DOCSLIB.ORG

Usenix Notes 55

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Usenix Notes 55 USENIX BOARD OF DIRECTORS Years and Years Ago Communicate directly with the USENIX Board of Directors by Peter H. Salus writing to [email protected]. [email protected] 2005! PRESIDENT Think about it! USENIX Michael B. Jones, Fifty years ago, 1955, IBM encour- [email protected] aged the first meeting of “users”— notes operators of the new 704—I guess, the first LISA. Also in 1955, VICE PRESIDENT USENIX MEMBER BENEFITS AT&T/Western Electric submitted Members of the USENIX Associa- Clem Cole, to the consent decree that, among tion receive the following benefits: [email protected] other things, barred them from entering any business other than FREE SUBSCRIPTION to ;login:, the Associ- telephony or telegraphy (when was ation’s magazine, published six times SECRETARY the last time you sent or received a a year, featuring technical articles, Alva Couch, system administration articles, tips telegram?). [email protected] and techniques, practical columns on A decade later, 1965. The Multics such topics as security, Perl, Java, and project has gotten underway. CTSS operating systems, book reviews, and TREASURER and DTS on the DEC-10 are the big summaries of sessions at USENIX thing. IBM is just coming out with conferences. Theodore Ts’o, [email protected] the 360. ACCESS TO ;LOGIN: online from October 1975! Big time. The Labs have 1997 to this month: www.usenix.org/publications/login/. DIRECTORS come out with V6. The “UNIX Users’ Group” (linear parent of ACCESS TO PAPERS from USENIX confer- Matt Blaze, USENIX) has met (nearly two ences online: [email protected] dozen in attendance!) and is brac- www.usenix.org/publications/ Jon “maddog” Hall, ing itself for another—publi- library/proceedings/ [email protected] cized—meeting to be held 18 June. THE RIGHT TO VOTE on matters affecting Geoff Halprin, (Attendance bounded to 40.) It the Association, its bylaws, and elec- [email protected] was the beginning of multiple tion of its directors and officers. meetings, too. Marshall Kirk McKusick, DISCOUNTS on registration fees for all [email protected] In October, Mel Ferentz chaired a USENIX conferences. meeting at CUNY, and Belton Allen DISCOUNTS on the purchase of proceed- chaired one four days later at the ings and CD-ROMs from USENIX EXECUTIVE DIRECTOR NPG in Monterey. conferences. Ellie Young, The year 1976 saw three meetings, SPECIAL DISCOUNTS on a variety of prod- [email protected] too: one in Berkeley in February ucts, books, software, and periodi- chaired by Bob Fabry; two at Har- cals. For details, see vard—April and October—both www.usenix.org/membership chaired by Lew Law. /specialdisc.html. The second of these was the first FOR MORE INFORMATION regarding meeting to top 100 attendees, but membership or benefits, please see the next May, Steve Holmgren www.usenix.org/membership/ or contact offi[email protected]. chaired the first Midwest meeting Phone: 510-528-8649 at UIUC and eclipsed that with 250 in attendance. In September, Oliver Whitby and John Bass ran a West Coast meet- ing at SRI with about 100 atten- dees. 54 ;LO GIN: V OL. 30, NO. 2 The organization was about to Thirtieth Anniversary, USENIX held a POSIX workshop become USENIX. The publication in 1987. USENIX Association was about to become ;login:. And: An informal gathering of under Peter H. Salus 1st Graphics Workshop, 1985 two dozen people in May 1974 [email protected] had turned into a semiannual 1st C++ Workshop, 1987 event of major proportions. June 18, 1975. CUNY in Manhat- 1st Supercomputing Conference, [Blatant advertisement: Tom tan. Mel Ferentz runs the first 1987 Limoncelli and I have edited an USENIX conference. Of course, it 1st Security Workshop, 1988 anthology of all the April Fool’s wasn’t called USENIX then, it was a UNIX users’ group, until the 1st Mobile Computing Workshop, Day RFCs. No Starch Press, out in 1993 July.] lawyers at AT&T got tough about that (tm). And it wasn’t the first 1st OSDI, 1996 meeting, either, as Lou Katz had 1st Electronic Commerce Work- Short Topics Booklets run a small meeting in a confer- shop, 1998 ence room at Columbia in May 1st Embedded Systems Workshop, Rik Farrow, Short Topics Editor 1974. 1999 [email protected] But there were “about 40 people SAGE became a Special Technical The Short Topics in System from 20 institutions” at the 1975 Group of USENIX in 1992. Administration series of booklets meeting. are intended to fill a void in the Look around at any USENIX con- USENIX has brought together the current information structure, pre- ference, workshop, symposium. core of the Linux Kernel develop- senting topics in a thorough, refer- There’ll be many times 40 folks. ment team in the Linux Kernel eed fashion but staying small and Yes, it has been 30 years, but the Developers Summit, held annually flexible enough to grow with the growth has come because USENIX since 2001. community. has been where it’s happening. USENIX is where Ken Thompson Number 12 in the series, Building And still is. spoke in 1974; where Steve Jobs a Logging Infrastructure, by Abe USENIX is where Kirk McKusick spoke in 1987; where Stu Feldman Singer and Tina Bird, appeared talked about memory manage- lectured us on architecture; where first at LISA ’04. As of March 1, a ment. we learned how Google works. new booklet, A Sysadmins’ Guide to USENIX is where Tom Ferrin told Oh, yeah. And how to fix your Oracle, by Ben Rockwood, was us how to “cut this foil etch” and PDP-11 with this 98-cent resistor. beginning a technical review, and “insert this jumper wire.” In 1966 BU (Before UNIX), it should be in production early Crispian St. Peters sang, “Follow this summer. A contract has gone USENIX is where we first heard me, I’m the Pied Piper . .” It made out to Xev Gittler and William about Tcl and OAK (= Java) and it to #4 on Billboard’s list. Charles for a booklet with the Perl and GNOME. working title Being Root, providing USENIX is where, in 1980 in But if you follow USENIX, you’ll guidelines for working as the root Boulder, Colorado, Jim Ellis really know where it’s at. user, focusing on best practices, announced USENET. tools, and ethics. USENIX is where UUNET began. USENIX welcomes suggestions for USENIX is where portability has topics, proposals for booklets, and been supported for 30 years. technical reviewers. Interested? Please send email to USENIX has been sponsoring [email protected]. redistributable software since 1976. For the list of booklets, see www.usenix.org/publications/. USENIX held its first security workshop in 1988. ;LOGIN: APRI L 2005 USENIX NOTES 55.
Load more

Recommended publications
  • The Design of the Openbsd Cryptographic Framework
    The Design of the OpenBSD Cryptographic Framework Angelos D. Keromytis Jason L. Wright Theo de Raadt Columbia University OpenBSD Project OpenBSD Project [email protected] [email protected] [email protected] Abstract the design of these systems is intended to impede sim- ple, brute-force, computational attacks. This complexity drives the belief that strong security is fundamentally in- Cryptographic transformations are a fundamental build- imical to good performance. ing block in many security applications and protocols. To improve performance, several vendors market hard- This belief has led to the common predilection to avoid ware accelerator cards. However, until now no operating cryptography in favor of performance [22]. However, system provided a mechanism that allowed both uniform the foundation for this belief is often software imple- and efficient use of this new type of resource. mentation [8] of algorithms intended for efficient hard- ware implementation. To address this issue, vendors We present the OpenBSD Cryptographic Framework have been marketing hardware cryptographic acceler- (OCF), a service virtualization layer implemented in- ators that implement several cryptographic algorithms side the kernel, that provides uniform access to accel- used by security protocols and applications. However, erator functionality by hiding card-specific details be- modern operating systems lack the necessary support hind a carefully-designed API. We evaluate the impact to provide efficient access to such functionality to ap- of the OCF in a variety of benchmarks, measuring over- plications and the operating system itself through a all system performance, application throughput and la- uniform API that abstracts away device details. As tency, and aggregate throughput when multiple applica- a result, accelerators are often used directly through tions make use of it.
    [Show full text]
  • Usenetdht: a Low-Overhead Design for Usenet
    UsenetDHT: A low-overhead design for Usenet Emil Sit, Robert Morris, and M. Frans Kaashoek MIT CSAIL Abstract traction of Usenet is the incredible diversity and volume of content that is available. Usenet is a popular distributed messaging and file sharing ser- vice: servers in Usenet flood articles over an overlay network to Usenet is highly popular and continues to grow: one fully replicate articles across all servers. However, replication Usenet provider sees upwards of 40,000 readers reading at of Usenet’s full content requires that each server pay the cost of an aggregate 20 Gbit/s [35]. Several properties contribute receiving (and storing) over 1 Tbyte/day. This paper presents the to Usenet’s popularity. Because Usenet’s design [1, 19] design and implementation of UsenetDHT, a Usenet system that aims to replicate all articles to all interested servers, any allows a set of cooperating sites to keep a shared, distributed Usenet user can publish highly popular content without copy of Usenet articles. UsenetDHT consists of client-facing the need to personally provide a server and bandwidth. Usenet NNTP front-ends and a distributed hash table (DHT) that Usenet’s maturity also means that advanced user inter- provides shared storage of articles across the wide area. This faces exist, optimized for reading threaded discussions or design allows participating sites to partition the storage burden, streamlining bulk downloads. However, providing Usenet rather than replicating all Usenet articles at all sites. service can be expensive: users post over 1 Tbyte/day of UsenetDHT requires a DHT that maintains durability despite transient and permanent failures, and provides high storage per- new content that must be replicated and stored.
    [Show full text]
  • Interview with Mary Ann Horton RIK FARROWPROGRAMMING
    Interview with Mary Ann Horton RIK FARROWPROGRAMMING Mary Ann Horton has been a met Mary Ann Horton at USENIX ATC ’19 in Seattle. I didn’t know who UNIX developer and sysadmin she was, but somehow discovered that she worked on the control systems since 1977. She contributed to for the grid in the San Diego area, and we exchanged email addresses so Berkeley UNIX, creating the I we could continue the conversation. Later, I read her Wikipedia page [1] and first email attachments and enhancing vi. Her PhD dissertation at Berkeley learned much more about her. led to IDE editors that check your program for Rik Farrow: You have been working with UNIX since its earliest days. errors. While at Bell Labs, she led the UUCP Mapping Project and brought .com domains Mary Ann Horton: I fell in love with UNIX earning my master’s degree at Wisconsin in to UUCP email. She led the growth of Usenet, 1977, but my big break came in 1978 when I transferred to Berkeley for my PhD. We got a an early social media network, in the early VAX, initially with VMS, but quickly changed to UNIX 32/V. There were many amazing grad 1980s. Her EMS email system allowed email students contributing tools to BSD, and it was a treat to get to be part of this effort. It seemed addressing by database query. As a transgender like about half the code was written by Bill Joy, including vi. I got to enhance vi, nurture activist in the 1990s, she convinced Lucent it, and port it to all sorts of UNIX clones.
    [Show full text]
  • Proceedings of the 12Th USENIX Security Symposium
    USENIX Association Proceedings of the 12th USENIX Security Symposium Washington, D.C., USA August 4–8, 2003 THE ADVANCED COMPUTING SYSTEMS ASSOCIATION © 2003 by The USENIX Association All Rights Reserved For more information about the USENIX Association: Phone: 1 510 528 8649 FAX: 1 510 548 5738 Email: [email protected] WWW: http://www.usenix.org Rights to individual papers remain with the author or the author's employer. Permission is granted for noncommercial reproduction of the work for educational or research purposes. This copyright notice must be included in the reproduced paper. USENIX acknowledges all trademarks herein. Preventing Privilege Escalation Niels Provos Markus Friedl Peter Honeyman CITI, University of Michigan GeNUA CITI, University of Michigan Abstract gain extra privilege after successful authentication lim- its the degree of escalation because the user is already Many operating system services require special priv- authorized to hold some privilege. On the other hand, ilege to execute their tasks. A programming error in a a remote adversary gaining superuser privilege with no privileged service opens the door to system compromise authentication presents a greater degree of escalation. in the form of unauthorized acquisition of privileges. In For services that are part of the critical Internet the worst case, a remote attacker may obtain superuser infrastructure is it particularly important to protect privileges. In this paper, we discuss the methodology against programming errors. Sometimes these services and design of privilege separation, a generic approach need to retain special privilege throughout their life- that lets parts of an application run with different levels time. For example, in SSH, the SSH daemon needs to of privilege.
    [Show full text]
  • USENIX Board of Directors Interview with Clem Cole ;Login
    NOTES USENIX Board of Directors ;login: Enters a New Phase Interview with Clem Communicate directly with the USENIX of Its Evolution Cole Board of Directors by writing to board@ Cat Allman, Rik Farrow, Casey Henderson Rik Farrow Arvind Krishnamurthy, and Laura Nolan usenix.org. Clem Cole is an old school hacker For over 20 years, ;login: has been a print and “Open Sourcerer” with more PRESIDENT magazine with a digital version; in the two than 45 years of free and open source system Amy Rich, Redox decades previous, it was USENIX’s news- development experience. Clem has held practically [email protected] letter, UNIX News. Since its inception 45 every position in the computer field from operator, VICE PRESIDENT years ago, it has served as a medium through programmer, and designer to VP of Engineering, Arvind Krishnamurthy, University which the USENIX community learns about CTO, and startup founder. He first encountered of Washington useful tools, research, and events from one the early editions of UNIX in the 1970s while at [email protected] another. Beginning in 2021, ;login: will Carnegie Mellon University, later doing his graduate SECRETARY no longer be the formally published print work at the University of California, Berkeley. He Kurt Andersen, LinkedIn magazine as we’ve known it most recently, has been designing and developing operating [email protected] but rather reimagined as a digital publica- systems and technical computing systems ever tion with increased opportunities for inter- since, currently leading an international team of TREASURER activity among authors and readers. engineers. He helped to write one of the original Kurt Opsahl, Electronic Frontier Foundation TCP/IP implementations in the late 1970s, and is [email protected] Since USENIX became an open access pub- known as one the authors of the precursor to IM, lisher of papers in 2008, ;login: has remained DIRECTORS the UNIX talk program, as well as other more Cat Allman, Google our only content behind a membership humorous and notorious hacks.
    [Show full text]
  • User Groups, Conferences, and Workshops PETERHISTORY H
    User Groups, Conferences, and Workshops PETERHISTORY H. SALUS Peter H. Salus is the author of ser groups go back to the beginning of “mass”-produced computers A Quarter Century of UNIX in 1953–54. In this column, I sketch that history, starting with (1994), Casting the Net (1995), mainframe groups that shared code and tips, up to the founding (and and The Daemon, the Gnu and U the Penguin (2008). naming) of USENIX and its conferences and workshops. [email protected] The first commercial computer, the IBM 701, wasn’t completed until late in 1952. The first production machine was shipped from Poughkeepsie to IBM headquarters in Manhattan that December. Prior to the 701, all computers—Aiken’s, Wilkes’, ENIAC, etc.—had been one-offs; each was sui generis. The 701 was a genuine breakthrough, and IBM built 18 of them! (By way of comparison, Apple announced that it had sold a total of 19 million Macs over fiscal 2014.) On May 7, 1954, the redesigned 701 was announced as the IBM 704. It was more than merely a redesign. The 704 was incompatible with the 701. It had 4096 words of magnetic-core memory. It had three index registers. It employed the full, 36-bit word (as opposed to the 701’s 18-bit words). It had floating-point arithmetic. It could perform 40,000 instructions per second. While deliveries began in late 1955, the operators (today we would think of them as system admin- istrators) of the eighteen 701s were already fretful months earlier. IBM itself had no solution to the problem.
    [Show full text]
  • Volume 14, Number 6 December 1993
    ISSN 1035-7521 AUUG Inc. Newsletter Volume 14, Number 6 December 1993 Registered by Australia Post, Publication Number NBG6524 The AUUG Incorporated Newsletter Volume 14 Number 6 December 1993 CONTENTS AUUG General Information . , . ¯ . 3 Editorial ......... AUUG Institutional Members . ¯ AUUG President’s Page ..... ¯ . 10 Minutes of the 1993 AGM .... ¯ . ll AUUG articles for the Australian . ¯ Financial Statement ............. .... ¯ 14 How did AUUG spend the monies at AUUG’93? Liz Fraumann ¯ . 22 Opportunity knocks softly sometimes ............... ¯ . 25 Conference Announcements Berkeley 4.3 - 4.4BSD Workshops ................. 26 AUUG Summer Conference Series ................. 30 Fifth Annual Canberra Conference and Workshops .............32 Annual AUUG-NSW Summer Conference ............... 33 ¯Northern q;emtory Chapter .................... 35 AUUG Queensland Chapter .................... 38 South Australian Summer Conference ................ 41 Victo~ Summer Technical Conference .............. ’ ¯ 43 46 Perth Summer Technical Conference ................. Tasmanian Summer Conference .................. 49 LIPDATE - USENIX Conference .................. 52 UniForum NZ’94 - Conference Countdown ............... AUUG Local Chapters AUUG Inc. - Victorian Chapter .................. 55 Update on AUUG Inc. - Victorian Chapter Activities" Stephen Prince ..... 56 From the Western Front Janet Jackson ..... 57 WAUG - Meeting Reviews .................... 59 Open System Publications ...................... 60 ACSnet Survey ........................
    [Show full text]
  • Visualizing Web Sessions: Improving Web Browser History by a Better
    ii Visualizing Web Sessions: Improving Web Browser History by a Better Understanding of Web Page Revisitation and a New Session- and Task-Based, Visual Web History Approach Matthias Mayer 2007 University of Hamburg ii Universität Hamburg Department Informatik University of Hamburg Department of Informatics Dissertation zur Erlangung des Grades des Doktors der Naturwissenschaften, eingereicht am Department Informatik der Universität Hamburg. Dissertation in the Department of Informatics submitted to the faculty in partial fulfillment of the requirements for the degree of »Doctor rerum naturalium« at the University of Hamburg. Gutachter | Reviewers: Prof. Dr. Benjamin B. Bederson, University of Maryland, USA Prof. Dr. Horst Oberquelle, Universität Hamburg, Deutschland Prof. Dr. Christopher Habel, Universität Hamburg, Deutschland Tag der Disputation | day of disputation: 12. Dezember 2007 iii Widmung | Dedication Meiner Mutter Esther, meinem Vater Karl, meiner Schwester Gabriele. In großer Dankbarkeit. I am grateful to my mother, Esther, my father, Karl, my sister, Gabriele. iv Zusammenfassung Bei der Benutzung des World Wide Web spielen verschiedene Arten des Wiederbe- suchs von Webseiten (Revisit) eine wesentliche Rolle. Kurzzeit-Revisits entstehen häufig beim Explorieren von Websites und von Suchergebnis-Listen. Mittelfristig entstehen Revisits vorwiegend durch die Wiederbenutzung bevorzugter Werkzeuge und durch das Beobachten von Seiten, deren Inhalt sich häufig ändert. Langzeit- Revisits entstehen vor allem durch das erneute Lesen spezifischen Inhalts nach eini- ger Zeit. Die heute verfügbaren History-Werkzeuge wie der Zurück-Knopf (Back Button), Lesezeichen (Bookmarks), Verlauf (History List) und die automatische Vervollständigung einer Adresse (Auto-Complete) bilden ein heterogenes Umfeld unterschiedlichster Ansätze, die alle ihre eigenen Methoden des Zugriffs auf die persönliche Web-History bieten und mit den verschiedensten Benutzungsschnitt- stellen ausgestattet sind.
    [Show full text]
  • Cryptographic Device Support for Freebsd Samuel J
    Cryptographic Device Support for FreeBSD Samuel J. Leffler Errno Consulting [email protected] ABSTRACT FreeBSD recently adopted the OpenBSD Cryptographic Framework [Keromytis et al, 2003]. In doing so it was necessary to convert the core framework to function correctly in a fully-preemp- tive/multiprocessor operating system environment. In addition several issues with the basic design were found to cause significant performance loss. After addressing these issues we found that FreeBSD outperformed OpenBSD on identical hardware by as much as 100% in tests that exercise only the cryptographic framework. These optimizations result in similar performance improvements for facilities like IPsec that make heavy use of the cryptographic framework. We observed that FreeBSD’s Fast IPsec [Leffler, 2003] typically outperforms OpenBSD’s IPsec implementation [Miltchev et al, 2002] by more than 50% on identical hardware. We conclude that the OCF cryptographic API can be optimized and re-tuned to deliver substan- tially better performance than the original OCF implementation with large gains in both through- put and latency. Moreover these changes can be made with no impact on clients of the crypto- graphic framework: both user and kernel sofware designed for the original OCF is easily ported to the FreeBSD implementation of OCF. 1. Background and Introduction • The OCF was designed for a uniprocessor system Cryptographic transformations are an important com- without kernel preemption. The FreeBSD 5.0 ponent of security applications and protocols. operating system has fine-grained locking and the Because these operations are computationally expen- kernel is fully preemptive. This required a rewrite sive vendors have dev eloped products that accelerate of the core crypto functionality.
    [Show full text]
  • Unix: a View from the Field As We Played the Game
    111 Témoignage / Testimony Unix: A View from the Field as We Played the Game Clement T. Cole Intel Corporation. UNIX is a classic example of a the economics in a virtuous circle as this “Christensen Disruptive Technology.” It was community developed into a truly global one. a cost-effective solution, produced at the I will trace a little of the history of a small right time, built by researchers at AT&T for newsletter to today’s USENIX Association Abstract themselves, and was not originally considered and some of its wider social impact.. seriously by its competition. The UNIX Operating System had simple goals. It ran Keywords: UNIX; history of UNIX; on modest hardware, and was freely shared operating systems; open system; disruptive as a result of AT&T legal requirements. As innovation. a result, a new computing customer developed, a different one than was being targeted by the large firms of the day. UNIX was targeted at the academically-inclined; it was economically accessible, and since its Intellectual Property (IP) was published in the open literature and implementation was available to the academic community fundamentally without restriction, the IP was thus “free” and able to be examined/ discussed/manipulated/abused by the target users. While its creators wrote UNIX for themselves, because they freely shared it with the wider community, that sharing fed on A Brief Personal History The point is that when UNIX was originally written, the purchasers of com- In the mid-1970s I was a student puting equipment were not primarily made at Carnegie Mellon University (CMU), up of the actual programmers of the com- studying Electrical Engineering and puting system.
    [Show full text]