textbook mental and societal sustainability into computer examples, supplemental readings, exam questions, science curricula at colleges and universities, focus- and team projects. Here are a few examples: ing on intelligent systems educational materials. • A several-week project and sequence of exercises Integration Levels in a database course, taught by one of the au- Currently, integration of sustainability into com- thors and taken by the other two, has students puter science higher education happens at two levels reverse engineer the database used in a dorm en- of granularity—the course level and the course- ergy monitoring project.1 component level. • David Patterson and John Hennessey’s third edi- Course-level integration happens by introduction of Computer Organization and Design in- ing computer science courses that focus on topics cludes extended examples of computers in the real at the intersection of computing and sustainability. world, most of which have sustainability themes.2 For example, courses we survey here include Com- • Closer to the home of intelligent systems, home- puting, Energy, and the Environment and Intro- work exercises of varying complexity can have stu- duction to Computational Sustainability. At some dents apply a machine learning algorithm being institutions, it might also be benefi cial to consider studied (such as decision tree induction or support courses in other majors, such as Earth and environ- vector machines) to datasets in sustainability-rele- mental sciences, that include a considerable and in- vant areas such as smart electricity grids, transpor- tegral computing component. For example, a course tation, and fl ora and fauna preservation. on Agent-Based Modeling offered by the Depart- • Another intelligent systems assignment can have ment of Earth and Environmental Sciences might students design an optimization algorithm serve as a solid elective that adds a sustainability that constructs protected areas for wildlife con- component to a computer science curriculum. servation and then compare their algorithms In contrast, component-level integration happens to those in the research literature. by introducing course components, such as lectures, exercises, and projects, with sustainability themes Resources for sustainability-themed computer into computer science courses that do not have a science course components, and intelligent systems

SEpTEMbEr/ocTobEr 2016 1541-1672/16/$33.00 © 2016 IEEE 93 2469-7087/17/$33.00 © 2017 IEEE Published by Publishedthe IEEE Computer by the IEEE Society Computer Society June 2017 49 Table 1. Topics covered by computational sustainability courses discovered using Mallet.

Topic number Topic words Topic name 0 model, data, species, climate, prediction, ecology, monitoring, distribution, methods Ecological modeling 1 energy, emissions, energy-consumption, environmental, data, planetary-boundaries, systems, Energy reduce, smart-meter 2 environmental, food-web, health, economic, food, development, complex, services, Human health and sustainable wealth 3 infrastructure, fish, urban optimization, ocean, agriculture, water, cost, systems City and agriculture 4 networks, optimization, algorithms, conservation, habitat, wildlife-corridors, graph, budget, Optimization and planning planning

specifically, can be found on the Web; • Computational Methods in Sus- computational sustainability courses. see, for example, Artificial Intelligence tainable Energy (Carnegie Mellon In particular, we ran the Mallet for Computational Sustainability: A Lab University, Fall 2012). topic-modeling tool (http://mallet. Companion3 and “Teaching Sustainabil- • Computational Sustainability (Uni- cs.umass.edu) on all the abstracts (or ity in Computer Science” (http://serc. versity of British Columbia, Winter introductory sections) of the readings carleton.edu/sisl/sustain_in_cs.html). 2013–2014). from these courses. Mallet uses word These follow in the tradition of more gen- • Computational Sustainability (Geor- frequencies and other factors to find eral resource repositories, such as Nifty gia Tech, Spring 2014). topics that cover the document collec- Assignments (http://nifty.stanford.edu), • Seminar on Computational Sustain- tion. The details are not vital for our IEEE Real World Engineering Projects ability: Algorithms for Ecology and purposes, but we experimented with (www.realworldengineering.org), and Conservation (University of Massa- various numbers of topics, then used MERLOT (Multimedia Educational chusetts Amherst, Spring 2014). Mallet to discover the five best topics Resource for Learning and Online • Topics in Computational Sustain- as reflected in the abstracts, after we Teaching; www.merlot.org). Ideally, fine- ability (Stanford University, Spring did some basic preprocessing (such as grained educational materials in sustain- 2016). stemming and hyphenation). Table 1 ability-focused computer science will lists characteristic words for each find their way into these more general These courses typically have a semi- topic found using Mallet. We named repositories as well. nar format, with pedagogical elements each topic. The topic words are domi- such as readings from the research and nated by sustainability-related words, Course-Level Integration popular literature, responses to these rather than computing words, which At a higher granularity level is the in- readings through short papers and in- merits further investigation; using the troduction of entire sustainability- class discussions, and semester proj- entirety of articles from courses’ read- themed courses into computing curricula. ects. Most of the courses are for up- ing lists, rather than just abstracts or Again, this might include courses in per-division undergraduate and grad- introductory sections as we did, could sustainability-area disciplines that uate students. yield somewhat different results. have a significant computing compo- Mallet also can take a single doc- nent. Here, we limit our attention to Content of Sustainability-Themed ument and characterize it by topics computing courses offered in computer Computing Courses that it has discovered. We appended science or similar departments. Collectively, the courses cover sustain- the abstracts of each course into a ability areas of wildlife conservation, single “document” per course and Courses to Date water distribution and safety, climate used Mallet to assign each course Our survey of sustainability-themed com- change, energy, agriculture, urban design, proportions using the topics from puting courses includes the following: disaster response, poverty, and human Table 1. For a given course, the topic health. Computing areas that are cov- proportions sum to 1.0. Table 2 • Sustainability and Assistive Comput- ered collectively across all courses in- gives a visual representation of each ing (Bryn Mawr College, Fall 2010). clude machine learning, optimization, course’s topic coverage. Bolded text • Computing and the Environment simulation, systems, hardware, and al- illustrates the topics of each course (Vanderbilt University, Spring 2011). gorithmic analysis. The reading lists of with a proportion that exceeds the • Computing, Energy, and the Envi- these courses can be found online,4 and average of 0.20 = 1.0 divided by 5 ronment (Vanderbilt, Fall 2016). they are dominated by intelligent systems. (topics), which might be considered • Topics in Computational Sustainabil- Table 1 gives a list of topics for the most characteristic topics of the ity (Cornell University, Spring 2011). characterizing the previous crop of course. For example, Vanderbilt’s

94 www.computer.org/intelligent IEEE INTELLIGENT SYSTEMS 50 Computing Edge June 2017 Table 1. Topics covered by computational sustainability courses discovered using Mallet. Table 2. Topic proportions of previous computational sustainability courses. Syllabi and reading lists of these and other courses are available online.* Topic number Topic words Topic name 0 model, data, species, climate, prediction, ecology, monitoring, distribution, methods Ecological modeling Course content breakdown 1 energy, emissions, energy-consumption, environmental, data, planetary-boundaries, systems, Energy Academic institution offering course Topic 0 Topic 1 Topic 2 Topic 3 Topic 4 reduce, smart-meter Bryn Mawr College 0.25 0.17 0.25 0.05 0.28 2 environmental, food-web, health, economic, food, development, complex, services, Human health and Vanderbilt University 0.15 0.33 0.17 0.05 0.29 sustainable wealth Cornell University 0.18 0.00 0.42 0.32 0.08 3 infrastructure, fish, urban optimization, ocean, agriculture, water, cost, systems City and agriculture University of British Columbia 0.05 0.52 0.10 0.29 0.03 4 networks, optimization, algorithms, conservation, habitat, wildlife-corridors, graph, budget, Optimization and planning planning Georgia Tech 0.27 0.10 0.12 0.22 0.30 University of Massachusetts Amherst 0.68 0.00 0.00 0.00 0.32 specifically, can be found on the Web; • Computational Methods in Sus- computational sustainability courses. *http://en.wikibooks.org/wiki/Artificial_Intelligence_for_Computational_Sustainability:_A_Lab_Companion/List_of_Computational_Sustainability_Courses. see, for example, Artificial Intelligence tainable Energy (Carnegie Mellon In particular, we ran the Mallet for Computational Sustainability: A Lab University, Fall 2012). topic-modeling tool (http://mallet. course is most strongly character- Online courses and modules are an- tracts, or concentrations, and perhaps Companion3 and “Teaching Sustainabil- • Computational Sustainability (Uni- cs.umass.edu) on all the abstracts (or ized by topics 1 and 4. other avenue for creating educational even entire majors, for on-campus, ity in Computer Science” (http://serc. versity of British Columbia, Winter introductory sections) of the readings Instructors of subsequent courses computational sustainability materi- online, and hybrid settings. For ex- carleton.edu/sisl/sustain_in_cs.html). 2013–2014). from these courses. Mallet uses word can add to the online list of course als. They allow instructors to share ample, a specialized track might cover These follow in the tradition of more gen- • Computational Sustainability (Geor- frequencies and other factors to find readings and other resources as their lectures and other resources Machine Learning for Computational eral resource repositories, such as Nifty gia Tech, Spring 2014). topics that cover the document collec- appropriate. This list is intended as with a global audience. Many online Sustainability, with courses within the Assignments (http://nifty.stanford.edu), • Seminar on Computational Sustain- tion. The details are not vital for our a resource for instructors to compare courses follow a formal undergradu- track differentiated by sustainability IEEE Real World Engineering Projects ability: Algorithms for Ecology and purposes, but we experimented with and select readings as they plan and ate course design, with lectures, as- topics (for example, Machine Learning (www.realworldengineering.org), and Conservation (University of Massa- various numbers of topics, then used revise their own courses. signments, and exams that are auto- for Urban Operations and Machine MERLOT (Multimedia Educational chusetts Amherst, Spring 2014). Mallet to discover the five best topics or peer-graded, but examples exist Learning for Wildlife Conservation). Resource for Learning and Online • Topics in Computational Sustain- as reflected in the abstracts, after we Evolution of Computational of successful online courses in areas But we can expect more than courses Teaching; www.merlot.org). Ideally, fine- ability (Stanford University, Spring did some basic preprocessing (such as Sustainability Education such as literary analysis that have sig- and programs that are obvious inter- grained educational materials in sustain- 2016). stemming and hyphenation). Table 1 We have briefly surveyed educational nificant seminar design components sections of the parent areas of com- ability-focused computer science will lists characteristic words for each materials and courses to date and like the computational sustainability puting and sustainability, and more too find their way into these more general These courses typically have a semi- topic found using Mallet. We named pointed to online resources that in- courses of today, including live, vir- than prerequisite courses from each area repositories as well. nar format, with pedagogical elements each topic. The topic words are domi- structors and students can share and tual seminar discussions by students (such as Data Structure Tech-niques such as readings from the research and nated by sustainability-related words, expand. We close with some thoughts and instructors across the globe. from computing and Basic Hydrology Course-Level Integration popular literature, responses to these rather than computing words, which on the future of education at the nexus The online medium also offers unique from sustainability). At a higher granularity level is the in- readings through short papers and in- merits further investigation; using the of computing and sustainability. Com- opportunities for both instructors and What will be the design principles troduction of entire sustainability- class discussions, and semester proj- entirety of articles from courses’ read- putational sustainability is quickly be- students to collaborate across tradi- for larger curricular constructs in com- themed courses into computing curricula. ects. Most of the courses are for up- ing lists, rather than just abstracts or coming the commonly accepted name tional institutional boundaries.6 The putational sustainability education? Again, this might include courses in per-division undergraduate and grad- introductory sections as we did, could for this nexus,5 and we use it here to time seems right for a global course in One ideal is that computational sus- sustainability-area disciplines that uate students. yield somewhat different results. signify a maturing field. computational sustainability, engaging tainability majors be designed around have a significant computing compo- Mallet also can take a single doc- many of the researchers and instructors frameworks of evidence-based decision nent. Here, we limit our attention to Content of Sustainability-Themed ument and characterize it by topics On-Campus and Online Courses who have pioneered the field to date. making.7 After all, it is the myopia of computing courses offered in computer Computing Courses that it has discovered. We appended A seminar format, which dominates Such a course would undoubtedly be hy- human decision making that compu- science or similar departments. Collectively, the courses cover sustain- the abstracts of each course into a current computational sustainabil- brid, with much of the content delivered tational sustainability is intended to ability areas of wildlife conservation, single “document” per course and ity courses, is typical in a new area online, but with local cohorts meeting correct. Courses to Date water distribution and safety, climate used Mallet to assign each course in which the instructor is exploring physically and perhaps customizing the Our survey of sustainability-themed com- change, energy, agriculture, urban design, proportions using the topics from the territory, at least pedagogically. As computational sustainability content to puting courses includes the following: disaster response, poverty, and human Table 1. For a given course, the topic computational sustainability matures, the environmental and infrastructure The nexus between sustainability health. Computing areas that are cov- proportions sum to 1.0. Table 2 we can expect that courses with more challenges of their region. and computing is natural and impor- • Sustainability and Assistive Comput- ered collectively across all courses in- gives a visual representation of each formal designs will emerge, presumably tant—computing is ubiquitous, and ing (Bryn Mawr College, Fall 2010). clude machine learning, optimization, course’s topic coverage. Bolded text following along with the bottom-up de- Curriculum-Level Integration: sustainability concerns, knowledge, • Computing and the Environment simulation, systems, hardware, and al- illustrates the topics of each course velopment of a rich set of educational Tracks, Minors, and Majors and strategies should be ubiquitous, (Vanderbilt University, Spring 2011). gorithmic analysis. The reading lists of with a proportion that exceeds the resources as part of component-level in- As more courses develop and become too. Just as many fields are integrat- • Computing, Energy, and the Envi- these courses can be found online,4 and average of 0.20 = 1.0 divided by 5 tegration. Most importantly, textbooks differentiated in computational sus- ing computing into their educational ronment (Vanderbilt, Fall 2016). they are dominated by intelligent systems. (topics), which might be considered in computational sustainability can tainability, instructors will recognize curricula, it seems wise to integrate • Topics in Computational Sustainabil- Table 1 gives a list of topics for the most characteristic topics of the be both a driver and manifestation of and design complementary courses sustainability science and engineering ity (Cornell University, Spring 2011). characterizing the previous crop of course. For example, Vanderbilt’s greater maturity in the field. that can be arranged into minors, into computing education.

94 www.computer.org/intelligent IEEE INTELLIGENT SYSTEMS SEpTEMbEr/ocTobEr 2016 www.computer.org/intelligent 95 www.computer.org/computingedge 51 This article originally appeared in IEEE Intelligent Systems, vol. 31, no. 5, 2016.

We have surveyed existing resources 2. D.A. Patterson and J.I. Hennessy, Com- 6. D.H. Fisher and A. Fox, Report on the and courses that integrate environmen- puter Organization and Design: The CCC-CRA Workshop on Multidisci- tal and societal sustainability into com- Hardware/Software Interface, 3rd ed., plinary Research for Online Education, puting higher education, particularly Morgan Kaufmann, 2005. Feb. 2013; cra.org/ccc/wp-content/ with respect to intelligent systems. In- 3. Artificial Intelligence for Computa- uploads/sites/2/2013/02/CCC-MROE- terested readers are encouraged to use tional Sustainability: A Lab Com- Report.pdf. the online resources for more details panion; http://en.wikibooks.org/ 7. D. Fisher, “Recent Advances in AI for and to add to them. wiki/Artificial_Intelligence_for_ Computational Sustainability,” IEEE Computational_Sustainability:_A_ Intelligent Systems, vol. 31, no. 4, Lab_Companion. 2016, pp. 56–61. Acknowledgments 4. “List of Computational Sustainability Douglas Fisher, Zimei Bian, and Se- Courses,” Artificial Intelligence for is an associate professor in lina Chen are supported by NSF grant Douglas H. Fisher #1521672 for research, education, and Computational Sustainability: A Lab the Department of Electrical Engineering and reporting on advances in computing and Companion; http://en.wikibooks.org/ Computer Science at Vanderbilt University. Con- sustainability. wiki/Artificial_Intelligence_for_ tact him at [email protected]. Computational_Sustainability:_A_ References Lab_Companion/List_of_Computational_ Zimei bian is a senior undergraduate student 1. J. Petersen et al., “Dormitory Residents Sustainability_Courses. in computer science at Vanderbilt University. Reduce Electricity Consumption When 5. C. Gomes, “Computational Sustain- Contact her at [email protected]. Exposed to Real-Time Visual Feedback ability: Computational Methods for a and Incentives,” Int’l J. Sustainability Sustainable Environment, Economy, Selina chen is a senior undergraduate student in Higher Education, vol. 8, no. 1, and Society,” Bridge, vol. 39, no. 4, in computer science at Vanderbilt University. 2007, pp. 16–33. 2009, pp. 5–13. Contact her at [email protected].

Harlan D. Mills Award

Take the CS Library wherever you go! Call for SoftwareIEEE Computer Engineering Society magazines andAward Transactions Nominations are now available to subscribers in the portable ePub format. Established in Harlan D. Mills’ name to recognize researchers and practitioners Just download the articles from the IEEE Computer Society Digital whoLibrary, have and demonstrated you can read them long-standing, on any device that sustained,supports ePub. and For meaningfulmore contributionsinformation, toincluding the theory a list of compatibleand practice devices, of visit the information sciences, focusing on contributions to the practice of software engineering through the applicationwww.computer.org/epub of sound theory. The award consists of a $3,000 honorarium, plaque, and a possible invited talk during the week of the annual International Conference on Software Engineering (ICSE), co-sponsored by the Deadline for 2018 Nominations: IEEE Computer Society Technical Council on Software Engineering. 15 October 2017 The award nomination requires at least 3 endorsements. Nomination site: Self-nominations are not accepted. awards.computer.org Nominees/nominators do not need to be IEEE or IEEE Computer Society members.

96 www.computer.org/intelligent IEEE INTELLIGENT SYSTEMS 52 Computing Edge June 2017 SkillChoice™ Complete Now with expanded libraries and an upgraded platform! Valued at $3,300! 3,000+ MENTORSHIP u online courses 28,000+ books 6,000+ Practice videos Exams 15,000+ Books24x7 titles

OVER 20x as many resources as before

One membership. Unlimited knowledge.

Did you know IEEE Computer Society membership comes with access to a high-quality, interactive suite of professional development resources, available 24/7?

Powered by Skillsoft, the SkillChoice™ Complete library contains more than $3,000 worth of industry-leading online courses, books, videos, mentoring tools and exam prep. Best of all, you get it for the one low price of your Preferred Plus, Training & Development, or Student membership package. There’s something for everyone, from beginners to advanced IT professionals to business leaders and managers.

The IT industry is constantly evolving. Don’t be left behind. Join the IEEE Computer Society today, and gain access to the tools you need to stay on top of the latest trends and standards.

Learn more at www.computer.org/join.

ACCESS TO SKILLSOFT IS AVAILABLE WITH

NEW PREFERRED PLUS MEMBERSHIP

OPTIONS TRAINING & DEVELOPMENT FOR A BETTER FIT. RESEARCH

BASIC

STUDENT

And a better match for your career goals. IEEE Computer Society lets you choose your membership — and the benefits it provides — to fit your specific career needs. With four professional membership categories and one student package, you can select the precise industry resources, offered exclusively through the Computer Society, that will help you achieve your goals.

Learn more at www.computer.org/membership. Achieve your career goals with the fit that’s right for you. Explore your options below.

Training & Preferred Plus Research Basic Student Select your Development $60 $126 $55 $115 $55 $115 $40 $99 $8 membership IEEE Affiliate IEEE Affiliate IEEE Affiliate IEEE Affiliate Does not include Member Member Member Member Member Member Member Member IEEE membership

Computer magazine (12 digital issues)*

ComputingEdge magazine (12 issues)

Members-only discounts on conferences and events

Members-only webinars

Unlimited access to Computing Now, computer.org, and the new mobile-ready myCS

Local chapter membership

Skillsoft’s Skillchoice™ Complete with 67,000+ books, videos, courses, practice exams and mentorship resources Books24x7 on-demand access to 15,000 technical and business resources

Two complimentary Computer Society magazine subscriptions

myComputer mobile app 30 tokens 30 tokens 30 tokens

12 FREE 12 FREE Computer Society Digital Library Member pricing Member pricing Included downloads downloads

3 FREE 3 FREE Training webinars Member pricing Member pricing Member pricing webinars webinars

Priority registration to Computer Society events

Right to vote and hold office

One-time 20% Computer Society online store discount

* Print publications are available for an additional fee. See catalog for details.

www.computer.org/membership COMPUTING CAREERS

Careers in Emerging Technologies for People Changing Jobs from Industry to Academia

or this ComputingEdge issue, we focus promotions. However, work in industry proved on emerging technologies as they relate unstable, and unsuccessful company campaigns F to an increasing popular career transi- and projects encouraged me to change jobs sev- tion for computing professionals—the shift from eral times. industry to academia. Prior to obtaining a full Eventually, a friend who was a physics profes- professorship in information systems at Califor- sor told me about a tenure-track position in Cal nia State University Fullerton, Sorel Reisman held State Fullerton’s business school. Having had senior management positions at IBM, Toshiba, enough of the insecurity of everything related to and EMI in the US and Canada. He served as 2011 the computer industry, I applied for and was hired IEEE Computer Society president and is currently as an associate professor. Seven years later, I got a member of both the IEEE Publications Services tenure and was promoted to full professor. and Products Board and the IEEE Education Activ- ities Board. ComputingEdge: What are the opportunities in academia for pursuing an interest in emerging ComputingEdge: You spent considerable time technologies? in industry working for multinational companies, starting as an engineer and rising to vice president Reisman: An academic can pursue an interest of development. Why did you leave for academia? in emerging technologies in three areas, which align with the three criteria used to assess a pro- Reisman: When I fi nished graduate school, I fessor’s work performance: research, teaching, fully intended to pursue an academic career. But and service. academic positions were in short supply at the When I entered academia, I didn’t have a spe- time, so I looked for a job in industry. And once cifi c research agenda. However, I’d been involved there, I got caught up in the dynamic of raises and with multimedia computing in industry, and it

56 June 2017 Published by the IEEE Computer Society 2469-7087/17/$33.00 © 2017 IEEE seemed reasonable to work in that area again as Also, in industry, you usually work on common a professor. How much real research you can do objectives as a team with others. As a professor, in emerging technologies depends on the funding you typically don’t work in teams, and your objec- available for that kind of work. In industry, com- tives may be completely diff erent from those of panies typically pay for their own projects. In aca- your colleagues. demia, on the other hand, external grants typically fund research. Your success in obtaining a grant ComputingEdge: Are you glad to have moved determines your ability to pursue your interest in from industry to academia? emerging technologies. In terms of instruction, undergraduate courses Reisman: Absolutely! Probably the best thing off er limited opportunities to teach emerging tech- about academia is the freedom to pursue your own nologies, but there’s more freedom to do this with interests, whether personal or professional. Also, graduate level courses. At the undergraduate level, unlike industry, whose projects go on for a long accredited academic departments must adhere to time, academic work is divided into shorter seg- a prescribed set of topics to provide students with ments—semesters. If you’re teaching a class you foundational knowledge. You can introduce newer, don’t like, it ends with the semester. And you can advanced topics in some courses, but you don’t change courses over time to make the experience have time to deal with them in depth. better. In addition, breaks between semesters and You can do as I did and propose optional school years enable you to recharge your batteries. courses related to an emerging technology, but there might not be enough interested students to ComputingEdge: What important advice would justify the university off ering the class. I created you give colleagues considering moving from a course called “personal computer systems and industry to academia? architectures” that became very popular, just as PCs were being widely adopted. I also designed Reisman: Don’t assume that because you held a class on e-commerce systems—which was just senior management positions in industry that starting to become an emerging topic of interest— your academic colleagues will value your indus- that was popular with graduate students. try accomplishments. The kinds of skills that Service means volunteer personal or profes- helped you achieve those positions are mostly sional community work. I chose to invest my ser- irrelevant in academia, unless you take an admin- vice time with the IEEE Computer Society, which istrative job. has paid off immeasurably. I’ve learned about many new technologies, and my involvement with the IEEE and Computer Society digital libraries omputingEdge’s Lori Cameron inter- has enabled me to bring new ideas and concepts viewed Reisman for this article. Contact to the work I do it at the university. C her at [email protected] if you would like to contribute to a future ComputingEdge ComputingEdge: Can you describe one or two article on computing careers. Contact Reisman at major diff erences between working in industry [email protected]. versus working in an academic environment

Reisman: The environments are completely diff erent. For example, they don’t work with the same decision-making time frames. Industry tends Read your subscriptions through to make decisions and act on them much more the myCS publications portal at http://mycs.computer.org. quickly than academia. www.computer.org/computingedge 57 CAREER OPPORTUNITIES

SOFTWARE DEVELOPER IT Company master, business analyst, &developer), improve current mortgage LOS &dvlp (So. Plainfield, NJ) seeks software devel- planning the deliverable by sprint or bi- new LOS. Mail resume to Ref#1845, 1414 oper to participate in the full Software weekly cycle, execution of the dvlpmt, E Maple Rd, Troy, MI 48083. Development Lifecycle (“SDLC”) involv- &delivery of the product to production ing design, development, testing and im- environment. Use SQL stored procedure, SR. QUALITY ASSURANCE ANALYST: plementation in n-tier development envi- SQL server agent job, restful API, Asp. Peterson Technology Partners Inc. seeks ronment by utilizing ASP.Net, C#, HTML, Net 3.5/4.0 .NET framework, C#, HTML, qualified Sr. Quality Assurance Ana- CSS, JQuery, SQL Server, ASP.Net MVC. Scheme, CSS, Ajax, bootstrap &JavaS- lyst for its headquarters located in Park Maximize customer satisfaction by main- cript, to improve current mortgage Loan Ridge, IL & various & unanticipated work taining, understanding, adding function- Origination System (LOS) &dvlp new Loan locations thruout the U.S. Resp. for lead- ality and meeting end-user acceptance Origination System. Perform Web dvlpmt ing testing of multiple, large, complex, or criteria through the design and imple- in .Net using Microsoft Visual Studio 2013 strategic software devel. apps using QA mentation of software test strategies for &SQL Server 2008/2012. Design &dvlp testing practices & frameworks. Bache- new features. Analyze, participate, re- apps using Asp.net, C# using domain lor’s degree in Comp Sci, Comp Engg, or view the requirements, create and assist layer architecture which includes repos- a closely related field of study required in performing various types of tests and itory, services layer, comm&bus frame- w/ at least 2 yrs exp in: (i) developing & code review. Mail resumes to HRD, Code work, object oriented programming. executing test scenarios, test scripts, Ace Solutions, Inc., 50 Cragwood Rd., Conduct code reviews for team members test data docs based on design in agile Ste. 217, So. Plainfield, NJ 07080. to make sure code is following standard environment as well as testing web ser- both on C# side &HTML, CSS, JavaScript, vices utilizing MTM, TFS, Protractor 2.0; PROGRAMMER ANALYST, United Shore &JQuery. Bachelor, Computer Informa- (ii) experience w/ test automation; & (iii) Financial Services, Troy, MI. Follow Agile tion Systems, Business Information Sys- working on different technologies during methodology for the process of scrum to tems, or Computer Science. 12 mos exp as devel. Incl. JavaScript, HTML, XML, PHP, determine the complexity of Loan Origi- Programmer Analyst or System Engineer, Python & JSON, MS Visio & Crystal Re- nation System projects, initiating meet- using SQL stored procedure, SQL server ports. An EOE. Respond by mail to Peter- ings with business partners to determine agent job, restful API, Asp.Net 3.5/4.0 son Tech Partners, 1030 W Higgins Rd. the minimum viable product, determine .NET framework, C#, HTML, Scheme, Ste 230, Park Ridge, IL 60068. Refer to the necessary resources (such as scrum CSS, Ajax, bootstrap &JavaScript, to ad code: PTP-0417.

SOFTWARE SOFTWARE TECHNICAL Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for TECHNICAL SOFTWARE SOFTWARE ANALYSTS- DEVELOPER SUPPORT DEVELOPER positions in positions in Burlington, MA. Colorado Springs, Colorado. positions in Cambridge, MA. Job duties include: Design, develop, Job duties include: deliver solutions to troubleshoot and/or test/QA Job duties include: Design, develop, trouble- the Oracle customer base while serving software. shoot and/or test/QA software. as an advocate for customer needs; offer strategic technical support to assure the Apply by e-mailing resume to Apply by e-mailing resume to highest level of customer satisfaction. [email protected], [email protected], referencing 385.20153. Apply by e-mailing resume to referencing 385.15409. [email protected], Oracle supports workforce diversity. referencing 385.18255. Oracle supports workforce diversity. Oracle supports workforce diversity.

SENIOR SOFTWARE ENGINEER, Warren, present the user to update the software application, migration, infrastructure/ MI, General Motors. Design, dvlp code, when software revived Over the Air (OTA) operations &architecture teams to tech- review, test, integrate &verify embedded or removable disk with the software nically enable migrations. Ensure Appli- automotive ECU software incldg infotain- (USB). Bachelor, Electronics &Communi- cation teams adequately document &ad- ment &safety modules for vehicles to play cation Engrg, Electrical Engrg or related. here to internal process in building UNIX tuner audio sources such as Amplitude 60 mos of post baccalaureate &progres- &Windows servers as per GM require- modulation (AM), Frequency modulation sive exp as Software Engineer, Technical ments. Use Project Management (incldg (FM), Sirius Radio (SiriusXm), Media, Nav- Lead, or related, designing/dvlping, re- project charters, plans &schedules) &Mi- igation, OnStar audio, Android auto &car viewing, integrating, &verifying embed- crosoft tools to capture business impact play using Android, real time OS (RTOS), ded automotive ECU infotainment &/or detection &response project scope, time C, CAPL, &CAN. Dvlp, configure &exe- safety system software for vehicles, us- &cost. Master, Information Technology, cute software to play safety &driver alert ing C, CAPL, RTOS, &CAN. Mail resume Management Information Systems or chimes for different vehicle architectures, to Ref#2791, GM Global Mobility, 300 related. 2 mos exp as Business Analyst global market (Europe, China, NA &rest Renaissance Center, MC:482-C32-D44, or Systems Analyst, using Project Man- of the world) using calibrations, C, C++, Detroit, MI 48265. agement (incldg project charters, plans Java programming languages &vehicle &schedules) &Microsoft tools to capture protocols such as CAN. Lead, dvlp, exe- DATA CENTER TRANSFORMATION IT project scope, time &cost. Mail resume cute, review, test &integrate the Android MANAGER, Warren, MI, General Motors. to Ref#34285, GM Global Mobility, 300 application software for electric &hybrid Work with Innovation team members to Renaissance Center, MC:482-C32-D44, vehicles to provide energy performance understand data center migration ap- Detroit, MI 48265. &charging related information in car radio proaches, business continuity switchover for current &future model year electric plans, architectures &designs to migrate CALYPSO TECH. seeks Quality Assu- &extended range electric vehicles. Lead, data center infrastructure among multi- ance Analyst in SF, CA to guarantee qual- estimate, conduct scrum &review work ple data centers. Use MS Visio to review ity & persistence of SW by testing correc- of 10-member scrum team to achieve the data center architectures. Identify, triage tions of bugs & enhancmnts. Ref Job ID: project goals based on Agile software &work with support providers to iden- 9GYTTG & mail res. to Calypso, Attn: B. methodologies. Dvlp Human Machine In- tify solutions to technical obstacles. Act Grandin, 595 Market St, Ste. 1800, San terface Screens using Android &Java to as technical liaison with &between the Francisco, CA 94105.

CLASSIFIED LINE AD SUBMISSION DETAILS: Rates are 2018-2019 $425.00 per column inch ($640 minimum). Eight lines per col- IEEE-USA Government umn inch and average five typeset words per line. Send copy Fellowships at least one month prior to publication date to: Debbie Sims, Classified Advertising, Computing Edge Magazine, 10662 Los Congressional Fellowships Seeking U.S. IEEE members interested in Vaqueros Circle, Los Alamitos, CA 90720; (714) 816-2138; fax spending a year working for a Member of (714) 821-4010. Email: [email protected]. Congress or congressional committee.

In order to conform to the Age Discrimination in Employment Engineering & Diplomacy Fellowship Act and to discourage age discrimination, Computing Edge Seeking U.S. IEEE members interested in may reject any advertisement containing any of these phras- spending a year serving as a technical adviser es or similar ones: “…recent college grads…,” “…1–4 years at the U.S. State Department. maximum experience…,” “…up to 5 years experience,” or “…10 years maximum experience.” Computing Edge reserves the USAID Fellowship right to append to any advertisement without specific notice Seeking U.S. IEEE members who are interested in serving as advisors to the U.S. government to the advertiser. Experience ranges are suggested minimum as a USAID Engineering & International requirements, not maximums. Computing Edge assumes that Development Fellow. since advertisers have been notified of this policy in advance, The application deadline for 2018-2019 they agree that any experience requirements, whether stat- Fellowships is 8 December 2017. ed as ranges or otherwise, will be construed by the reader as For eligibility requirements and application information, go to minimum requirements only. Computing Edge encourages em- www.ieeeusa.org/policy/govfel or contact Erica Wissolik by emailing ployers to offer salaries that are competitive, but occasionally [email protected] or by calling +1 202 530 8347. a salary may be offered that is significantly below currently acceptable levels. In such cases the reader may wish to inquire of the employer whether extenuating circumstances apply.

www.computer.org/computingedge 59 CAREER OPPORTUNITIES

HMI MODELING & PROTOTYPING infotainment prototypes using Android, MPHASIS CORP has multi openings at SOFTWARE LEAD, Warren, MI, General QNX &proprietary OS platforms using various levels for the follow’g positions Motors. Lead member of cross-func- C/C++, Java, HTML5. Master, Computer at its office in NY, NY & unanticipated tional team, dvlp emerging embedded Science, Information Technology, Soft- client sites thr/o the US 1. Info. Sys. telematics features in ECUs focused on ware Systems or related. 12 mos exp as Anyst* - Ana. & provide sys req & spec. Human Machine Interfaces (HMI) across Engineer, dvlpg HMI embedded soft- 2. SW Dvlper* - Design, dvlp & modify multiple delivery channels, incldg multi- ware for psgr vehicle infotainment sys- SW sys. 3. Sys. Architect Dvlper* - Dvlp modal voice response systems, mobile tems incldg instrument clusters, center IT architecture 4. Graphic UI Desgr* - devices, &in-vehicle systems. Dvlp hu- stack modules incldg radios, integrating Design UI & perform UAT 5. N/W Infra man HMI embedded software for psgr CAN, LIN, Bluetooth handsfree apps, Eng* - Maintain & TRBL n/w, design, vehicle infotainment systems incldg in- &internet apps into automotive systems. dvlp, install n/w infra appl. 6. Business strument clusters, center stack modules Mail resume to Ref#44831, GM Global Operation Anyst* - Ana bus process incldg radios. Integrate CAN, LIN, Blue- Mobility, 300 Renaissance Center, thru app of s/w sol. 7. IT Mgr* - Plan & tooth handsfree apps, &internet apps MC:482-C32-D44, Detroit, MI 48265. manage the delivery of IT proj. 8. En- into automotive systems. Lead &mentor terprise Svc Engagem’t Mgr* - E2E sale team of 7 external engineers in embed- MOBILE APPLICATION IOS DEVEL- of IT svc/prod. 9. Eng Engagem’t Mgr* ded prototype software dvlpmt group OPER. Position available in Boston, MA. - Manage & direct business integration to dvlp future automotive infotainment Design and develop mobile applications of proj activities. 10. Mkt Dvlpt Mgr* - software. Research &dvlp advanced en- for iPads/iPhones using Xcode and In- Promote IT svc/prod. & impl bus plans. grg &research prototype software for fu- terface Builder. Create SOAP/REST web Must have a Bachelor/equiv and prior ture psgr vehicle infotainment systems. services. Incorporate current front-end rel. exp, Master/equiv, or Master/equiv Use IBM Rational Rhapsody tool for design patterns using JAVA and C. Col- and prior rel. exp. Edu/exp req vary de- modeling, visualizing, specifying, con- laborate with sponsors and develop- pending on position level/type. *Lead structing &documenting artifacts of soft- ment technicians to improve existing positions in this occupation must have ware intensive user interface systems. programs. Apply: C. Andre, MIP F115, Master/equiv+2yr or Bach/equiv+5yr Create reusable software component Massachusetts Mutual Life Insurance progressive exp. Travel/relo req. Send frameworks &standard interfaces using Company, 1295 State Street, Spring- resume & applied position to: recruit- Real Time Operating Systems for vehicle field, MA 01111; Please Reference Job ID: [email protected] or 460 Park Ave. display prototypes. Design &dvlp rapid 708203800. S., Ste# 1101, New York, NY 10016 Attn: Recruit.

SOFTWARE 23ANDME, INC. has a Business In- telligence Engineer position in Moun- tain View, CA. Conduct MicroStrategy Oracle America, Inc. development work including the development of schema, applications, has openings for configuration objects, and distribution services. To apply, please mail resume to 23andme, Inc., Attn: Grace Moeller, Juniper Networks is recruiting 899 W Evelyn Ave, Mountain View, CA SOFTWARE 94041. for our San Francisco, CA office: Software Engineer #38229: De- INFOSYS LIMITED, Associate Vice DEVELOPER sign, develop, troubleshoot and President Senior Global Sourcing Man- implement scalable cloud-based ager(s)- multiple, full-time openings in security features for Company Plano, TX to lead IT business planning positions in Broomfield, CO. products using Python and Java. for bid management and transition solution planning targeting Fortune 500 Juniper Networks is recruiting companies. Develop and manage strong for our Quincy, WA office: peer-level relationships with sourcing advisors and senior-level executives. Job duties include: Design, develop, Lab SW Services Admin #14086: 10% travel to various & unanticipated troubleshoot and/or test/QA Install network and/or computer client site locations throughout the equipment. Troubleshoot hard- U.S. required. All employment-related software. ware/software issues on installed travel expenses to be paid by employer equipment. pursuant to internal policy. Equal Op- Apply by e-mailing resume to portunity Employer M/F/D/V. Apply to Mail single-sided resume with Infosys Limited online at: https://www .infosys.com/careers/job-opportunities [email protected], job code # to /pages/index.aspx. Select the box la- referencing 385.20140. Juniper Networks beled “Technical” to access the “Ex- Attn: MS A.4.435 perienced Professionals” link in right 1133 Innovation Way column. Click on “Experienced Pro- fessionals” link. Create user account Oracle supports workforce diversity. Sunnyvale, CA 94089 & follow link for ‘Search Openings’ & click on ‘Advanced Search.’ Enter Job #22075BR in “Auto Req ID box.”

60 ComputingEdge June 2017 CAREER OPPORTUNITIES

METAMARKETS GROUP, INC. seeks a SOFTWARE DEVELOPER. Central NJ IT Software Engineer in San Francisco, CA. Consulting Company requires Software Contrib to a fluid set of distrib systms Developer at their primary Princeton, that pwr Metamarkets’ petabyte-scale NJ location, to Design, develop & im- data platform. Mail resume to Metamar- plement business software apps using kets Group, Inc., ATTN: E. Choung, 625 MSSQL Server/BI Developer ,T-SQL CLOUDERA, INC. IS RECRUITING 2nd St., St. 230, San Francisco, CA (DDL and DML) PL/SQL, SQL Server FOR OUR PALO ALTO, CA 94107. Reference code: 9ZX4KE. 2005/08/14,DTS, with experience in OFFICE: designing high level ETL architecture, HADOOP/JAVA ENGINEER #37533: creating and fine tuning ETL workflows. CUSTOMER ENGAGEMENT MAN- Work with Cloudera’s full stack of Candidates must have 1 year of manda- AGER – Technical Sales. Sell software technology, designing these systems tory exp in related -field. All positions & related tech. Develop proofs of con- on top of Hadoop, Hive, & HBase. cept in .Net framework & C# language. require: MS in CS/Engineering/Comp Travel to various & unanticipated loca- Applications/Business or related. BS SOFTWARE ENGINEER #37984: De- tions throughout continental U.S. req’d degree + 5yrs experience can be sub- velop software infrastructure neces- on bi-weekly basis. U.S. Master’s deg. in stituted for the MS requirement . Any sary to run large scale data process- Computer Science (Applied) or foreign combination of foreign edu + related ing service. equiv. req’d. Min 2 years’ exp. in sales exp equivalent to a US Masters, or any engineer pos’n(s) involving develop of combination of foreign edu +related exp UX DESIGN ENGINEER #38258: Re- proofs of concept in .Net framework and equivalent to a BS Degree will be ac- sponsible for user experience & de- C# language req’d. Great Software Lab- cepted. Travel to several unanticipated sign projects for Cloudera’s products oratory, Inc., Chicago, IL. Resumes to: locations all over US and might involve to successful delivery & completion. Recruiting, Great Software Laboratory, relocation consistent business needs. Inc., 401 S. LaSalle Street, Suite 801-W, Mail your resume to: Evantage Solutions SOFTWARE ENGINEER #37293: Chicago, IL 60605. Inc., Attn: HR, 3 Independence way, STE Write, maintain & run software for 209, Princeton NJ 08540. building & releasing Cloudera’s software.

CLOUDERA, INC. IS RECRUITING FOR OUR AUSTIN, TX OFFICE: TECHNICAL TECHNICAL SOLUTIONS CONSULTANT #37577: Develop, implement & test Ha- Oracle America, Inc. doop-based solutions for Cloudera Oracle America, Inc. customers. Travel required. Telecom- has openings for has openings for muting allowed. CLOUDERA, INC. IS RECRUITING FOR OUR NEW YORK, NY TECHNICAL TECHNICAL OFFICE: SOLUTIONS ARCHITECT #37609: ANALYST- Drive technical customer conversa- ANALYST tions, understand customer requirements, either explicit or inferred, and SUPPORT positions in Lehi, UT. transfer that knowledge into a technical plan of action. Travel required. positions in Morrisville, North Carolina. Job duties include: deliver solutions to CLOUDERA, INC. IS RECRUITING Job duties include: deliver solutions to the Oracle customer base while serving FOR OUR RALEIGH, NC OFFICE: SOLUTIONS ARCHITECT #36922: the Oracle customer base while serving as an advocate for customer needs; Drive technical customer conversa- as an advocate for customer needs; offer tions, understand customer require- offer strategic technical support to ments, either explicit or inferred, and strategic technical support to assure the transfer that knowledge into a tech- assure the highest level of customer nical plan of action. Travel required. highest level of customer satisfaction. satisfaction. MAIL RESUME WITH JOB Apply by e-mailing resume to Apply by e-mailing resume to CODE # TO [email protected], [email protected], CLOUDERA referencing 385.19568. referencing 385.20046. ATTN: HR Oracle supports workforce diversity. 1001 PAGE MILL RD., BLDG. 3 Oracle supports workforce diversity. PALO ALTO, CA 94304

www.computer.org/computingedge 61 TECHNICAL TECHNICAL TECHNICAL Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for TECHNICAL TECHNICAL TECHNICAL ANALYST- ANALYST ANALYSTS positions in Lehi, Utah. SUPPORT positions in Colorado Springs, Colorado. Job duties include: Deliver solutions to positions in Colorado Springs, CO. the Oracle customer base while serving Job duties include: deliver solutions to as an advocate for customer needs. Job duties include: Deliver solutions the Oracle customer base while serving as an advocate for customer needs; offer Offer strategic technical support to to the Oracle customer base while strategic technical support to assure the assure the highest level of customer serving as an advocate for customer highest level of customer satisfaction. satisfaction. needs. Apply by e-mailing resume to Apply by e-mailing resume to Apply by e-mailing resume to [email protected], [email protected], [email protected], referencing 385.19392. referencing 385.20112. referencing 385.19018. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

TECHNOLOGY TECHNICAL TECHNICAL Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for APPLICATIONS TECHNICAL TECHNICAL DEVELOPER ANALYST ANALYST- positions in Orlando, FL. positions in Lehi, Utah. SUPPORT Job duties include: Analyze, design, develop, troubleshoot and debug Job duties include: deliver solutions to positions in Lehi, Utah. software programs for commercial or the Oracle customer base while serving Job duties include: Deliver post-sales end-user applications. Write code, as an advocate for customer needs; complete programming and perform offer strategic technical support to support and solutions to the Oracle testing and debugging of applications. assure the highest level of customer customer base while serving as an satisfaction. advocate for customer needs. Apply by e-mailing resume to [email protected], Apply by e-mailing resume to Apply by e-mailing resume to referencing 385.20299. [email protected], [email protected], referencing 385.20583. referencing 385.19912. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

62 ComputingEdge June 2017 CAREER OPPORTUNITIES

SOFTWARE DEVELOPER Central NJ IT CONSULTANT, PISCATAWAY, NJ, busi- content mngmnt sys & dev web apps. Consulting Company requires Software ness consulting services using business Software Engineer (Req#A7Y2QT) De- Developer at their primary Princeton, NJ analytics & statistical modeling. Up to sign & dev features & improvements for location, to Design, develop & implement 100% travel. Mail resume to Ericsson Inc. indexer clustering capability of Splunk business software apps using C#,ASP. 6300 Legacy Dr, R1-C12 Plano, TX 75024 Enterprise. Refer to Req# & mail resume NET MVC4.0,5.0, JavaScript, jQuery, ID# 16-NJ-3348. PROJECT MANAGER, to Splunk Inc., ATTN: J. Aldax, 270 Bran- Angular JS, JSON, HTML5, SSIS tools PLANO, TX, program lifecycle mgmnt, nan Street, San Francisco CA 94107. In- and Net Framework4.5 with experience E2E biz perf mgmnt & analytics projects. dividuals seeking employment at Splunk in SCRUM Framework and SQL SERVER Mail resume to Ericsson Inc. 6300 Leg- are considered without regards to race, 2012 Candidates must have 1 year of acy Drive, R1-C12 Plano, TX 75024 ID# religion, color, national origin, ancestry, mandatory experience in related field. 16-TX-166. BUSINESS CONSULTANT, sex, gender, gender identity, gender All positions require: MS in CS/Engineer- PLANO, TX business consulting ser- expression, sexual orientation, marital ing/Comp Science/Science or related. vices using business analytics & statis- status, age, physical or mental disability BS degree + 5yrs exp can be substituted tical modeling. Up to 100% travel. Mail or medical condition (except where phys- for the MS requirement any combination resume to Ericsson Inc. 6300 Legacy Dr, ical fitness is a valid occupational qual- of foreign edu +related exp equivalent to R1-C12 Plano, TX 75024 ID# 16-TX-3358. ification), genetic information, veteran a US Masters, or any combination of for- ENGINEER-SYSTEMS, PLANO, TX pro- status, or any other consideration made eign edu and related experience equiv- vide installation & network integration, unlawful by federal, state or local laws. alent to a BS Degree will be accepted. technical documentation to support sol’n To review US DOL’s EEO is The Law no- Travel to several unanticipated locations Life Cycle Mgmt. Travel required. Mail tice please visit: https://careers.jobvite. all over US, might involve relocation resume to Ericsson Inc. 6300 Legacy com/Splunk/EEO_poster.pdf. To review consistent business needs. Mail your Drive, R1-C12 Plano, TX 75024 Job # Splunk’s EEO Policy Statement please resume to: Kellton Tech Inc., Attn: HR, 3 17-TX-2193. visit: http://careers.jobvite.com/Careers Independence Way, STE 209, Princeton /Splunk/EEO-Policy-Statement.pdf. Pur- NJ 08540. SPLUNK INC. has the following job oppor- suant to the San Francisco Fair Chance tunities in San Francisco, CA: Software Ordinance, we will consider for employ- ERICSSON INC. has open positions Engineer, Applications (Req#9YQ425) ment qualified applicants with arrest and for the following positions: BUSINESS Design & dev web properties in new conviction records.

TECHNICAL INFO TECH Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for SOFTWARE TECHNICAL IT BUSINESS DEVELOPMENT ANALYST- IMPLEMENTATION MANAGER ANALYST SUPPORT positions in Rocklin, California. positions in Redwood Shores, CA. positions in Lehi, Utah. Job duties include: As part of a software Job duties include: Manage a team that Job duties include: Deliver solutions to project implementation team assists in designs, develops, troubleshoots and the Oracle customer base while serving debugs software programs for producing implementing software projects by automation. Monitor software build and as an advocate for customer needs. mapping business processes, producing test pipelines and resolve issues. documentation, establishing acceptance Apply by e-mailing resume to testing criteria. Apply by e-mailing resume to [email protected], Apply by e-mailing resume to [email protected], referencing 385.19858. referencing 385.15380. [email protected], referencing 385.19829. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

www.computer.org/computingedge 63 CONSULTANT PRODUCT MANAGER TECHNOLOGY Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for PRODUCT SYSTEMS CONSULTANT ANALYST- positions in Redwood Shores, CA. MANAGER positions in Redwood Shores, CA. SUPPORT Job duties include: Analyze requirements positions in Redwood Shores, CA. and deliver functional and technical Job duties include: Participate in all software solutions. Implement products and and/or hardware product development life Job duties include: Analyze user require- technologies to meet post-sale customer cycle activities. Move software products ments to deliver post-sales support and needs. Travel to various unanticipated sites through the software product development solutions to the Oracle customer base while throughout the United States required. May cycle from design and development to serving as an advocate for customer needs. telecommute from home. implementation, testing and/or marketing. Travel to various unanticipated sites May telecommute from home. throughout the United States required. Apply by e-mailing resume to Apply by e-mailing resume to Apply by e-mailing resume to [email protected], [email protected], referencing 385.18107. [email protected], referencing 385.19824. referencing 385.15540. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

TECHNICAL TECHNICAL QA ANALYST Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for TECHNICAL TECHNICAL QA ANALYST ANALYST ANALYST positions in Orlando, FL. positions in Orlando, FL. positions in Redwood City, California.

Job duties include: Deliver solutions to Job duties include: Deliver solutions to Job duties include: responsible for develop- the Oracle customer base while serving the Oracle customer base while serving ing, applying and maintaining quality as an advocate for customer needs. as an advocate for customer needs. standards for company products with Offer strategic technical support to Offer strategic technical support to adherence to both internal and external assure the highest level of customer assure the highest level of customer standards; develop and execute software test plans. satisfaction. satisfaction. Apply by e-mailing resume to Apply by e-mailing resume Apply by e-mailing resume to [email protected], [email protected], [email protected], referencing 385.18346. referencing 385.19798. referencing 385.19812. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

64 ComputingEdge June 2017 TECHNICAL Oracle America, Inc. CAREER OPPORTUNITIES has openings for

IT FINANCE PROJECT MGR.: Coord. SOFTWARE DEVELOPER: Design, de- finance/ops IT proj. for FP&A, reporting; velop, test and implement s/w using TECHNICAL ERP sol’ns e.g. MS Dynamics Axapta, knowledge in Java, C++, JNI, Android SAP. Ensure compliance w/ accounting SDK, NDK, ADB, DDMS, Logcat, Trace- stds. Use SAP, MS Dynamics Axapta, view, GIT, JIRA and Agile Development ANALYST Cognos, Tableau, SQL, 6 Sigma, Lean. Lifecycle. Must be willing to travel & reloc positions in Orlando, FL. Travel 30%, US only. Req. BS in BusAd- to unanticipated client locations through- min, MIS or Finance; 7 yrs exp. Apply to out the US. Reqs MS in comp sci, eng or [email protected] w/ ITFPM in subj rel. Mail resumes to Informatic Technol- line. GEA Process Engineering Inc. Job in ogies Inc. 825 Georges Road, 3rd Floor, Job duties include: deliver solutions to Columbia MD. North Brunswick, NJ 08902. the Oracle customer base while serving as an advocate for customer Cisco Systems, Inc. is accepting resumes needs; offer strategic technical for the following positions: support to assure the highest level of AUSTIN, TX: Solutions Architect (Ref.# AUS16): Responsible for IT advisory and customer satisfaction. technical consulting services development and delivery. Telecommuting permitted. BELLEVUE/SEATTLE, WA: Software Engineer (Ref.# SEA1): Responsible for the definition, design, development, test, debugging, release, enhancement or maintenance of networking software. Network Consulting Engineer (Ref.# BEL6): Apply by e-mailing resume to Responsible for the support and delivery of Advanced Services to company’s [email protected], major accounts. Telecommuting permitted and Travel may be required to various unanticipated locations throughout the United States. referencing 385.20769. FULTON, MD: Manager, Technical Marketing (Ref.# FUL3): Responsible for Oracle supports workforce diversity. leading Technical Marketing Engineering teams in enlarging company’s market and increasing net revenue by marketing, supporting and promoting company’s technology to customers and other architectures as appropriate. RESEARCH TRIANGLE PARK, NC: IT Engineer (Ref.# RTP13): Responsible for development, support and implementation of major system functionality of TECHNICAL company’s proprietary networking products. RICHARDSON, TX: Customer Support Engineer (Ref.# RIC1): Responsible Oracle America, Inc. for providing technical support regarding the company’s proprietary systems and has openings for software. SAN JOSE/MILPITAS/SANTA CLARA, CA: Systems Engineer (Ref.# SJ13): Provide business-level guidance to the account team or operation on technology trends and competitive threats, both at a technical and business level. TECHNICAL Hardware Engineer (Ref. #SJ557): Participate on development of Application Specific Integrated Circuit (ASIC) for next generation data center switch product family, with emphasis on routing/switching protocols. Technical Lead/Leader (Ref.# SJ14): Lead engineering groups on projects to design, develop or test hardware or ANALYST software products. Software Engineer (Ref.# SJ10): Responsible for the definition, design, development, test, debugging, release, enhancement or maintenance of positions in Lehi, UT. networking software. Software/QA Engineer (Ref.# SJ11): Debug software products through the use of systematic tests to develop, apply, and maintain quality standards for company products. Technical Marketing Engineer (Ref.# SJ15): Responsible for enlarging company’s market and increasing revenue by marketing, supporting, Job duties include: Deliver solutions to and promoting company’s technology to customers. Technical Marketing Engineer the Oracle customer base while serving (Ref.# SJ178): Responsible for enlarging company’s market and increasing revenue by marketing, supporting, and promoting company’s technology to customers. as an advocate for customer needs. Travel may be required to various unanticipated locations throughout the United States. Software Engineer (Ref.# SJ957): Responsible for the definition, design, development, test, debugging, release, enhancement or maintenance of networking software. Telecommuting permitted. Principal Engineer (Ref.# SJ673): Lead the Apply by e-mailing resume to team for developing and enhancing routing and switching protocols and features [email protected], for company’s products. PLEASE MAIL RESUMES WITH REFERENCE NUMBER TO CISCO referencing 385.20025. SYSTEMS, INC., ATTN: G51G, 170 W. Tasman Drive, Mail Stop: SJC 5/1/4, San Jose, CA 95134. No phone calls please. Must be legally authorized to work in the U.S. without sponsorship. EOE. Oracle supports workforce diversity. www.cisco.com www.computer.org/computingedge 65 SOFTWARE SOFTWARE TECHNOLOGY Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for SOFTWARE SOFTWARE has openings for DEVELOPER DEVELOPER APPLICATIONS positions in Bedford, MA. positions in DEVELOPER Cambridge, Massachusetts. Job duties include: Oracle America, Inc. has positions in Irvine, CA. openings for Software Developer positions Job duties include: design, develop, trouble- in Bedford, MA. Design, develop, trouble- Job duties include: Analyze, design, shoot and/or test/QA software. As a shoot and/or test/QA software; as a member develop, troubleshoot and debug member of the software engineering of the software engineering division, apply division, apply knowledge of software knowledge of software architecture to software programs for commercial or architecture to perform tasks associated perform tasks associated with developing, end-user applications. with developing, debugging, or designing debugging, or designing software applica- software applications or operating systems tions or operating systems according to provided design specifications. Apply by e-mailing resume to according to provided design specifications. [email protected], Apply by e-mailing resume to Apply by e-mailing resume to referencing 385.21134. [email protected], [email protected], referencing 385.19249. referencing 385.19784. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

TECHNICAL TECHNICAL TECHNICAL Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for TECHNICAL TECHNICAL TECHNICAL ANALYST- ANALYST ANALYST SUPPORT positions in Lehi, UT. positions in Lehi, UT. positions in Lehi, UT. Job duties include: Deliver solutions to Job duties include: Deliver solutions the Oracle customer base while serving to the Oracle customer base while Job duties include: Deliver post-sales as an advocate for customer needs. support and solutions to the Oracle Offer strategic technical support to serving as an advocate for customer customer base while serving as an assure the highest level of customer needs. advocate for customer needs. satisfaction. Apply by e-mailing resume to Apply by e-mailing resume to Apply by e-mailing resume to [email protected], [email protected], [email protected], referencing 385.19989. referencing 385.19219. referencing 385.19586. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

66 ComputingEdge June 2017 TECHNOLOGY LinkedIn Corp. has openings in our Sunnyvale, CA location for: Software Engineer (All Levels/Types) (SWE0517SV) Design, develop & integrate cutting-edge software technologies; Products Operations Manager (6597.1532) Responsible for gathering key member/ customer insights & feedback for the Product team to influence the roadmap/prioritization of potential product changes; User Experience Designer (6597.1372) Design solutions that address business, brand & user requirements; Senior Site Reliability Engineer (6597.1957) Serve as primary point responsible for the overall health, performance, & capacity of one or more Internet-facing services; Associate User Experience Designer (6597.1487) Design solutions that address business, brand & user requirements; Senior Data Scientist (6597.96) Own & implement novel tracking and classification approaches to enhance measurement of site traffic, ensuring highest fidelity of executive KPIs; User Experience Designer (6597.1958) Create holistic design solutions that address business, brand & user requirements; Engineering Manager- Growth Engineering (6597.1886) Mentor, coach, develop & grow top engineering talent; Sr. Service Now Developer (6597.1860) Serve as domain expert & point of contact for Service Platform architectural decisions; Salesforce.com Developer (6597.1706) Develop, enhance, debug, support, analyze, maintain & test new/existing functionality which supports internal business units or supporting functions.

LinkedIn Corp. has openings in our San Francisco, CA location for: Software Engineer (All Levels/Types) (SWE0517SF) Design, develop & integrate cutting-edge software technologies; Manager, Software Engineering (6597.1438) Design, develop & integrate cutting-edge software technologies; Associate Technical Services Manager (6597.1757) Working under direct supervision, develop & maintain corporate solutions & products.

LinkedIn Corp. has openings in our New York, NY location for: Software Engineer (All Levels/Types) (SWE0517NY)

Please email resume to: [email protected]. Must ref. job code above when applying.

TECHNOLOGY Intuit Inc. has openings for the following positions in Mountain View, California: Senior Interaction Designers (Job code: I-2383): Exercise senior level knowledge to provide design direction for interaction concepts, flows and standards for both web and mobile. Staff Data Engineers (Job code: I-3166): Apply master level knowledge to design, develop, and implement data movement and integration processes in preparation for analysis, data warehousing, and operational data stores, involving very large quantities of data. Staff Business Data Analysts (Job code: I-3209): Partner with product managers, marketing, and engineering to enable decision support and key customer insights through web tracking metrics. Product Managers (Job code: I-301): Design and bring to market revenue generating, customer-driven software products and services that deliver great customer experiences. Senior Product Managers (Job code: I-435): Design and bring to market revenue generating, customer-driven software products and services that deliver great customer experiences. Communication Services Engineers (Job code: I-2576): Responsible for level 2 and 3 network operations that support and manage Intuit’s very large-scale and complex network data centers to support many of our applications. Senior Business Data Analysts (Job code: I-2915): Work alongside analysts, business stakeholders and data partners to deliver quality data for customer insights and recommendations. Sr. Technical Data Analysts (Job code: I-3226): Provide critical data extraction, segmentation, and analysis for direct response marketers and market researchers.

Positions in San Diego, California: Staff Software Engineers (Job code: I-369): Apply master level software engineering and industry best practices to design, implement, and support software products and services.

Positions in Plano, Texas: Senior Software Engineers in Quality (Job code: I-437): Exercise senior level knowledge to partner with cross-functional leaders and team members to deliver Intuit products, with greater efficiency and speed.

To apply, submit resume to Intuit Inc., Attn: Olivia Sawyer, J203-6, 2800 E. Commerce Center Place, Tucson, AZ 85706. You must include the job code on your resume/cover letter. Intuit supports workforce diversity.

www.computer.org/computingedge 67 CAREER OPPORTUNITIES

Apple Inc. has the following job opportunities in Cupertino, CA: Project Coordinator (Req# Software Development Engi- charctrisc of acc’ssries for Apple 9P7MJJ) Rspnsble fr ensurng the neer (Req# 9GUV5Q) Des & dev mobile devs (iPod, iPhone, iPad readiness of data collction sys Maps SW for iOS. & Watch). at fctories focsd on new prdct Software Development Engineer Software Development Engineer actvty. (Req# AF4VBC) Des & dev data (Req# A6TVMN) Des & imple- Software Development Engineer mining pipeline to automate all ment OS SW framework for me- (Req# A6K54T) Improve cell backend services. dia streaming. perform of iPhone & iPad prods Software Engineer Applications Software Development Engineer thru lab tests focused on stress- (Req# 9GZ2BN) Des & dev (Req# AC8U4F) Code in C pr- ing wireless sys. prdctivity app w a focus on srvr grmng lang & assmbly lang. Software Development Engineer side techs & client/srvr intrction. Software Engineer Systems (Req# 9UJ4L8) Des & dev web- Software Development Engineer (Req# A95U6N) Dsgn & dev bsd tools &infrastructure for (Req# AA8MRN) Debug & re- ISP firmware for Apple product dev tools & oper sys. solve issues w/ NVIDIA GPU cameras. Software Engineer Applications drivers on macOS. Software Development Engineer [Multiple Positions Open] (Req# Commodity Analyst (Req# (Req# A352G7) Rspnsble for 9ZTVVX) Standardize & sim- 9E4QR8) Source, dev, manu- testing synchronization feats on plify SW systems. Provide prod fact, & prod the company’s diff iOS & Windows throughout the supp for iCloud & CloudKit ser- & unique mech enclosure parts. Apple SW Dev process. vices & back-end apps. Software Engineer Applications Product Design Engineer (Req# Operations Engineering Pro- (Req# AF4RL2) Des & implmnt 9TDQR3) Des, dev & ex’cte gram Lead (Req# A7VVX2) ID contact cntr staffing tools & comp based sim’s models of & execute optimizations of the wrkfrce mngmnt soltns. mech syst’s. iPhone manufact process. Dev Ops Engineer (Req# AAS- Business Systems Analyst (Req# Software QA Engineer [Multiple SGG) Sprt & optmze Java run- AF6VTC) Prfrm bus analysis Positions] (Req# A5BTDE) Des time environment build, dlvry, for AppleCare org to des sys & & dev automation test SW & test & dplymnt pipeline. procdres, & analyze cmplx busi plan docs for testing Ad-Serving probs to sug enhncmnts w/in au- Software Development Engineer app. tomtd sys’s. (Req# A742NV) Test highly per- Software Engineer Applications formant, large scale sys using cli- Software Engineer Applications (Req# 9R6TN8) Dsgn & dvlp ent & server side test automation. (Req# A833EX) Des & dev ana- productivity app in iOS/ macOS lytics & telemetry SW. / Webapp. Software Development Engineer (Req# A4Z4CV) Des SW auto- Software Development Engineer Software Quality Assurance En- mation for iOS Sys SW updates (Req# AEW3XA) Build SW to gineer (Req# A8AUH4) Build (Python, Javascript). bring Maps data to users, help app test cases (auto & manual) dsgn APIs & services for new Information Systems Engineer & exec req’d tests. feats, & spport existing users. (Req# ABYVES) Des, dev, im- Software Development Engi- plmnt & maintain packaged en- Software Development Engineer neer (Req# A3D44U) Investigate terprise SW apps & custom built (Req# AB5VTX) Prototype adv bugs filed for location-based SW SW apps. apps & combine machine learn- service. ing & multi-modal user inter- Software Engineer Applications faces to create intelligent assist Software Development Engineer (Req# 9G6SCS) Des & dev data- srvcs. (Req# AD4TUH) Des & dev test stores & sys’tms. frameworks. Software Development Engineer Software Quality Assurance En- (Req# A7FQU6) Work on im- Software Quality Assurance En- gineer (Req# A7NMHZ2nd) proving the cell performance of gineer (Req# A7242W) Automate Eval bat’ry life & pwr perfmnc end-to-end web UI test case flows. iPhone & iPad prdcts.

68 ComputingEdge June 2017 CAREER OPPORTUNITIES

Software Engineer Applications (Req# AKFM6L) Gather bus & prdcts to prvide data-driven (Req# A5G38L) Res for des & reqs, provide GAP analysis & solution using advanced statisti- implmntation of apps build, re- complete code dsgn & dev for cal techniques & mach learning lease, deployment, configuration the MFi App & the integration to var bus problems. partners of the MFi App. activities, & tools &framework Hardware Development Engineer dev. (Req# A282AN) Dsgn camera Software QA Engineer (Req# mod structures & packaging in 9UKU3P) Test SW and algorithm Apple Inc. has CAD. Travel required 25%. for comptr vision. the following job Engineering Project Specialist Software Development Engineer (Req# 9ZK3GV) Resp for ops & (Req# 9WXS8S) Dsgn & dvlp opportunities in San supply chain for NPI. Travel req SW for mobile devices. Francisco, CA: 30%. Engineering Project Coordinator Engineering Project Lead (Req# Software Development Engineer (Req# 9SY33N) Define scope of A66MZ4) Coord’nte des, dev, (Req# A6MVMQ) Des & dev work, schedule, & budget for en- rsrch & implmnt of search & SW for Apple Ad Platform. gineering projects. web brwsing SW. Software Development Engi- Software Engineer Applications neer (Req# A4UUNE) Des, dev, (Req# A7Q5G7) Dsgn & dvlp test, & maintain sw for inter- cmplex, highly available, & scal- Apple Inc. has net ad sys. able SW sys. the following job Software Engineer Applications Software Development Engineer (Req# 9TDUXH) Des & im- (Req# AC9UC3) Research, dsgn, opportunities in plmnr SW to aid in analysis & dvlp, implmnt & debg statistical c’rrctn of client-fcn’g prods, us- & deterministic NLP SW as part Vancouver, WA: ing machine lrning tech, on a of novel txt input & txt prcssng Software Development Engineer global scale. sys. (Req# A7P3YH) Dsgn & dev SW Hardware Development Engineer Software Engineer Applications fr iBooks on iOS & macOS. (Req# A84USB) Dev advanced [Multiple Positions] (Req# AK- display calib method leading to KUS3) Des, dev, & maintain product implementation. complx front-end code for Apple Hardware Development Engi- Online Store. Refer to Req# & neer (Req# 9UJV7F) Dsgn elec Software Engineer Applica- mail resume to arch & sys for embedded sys & tions (Req# AE83ZL) Monitor Apple Inc., consumer electronics for Apple & oversee large scale distributed ATTN: D.W., products. systems based on NoSQL tech. 1 Infinite Loop 104-1GM, ASIC Design Engineer (Req# Systems Design Engineer (Req# Cupertino, CA 95014. 9TTUWC) Dvlp coverage mon- 9YX42T) Rspnsbl for RF Regu- Apple is an EOE/AA m/f/ itors & anlyze coverage to ensure latory qualification, eval, & de- disability/vets. all the test cases in the plans are bugging root cause analysis to covered. adhere to regulatory & industry Software Quality Assurance En- standard reqs. gineer (Req# A8KW24) Test new IST Technical Project Specialist Apple Inc. has iTunes Store prods & services. (Req# AGK5W7) Mntr day-to- the following job day prblm queue across multiple SAP prdctn envrnmnts to ensure opportunities in that highest priorities are being addressed, & take corrective ac- Cupertino, CA: tion when req’d. Machine Learning Engineer (Req# 9T5537) Dev data assets Software Engineer Applications Continued on the next page …

www.computer.org/computingedge 69 CAREER OPPORTUNITIES

… Continued from the previous page.

Hardware Engineering Man- ager (Req# 9LV5EW) Lead dsgn & dvlpmnt of advanced thin film optical coatings for Take the CS Library Apple display, sensor, & camera systms. wherever you go! IEEE Computer Society magazines and Transactions are now available to subscribers in the portable ePub format. Apple Inc. has Just download the articles from the IEEE Computer Society Digital Library, and you can read them on any device that supports the following job ePub. For more information, including a list of compatible devices, visit opportunities in www.computer.org/epub San Francisco, CA: Software Development En- gineer (Req# 9TNTNV) Dsgn, dev & implmnt sys & algrthms for text mining & machne lrng over large data sets. Software Development Engi- neer (Req#9F224R) Dsgn, dev TECHNICAL TECHNICAL & test SW to improve quality Oracle America, Inc. Oracle America, Inc. of data & French language results for Apple Search users. has openings for has openings for Native fluency in French req. TECHNICAL TECHNICAL Apple Inc. has ANALYSTS ANALYST the following job positions in Orlando, FL. positions in Orlando, FL. opportunities in Job duties include: Analyze user require- Austin, TX: ments to develop, implement, and/or Job duties include: Deliver solutions to ASIC Design Engineer (Req# support Oracle’s global infrastructure. As the Oracle customer base while serving 9PV2XQ) Model tasks req a member of the IT organization, assist as an advocate for customer needs. for the dvlpmnt & verifictn of Offer strategic technical support to GPU dsgn. with the design, development, modifications, debugging, and evaluation of assure the highest level of customer satisfaction. Refer to Req# & mail programs for use in internal systems resume & transcript to within a specific function area. Apple Inc., ATTN: D.W., Apply by e-mailing resume to 1 Infinite Loop 104- Apply by e-mailing resume to [email protected], 1GM, Cupertino, CA referencing 385.19329. 95014. [email protected], Apple is an EOE/AA referencing 385.19712. m/f/disability/vets. Oracle supports workforce diversity. Oracle supports workforce diversity.

70 ComputingEdge June 2017 SOFTWARE TECHNICAL INFO TECH Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for TECHNICAL SYSTEM SOFTWARE ANALYST ADMINISTRATOR-IT DEVELOPER positions in Colorado Springs, CO. positions in Burlington, MA. positions in Edison, NJ. Job duties include: Analyze user require- Job duties include: Install, configure, ments to develop, implement, and/or update, maintain, and support all aspects support Oracle's global infrastructure. As a of a customer facing production Job duties include: Design, develop, member of the IT organization, assist with implementation of software produced for troubleshoot and/or test/QA the design, development, modifications, a specific cloud industry and delivered in a debugging, and evaluation of programs for software. SaaS based Cloud delivery model. May use in internal systems within a specific telecommute from home. Apply by e-mailing resume to function area. [email protected], Apply by e-mailing resume to Apply by e-mailing resume to referencing 385.19883. [email protected], [email protected], referencing 385.19993. referencing 385.19667. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

TECHNOLOGY Help build the next generation of systems behind Facebook's products. Facebook, Inc. currently has the following openings in Menlo Park, CA (multiple openings/various levels):

Data Engineer, Analytics (8392J) Design and build data reporting and visualization needs for a product or a group of products. Data Scientist, Online Safety (9140J) Apply your expertise in quantitative analysis, data mining, and the presentation of data to see beyond the numbers and understand how our users interact with our core products. Program Solutions Lead (1069J) Build tools to help partners and internal teams perform better, automating workflows where they see an opportunity and ‘hacking’ solutions where needed. Software Engineer (6872J) Develop highly innovative, consumer-facing mobile products. Prototype new features and redesign existing features for mobile apps. Solutions Engineer (9261J) Combine technical and business skills to make our partners successful and improve Facebook platform. Product Manager (7219J) Engage in product design and technical development of new products. Application Engineer (9263J) Develop and maintain integrated, scalable, corporate applications. Build solutions using Oracle technologies. Data Scientist, Analytics (6318J) Apply your expertise in quantitative analysis, data mining, and the presentation of data to see beyond the numbers and understand how our users interact with our core products.

Mail resume to: Facebook, Inc. Attn: SB-GIM, 1 Hacker Way, Menlo Park, CA 94025. Must reference job title & job# shown above, when applying.

www.computer.org/computingedge 71 TECHNOLOGY TECHNICAL SOFTWARE Oracle America, Inc. Oracle America, Inc. has openings for Oracle America, Inc. has openings for has openings for PROGRAMMER TECHNICAL ANALYST ANALYST SOFTWARE positions in Lehi, Utah. positions in Redwood Shores, CA. DEVELOPER Job duties include: Deliver solutions to positions in Seattle, Washington. Job duties include: Analyze user require- the Oracle customer base while serving Job duties include: Design, develop, ments to develop, implement, and/or as an advocate for customer needs. troubleshoot and/or test/QA software. support Oracle's global infrastructure. May Offer strategic technical support to telecommute from home. assure the highest level of customer Apply by e-mailing resume to satisfaction. Apply by e-mailing resume to [email protected], referencing 385.20863. [email protected], Apply by e-mailing resume to referencing 385.19856. [email protected], Oracle supports workforce diversity. Oracle supports workforce diversity. referencing 385.19252. Oracle supports workforce diversity.

TECHNICAL QA ANALYST TECHNICAL Oracle America, Inc. Oracle America, Inc. Oracle America, Inc. has openings for has openings for has openings for TECHNICAL QA TECHNICAL ANALYST- ANALYST ANALYST positions in Lehi, Utah. SUPPORT positions in Bedford, MA. positions in Lehi, UT. Job duties include: deliver solutions to Job duties include: Develop and execute the Oracle customer base while serving software test plans. Develop, apply and Job duties include: Deliver post-sales as an advocate for customer needs; maintain quality standards for company offer strategic technical support to support and solutions to the Oracle products with adherence to both internal assure the highest level of customer customer base while serving as an and external standards. advocate for customer needs. satisfaction. Apply by e-mailing resume to Apply by e-mailing resume to [email protected], Apply by e-mailing resume to [email protected], referencing 385.18708. [email protected], referencing 385.19912. referencing 385.19892. Oracle supports workforce diversity. Oracle supports workforce diversity. Oracle supports workforce diversity.

72 ComputingEdge June 2017 CAREER OPPORTUNITIES

IEEE-CS Charles Babbage Award CALL FOR AWARD NOMINATIONS Deadline 15 October 2017

ABOUT THE IEEE-CS CHARLES BABBAGE AWARD Established in memory of Charles Babbage in recognition of significant contributions in the field of parallel computation. The candidate would have made an outstanding, innovative contribution or contributions to parallel computation. It is hoped, but not required, that the winner will have also contributed to the parallel computation community through teaching, mentoring, or community service. CRITERIA This award covers all aspects of parallel computing including computational aspects, novel applications, parallel algorithms, theory of parallel computation, parallel computing technologies, among others. AWARD & PRESENTATION A certificate and a $1,000 honorarium presented to a single recipient. The winner will be invited to present a paper and/or presentation at NOMINATION SITE the annual IEEE-CS International Parallel and Distributed Processing awards.computer.org Symposium (IPDPS 2017). AWARDS HOMEPAGE NOMINATION SUBMISSION www.computer.org/awards Open to all. Nominations are being accepted electronically at www. computer.org/web/awards/charles-babbage. Three endorsements are required. The award shall be presented to a single recipient. CONTACT US [email protected]

SOFTWARE TECHNICAL Reputation.com Oracle America, Inc. has openings for has openings for SENIOR TECHNICAL SOFTWARE ANALYST- ENGINEERS SUPPORT (Job ID#: 11076.27) positions in Lehi, UT. in Redwood City, CA. Job duties include: Deliver solutions to Design and develop products that allow web the Oracle customer base while serving applications to be highly available, clearly as an advocate for customer needs. understandable and customer centric. Apply by e-mailing resume to NEXT ISSUE To apply, send resume to: Reputation.com, INTERNET ATTN: Archana Yenuga, 1400A Seaport Blvd, [email protected], Suite 401, Redwood City, CA 94063. Must referencing 385.20304. OF THINGS reference Job ID#. Oracle supports workforce diversity. CALL FOR NOMINEES Education Awards Nominations

Taylor L. Booth Education Award Computer Science and Engineering Undergraduate Teaching Award A bronze medal and US$5,000 honorarium are awarded for an outstanding record in computer science and A plaque, certificate and a stipend of US$2,000 engineering education. The individual must meet two or is awarded to recognize outstanding contributions more of the following criteria in the computer science and to undergraduate education through both teaching engineering field: and service and for helping to maintain interest, increase the visibility of the society, and making a • Achieving recognition as a teacher of renown. statement about the importance with which we view • Writing an influential text. undergraduate education. • Leading, inspiring or providing significant education content during the creation of a curriculum in the field. • Inspiring others to a career in computer science and The award nomination requires a minimum of three engineering education. endorsements.

Two endorsements are required for an award nomination.

See the award information at: See the award details at: www.computer.org/web/awards/booth www.computer.org/web/awards/cse-undergrad-teaching

Deadline: 15 October 2017 Nomination Site: awards.computer.org