Open Science Developments in Japan and Its Challenges —Towards Sustainable OS Infrastructure

Open Science Developments in Japan and its challenges —Towards sustainable OS infrastructure

NATIONAL INSTITUTE OF INFORMATICS • MIHO FUNAMORI

International College Open Science Committee, France September 13, 2019 Today’s Talk 1. Open Science policy developments in Japan 2. Introduction to NII and its Services 3. Engaging the Academia in Japan with RDM 4. Impact of Plan S on Japan and non‐Plan S countries 5. PROPOSAL: Future of Open Access Publishing

1 Open Science policy １ developments in Japan OA policy developments for publication

3 How it started: “Serials Crisis”

 Journal subscription cost rising faster than the inflation speed  Four times higher in 2011 than 1986

Source: ARL Statistics 2010‐11 Association of Research Libraries, Washington, D.C. *Includes electronic resources from 1999‐2011. 4 http://www.arl.org/storage/documents/monograph-serial-costs.pdf The situation in Japan

 Japan’s Yen steadily strengthening! (1980-90’)  Cancelled almost all of the cost increase of subscription!  Introduction of “Big Deal” which increased the number of serials at universities. (2000’-)  Big Deal: Commercial publishers selling online subscriptions to large bundles of electronic journals at a discount price. ⇒ Almost no serials crisis. ⇒ Awareness for OA very low. 5 Journal price mostly affected by currency rate and consumption tax rise. $-YEN Consumption Exchange tax rise! Rate 100mil YEN Hopefully, 390 YEN is strong 190 at point of 370 contract! 170

350 150 UK ￡ 330 130 Euro € 310 110

US ＄ 90 290 70 270

E-Journal cost Subscription in Yen 50 250 2009 2010 2011 2012 2013 2014 2015 2016 2017

6 Japanese universities slowly feeling the pain of journal subscription cost

260 Publisher A260 Publisher B 260 Publisher C

250 250 250 253 240 247 240 240 244 243 241 238 230 230 230 234 232 230 220 220 220 225 224 225 210 210 210

200 200 200

190 190 190

180 180 180

170 170 175 174 174 170 170 170 160 160 160 160 150 150 150 2012 2013 2014 2015 2016 2017 2012 2013 2014 2015 2016 2017 2012 2013 2014 2015 2016 2017

Provided by Japan Alliance of University Library Consortia for E‐Resources : JUSTICE

7 Japan’s Governmental OA policies (1)

1. Deployment and enrichment of institutional repositories  In gradual deployment since 2003.  Stated in the 4th Science and Technology Basic Plan (FY2011-fy2015)!

2. JST adopting OA policy. (2013)  Allowing both green and gold OA. 3. A new funding scheme to publish an OA journal by JSPS. (2013)

8 Japan’s Governmental OA policies (2)

4. A platform to make Japanese scholarly journals OA available created by JST (J-STAGE)

5. OA mandate for PhD theses. (2013)

6. JSPS OA policy. (2017-) ⇒ Not distinguishing green or gold OA. ⇒ Enlarging OA contents is the main target!

9 Number of OA policies adopted

WORLD: - 1000 institutions

JAPAN - 27 institutions (as whole) - 15 institutions (on Roarmap)

(as of 2019.9) We have more than 700 universities! This is too small!

Source: Registry of Open Access Repository Mandates and Policies (ROARMAP) 10 http://roarmap.eprints.org/ Japan, the No. 1 country by the number of institutional repositories

900 808 800 754 52 700 808 IRs 681 598 600 ■ by JAIRO Cloud: Pilot Operation ■ by JAIRO Cloud: Production Operation 526 by University On‐premise System 498 504 500 ■ 396 431 288 210 400 357 130 73 300 260 228 193 200 144 316 310 101 284 301 285 228 260 256 252 100 58 193 144 101 2 10 58 0 2 10 2005.3 2006.3 2007.3 2008.3 2009.3 2010.3 2011.3 2012.3 2013.3 2014.3 2015.3 2016.3 2017.3 2018.3 2018.4

11 Japanese IRs contributing to preserving gray literature

Contents within IRs (total: 2.06 mil) Goal achieved 紀要論⽂学術雑誌論文学位論⽂研究報告書 1. Less research articles 一般雑誌記事 University データ・データベース → Not much contributing Bulletins テクニカルレポート会議発表論文 to green OA PhD thesis 図書 Research 会議発表用資料 Articles 教材 2. Many university プレプリントソフトウェア bulletin papers その他 NII_IRDBコンテンツ分析 2017.10 http://irdb.nii.ac.jp/analysis/index.php → Contributing to preservation and We need more circulation of gray research articles for green OA! literature

12 Japan at OA crossroads?  OA in Japan has been promoted by policymakers and the library community without the awareness of general academia.  Japanese OA policies rather in favor for green OA but not restricting.  However, with OA2020 and Plan S, the world seems to be shifting strongly to gold OA.  As the academia is starting to feel the pain, it is a good time to start discussion on OA.  However, the OA landscape is complex and most university administration don’t take the time to understand!

13 OS policy developments for research data

14 Policy Trends on RDM and Open Science in Japan

 June 2013: A joint statement by the G8 Science Ministers on making research data open  March 2015: Cabinet Office, "Promoting Open Science in Japan“  January 2016: "The 5th Science and Technology Basic Plan"  February 2016: Council for Science and Technology, "Promoting Open Access to Academic Information"  July 2016: Science Council of Japan (SCJ), "Recommendations Concerning an Approach to Open Science that Will Contributes to Open Innovation"  June 2018: Headquarters for Japan's Economic Revitalization, "Growth Strategy 2018 ‐ Reform towards Society 5.0 and Data‐driven Society"  June 2018: Cabinet Office, "Integrated Innovation Strategy"  “Data infrastructure for Open Science” as one of three keys to make Japan an innovative country.

15 Open Science Report from Japanese Cabinet Office (2015)

http://www8.cao.go.jp/cstp/sonota/openscience/150330_openscience_summary_en.pdf

16 Framework of the Open Science in Japan

http://www8.cao.go.jp/cstp/sonota/openscience/150330_openscience_summary_en.pdf 17 Data Management Plan (DMP) required by funding agencies

 Japan Science and Technology Data Agency (JST) Gene- ration  FY 2016-

Data Data  DMP after grant awarded Process- Reuse ing  Japan Agency for Medical Data Research and Development Lifecycle  FY 2017-  Asking about “data scientist” at Data Data Sharing Analysis grant proposal

Data  New Energy and Industrial Preser- vation Technology Development 大学図書館が、 Organization (NEDO) 研究者にDMPの作成方法を助言  FY 2017- しているよ！  Funds also industries 18 Guideline for drafting “Data Policy” at National Research and Development Agency

 Items to be defined in the data policy 1. Institutional Aim for drafting the policy 2. Definition and Conditions of research data 3. Preservation/Management/Implementation/ Security policies 4. Metadata, Persistent identifiers, 5. Ownership and Intellectual Property Rights 6. Open/Close, Embargo, Citation policies

（出典）内閣府「国立研究開発法人におけるデータポリシー策定のためのガイドライン」（2018） 19 http://www8.cao.go.jp/cstp/stsonota/datapolicy/datapolicy.html Society 5.0 …Vision in Basic Act on S&T (2016-20)

Society 1.0: Hunter-gatherer Society

Society 2.0: Agrarian Society

Society 3.0: Industry Society

Society 4.0: Information Society

Society 5.0

No name? - Data-linked society? - Data-intensive society?

Japan, “Realizing Society 5.0” 20 https://www.japan.go.jp/abenomics/_userdata/abenomics/pdf/society_5.0.pdf In reality, the major driving force for RDM is scientific misconduct prevention  MEXT: “Guideline for Dealing with Scientific Misconduct” (2014) 「研究活動における不正行為への対応等に関するガイドライン」  Strengthening the guideline in 2006.  Holds institutions to be responsible for research transparency and preventing scientific misconduct.  Science Council of Japan: “Reply: For the Enhancement of Soundness of Scientific Research” (2015) 日本学術会議「（回答）科学研究における健全性の向上について」  “Ten‐Years Preservation Rule for Research Data” 研究データ10年保存ルール

21 Japanese academic institutions having data preservation policies (FY2014 survey)

Does your institution have an data preservation policy?

The data policies tend to follow MEXT and SCJ NO YES Guidelines.

FY2014 Survey

 Public institutions tend to have a data policy rather than private institutions.  Institutions larger in size tend to have a data policy rather than small institutions.

22 Implementation of data preservation at Japanese universities  Introducing university‐wide “Research materials preservation policy.”  Cascading of responsibilities:  University holds departments,  Departments holds Labs,  Labs holds researchers responsible for data preservation.  No long‐term storage, no infrastructural support

23 Reporting of evidence‐data for research articles at Japanese universities  Research office sends out Excel spread sheet to researchers to have them report evidence‐ data.  Only single row to report,  No direct link to data  Data difficult to find.

Research Data Preservation List (研究データ保管管理簿) 保管管理者： Name

関連No. 発表テーマ・タイトル発表会議名等発表日保存期間データ破棄予定日データの保管場所等保存する研究データ等データ破棄日備考 No. Title of Research Article Conf.Name Date Prsv. Perio Data delete planned Storage place Preserving data Data deleted dat eOther 5年 5年 5年

Source: Kyushu Institute of Technology, “Policy for the Prevention of Research Misconduct” 2014 24 https://db.jimu.kyutech.ac.jp/cgi-bin/cbdb/db.cgi?Page=DBRecord&did=206&rid=236 Open Science in Japan still at its infancy  OS in Japan in mainly driven by policymakers and infrastructural work by NII.  The term OS is becoming familiar but most people do not understand what it means.  Strong emphasis on research data preservation to prevent scientific misconduct.  In this case, data does not need to be open.  Need to merge these two issues and direct RDM in Japan for positive purpose.

25 Introduction to NII 2 and its Services History of NII

• Pre-history as Research Center for Library and Information Science (RCLIS, 1976-) and Center for Bibliographic Information (1983) as centers within the University of Tokyo. • Founded in 1986 as National Center for Science Information Systems (NACSIS) • Reorganized in 2000 as National Institute of Informatics (NII) Tandem Organization of NII

• The National Institute of Informatics (NII) seeks to advance integrated research and development activities in information-related fields, including networking, software, and content. NII also promotes the creation of a state-of-the-art academic-information infrastructure.

2 21st Century Academic Information Infrastructure SINET5 for Advancing Open Science

Collaboration and Promotion in Research and Education

Resource Federation  Promotion of academic information GakuNin  Collaborative Federation circulation and open access enhancement of  Collaborative promotion of authentication between institutional repository expansion universities

Cloud Security Flow Analysis  Dramatic cost reduction and  Network flow analysis enhancement of research and and dynamic control education environment by GakuNin-Cloud  Raise of security level tailored cloud services Direct Connection for SINET users VPN

Network  Nationwide 100-Gbps backbone network and scalable network expansion  High-speed direct international lines to USA, Europe, and Asia  Introduction of new technologies such as SDN in response to user needs

29 NII is the Japanese NREN

• SINET is a Japanese academic backbone network for more than 800 universities and research institutions, and for about 3 million users. • SINET covers 100% of national, 78% of municipal, and 55% of private universities.

Inter-Univ. National Municipal Private Junior Colleges of Labs and Research Universities Universities Universities Colleges Technology Others Total Institutes Sapporo Number of 86 71 348 62 55 16 179 817 Organizations (100%) (78%) (55%) (18%) (97%) (100%)

(As of March 2015) : SINET node : Domestic line (100Gbps or more) To Europe

: International line (100Gbps) : International line (10Gbps)

Fukuoka

Osaka

To US Tokyo

To Asia 30 Scholarly Information Infrastructure

Journal articles Journal articles Catalog information Research Information

CiNii CiNii KAKEN Articles JAIRO Books

Catalog of materials held Project reports of Metadata and links of Metadata and links of MEXT Japanese journal Japanese institutional by universities articles Bibliographic info 11 M supported scientific 19 M records repositories records 2.5 M records researches Holding information 137 820 K record M records

Compilation

NACSIS-CAT

Shared Repositories Compilation Digitization Integration

JSPS MEXT 学協会 J-Stage NDL 学協会Academic (JST) Societies InstitutionalI More than repository Univ. 1,300 libraries Library More than 800 Linkage to other DB Universities and Research Institutions Note: The record institutions numbers are as of services March 2017

31 Japan, the No. 1 country by the number of institutional repositories

32 OA publishing platform J-Stage run by JST

33 NII Research Data Cloud

Discovery Platform DOI Metadata Management ● Linking Func between Article and Data International Metadata ● Researcher and Research Project Identification and Management Func Subject Aggregator Discovery Service Repository ● Data Exchange with International Discovery Service Metadata Aggregation User Flow Re‐use Search/Find Data Flow Data User Journal Supplemental Article Research Data Mng User Interface Data

Access Control Metadata Mng Data Depositor Institutional Research Data Mng Exp/Store Archive Research Data Management System Research Data Repository

RDM Platform Exp Data Article Publication Platform Private Shared Public by High Speed Access using SINET5 ● Hot Hot Hot Data oriented Self‐Archiving Func Data Sharing Func using Storage Storage Storage ● ● Versioning and auto‐Packaging Func Virtual NW and ID Federation Cold Cold Cold ● Effective Data Storage Switcher Storage Storage Storage ● User Dependent Personal Data ● Pseudonym Func Storage Area for Long‐term Preservation 34 Publication Platform

900 Number of Repository in Japan • Current System WEKO2 800 75 700 808 IRs 600 NII Repository • Journal Article Repository 500 477 Cloud Service 288 396 400 210 130 • Add Functions more and more 300 On‐Premise 73 200 301 316 310 228 260 284 285 267 ✖ 100 144 193

Number of Repositories 101 Research Data Handling 0 2 10 58

• New System WEKO3 Year • Based on Invenio3 which is originally focused as Data Repository • Integrate WEKO2 Functions into Invenio3

WEKO3 Realize New Publication Platform based on sophisticated Invenio3 Architecture Data Repository （Invenio3 = our RDM Platform in Architecture） Effective Development and Operation

Strengthen Domain Use‐case by Extensibility Conventional Functions

35 Discovery Platform

• NII Knowledge Graph • Aggregate from various DBs • Define Entity Links • User Interface • Support Discovery Experience for Research Activity Itself

Research Project Researcher Coverage of Current CiNii Researcher ID,ORCID... Conventional Research Output Project ID DOI, Handle, URI, ISBN, ISSN...

Book Paper Dissertation Funding Agency Research Crossref Funder, GRID, ISNI... Activity Research Data DOI, URI... Research Institution Institutional ID, GRID, ISNI...

Exchange Data with International Discovery Services 36 RDM Platform New Service

Extension of Open Science Framework developed by COS, USA

Manage Research Data Connect Cloud Storage Share Research Data within by Research Project from Various Plugin Collaborators Authn by ID Fed

NII: Frontend Service RDM Platform

University: Backend Storage Cloud Storage NII Storage

Customize selectable Plugin depending on Default (minimum?) University Environment and Policy Storage by NII

Public Cloud Private Cloud Public Cloud (Provider DC) (On‐premise) (Provider DC) 37 NII, the leading voice in digital transformation in higher education Japan

Government

ICT Univ. Centers NII Libraries

NII in strong collaboration with all ICT centers and university‐libraries across Japan

38 Engaging the 3 Academia in Japan with RDM Necessity to engage the Japanese academia into Open Science 1. The idea of OA and OS not well understood. 2. RDM in an academic institute involves multi‐ stakeholder approach.  Meaning, nobody takes leadership to start OS.  Hiring RDM manager does not work if there is no RDM policy justifying his/her work. 3. Need to direct RDM at Japanese universities to positive direction.  Implementing RDM for the sake of scientific misconduct prevention does not make researchers happy.

40 Multi‐stakeholder Approach I want to make needed to implement RDM at universitiesthe university research

Administ Univ-wide policies & strategies competitive! ration

R Research Library CIO Integrity VP Professional President Assoc. Service Units Service Univ-wide Research Research ICT Data Support Unit Univ. Library Protection Admin Office （URA Station） Center R admin R evaluation D preservation E-infrastructure D publishing IT policies R integrity R support Data Curation Dept. admin offices: coordination Departments,

Labs Learnt Societies

ポリシー策定 Grad students, technicians, lab manager, etc.: data generation, RDM 専門的助言41

41 Why an RDM Charter?  Participants at AXIES‐RDM session started to claim,  “We need a charter in order to convince the university administration and to get the researchers and staff engaged!”  AXIES • Academic eXchange for Information Environment and Strategy • Community of CIOs and ICT centers of universities in Japan. • Counterpart to EDUCAUSE in the US

42 “RDM Charter for Academic Institutions”

 RDM Charter Don’t dare to tell me how to  Not for researchers, but manage my data! I know what  For academic institutions! I’m doing!

Researcher  Purpose of RDM Charter  Give university administration ideas and options to implement RDM in respective institutions.

43 RDM implementation in an academic institute

Univ. Library Decision-making on RDM implementation Running Inst. Repoistories Research Strength President Adding metadata Info. Research Strategy Acad. Inf. DMP support Discovery RDM Training etc. Drafting of research strategy Service Research Evaluation Finding research collaborators INST. REPOS. STORAGE Open/Closed/Embargo Closed R Data Info Search ・・・・・・DMP Info Research ・・・・ ---- Publishing func. Access control func. ---- R Integrity (10 yrs preserv) Administration ---- Storing func. Version control func. Preservation func. Office Publish Grant Mgmt (DMP) ・・・・ ---- ・・・・・・------・・・・ ------Lab

Researcher

ICT center Grad students, technicians, lab manager, etc. Provision of e-infrastructures

44 “RDM Recommendation for Academic Institutions”…Main Text 1. Need for RDM at academic institutions  Research data used at every research process in every discipline  Researcher who manages research data controls research competitiveness in the digital age 2. Infrastructure for RDM should be provided by institution  Inefficient if every researcher cares for his/her own environment 3. Efficiency that the RDM infrastructures are common across institutions and AXIES role 4. Expectation that this recommendation will be referred at academic institutions

45 “RDM Recommendation for Academic Institutions”…Composition  The Recommendation  Addresses the viewpoints why academic institutions needs to take RDM seriously.  Viewpoints in bullet points: I. Role of academic institutions in RDM II. Policies and organizations for RDM needed in acad. Institutions III. RDM procedures in acad. Institutions IV. RDM Purpose options in acad. Institutions V. Digital platform functions needed for RDM in acad. Institutions VI. Human resources development for RDM in acad. Institutions VII. Reuse and service options of research data in acad. institutions  Appendix  Glossary  References

46 Calling universities Next Steps of AXIES for collaboration! GOAL Make sure that RDM platform is not just provided as system but make it work within institution! Activities Designing questionnaire survey template Establishing RDM case studies Developing RDM policies and guidelines

47 Japan Consortium for Open Access Repository (JPCOAR)

 Launched in July 2016  580 member institutions as of November 2018  https://jpcoar.repo.nii.ac.jp/

Mission of JPCOAR 1. Reforming scholarly communication system by promoting Open Access 2. Operating JAIRO Cloud jointly with NII 3. Enriching IRs with a wide range of scholarly materials 4. Providing training courses 5. Contributing to international collaboration

48 JPCOAR Organization

Task Forces

Research Researcher Mid‐term Metadata SCPJ Data ID Plan Schema

Steering Committee General Assembly report

Strategy Making and Steering Chairperson Auditor appoint Working Groups

JAIRO Cloud Human Resources Outreach Operation Development

49 JPCOAR Research Data TF

• RDM Training Course 1. Review and Analyze existing Training Materials Essentials 4 Data Support, MANTRA, RDM Rose, RDM Support, Research Data Management and Sharing (COURSERA) 2. Define Syllabus 3. Develop Slides, Scripts and then Video Production 4. Provide from JPCOAR Site (Slides) and Japanese MOOC (Course)

JPCOAR SITE Access DL 2017.06 936 903 2017.07 286 342 2017.08 393 318 2017.09 429 308 2017.10 362 238 2017.11 515 301 Participants Completion Rate 2017.12 355 113 RDM Course 2,305 25% 2018.01 313 131 Total 3,589 2,654 JMOOC Ave. 4,145 15%

Now Developing Next Version, especially focusing on RDM Supporters 50 Impact of Plan S on Japan and 4 non‐Plan S countries 11 European research funders demand immediate OA…the Plan S of cOAlition S

 Declaring that publicly-funded research outputs from respective funders must be published OA immediately after 2020.  Articles can only be published on compliant OA journals or platforms. Hybrid journals are explicitly excluded.  Aiming to transform hybrid and subscription journals to OA journals. Prestigious high-quality journals to be  Supporting funders eliminated?!  Austria, France, Ireland, Italy, Luxembourg, Netherland, Norway, Poland, Slovenia, Sweden, UK  Remaining 18 European funders also expected to participate

Source: cOAlition S (2018.9.4) 52 https://www.scienceeurope.org/coalition-s/ Plan S compliant publication roads OAJ desired

A) OA journal B) Subscription journal – non OA  Without embargo period C) Hybrid journal  The journals must agree on transformative agreement and must become full OA journal within three years

（出典）cOAlition S, “Principles and Implementation” (2019.5.30) 53 https://www.coalition-s.org/principles-and-implementation/ Proportion of journals by OA status

The hybrid journals are problematic with double dipping charges for APC and subscription.

Source: Nature, “Radical open-access plan could spell end to journal subscriptions” (2018.9.4) 54 https://www.nature.com/articles/d41586-018-06178-7 Assessing Method

 Thought experiment conducted for each ten principles of revised Plan S on the its impact on non-Plan S countries.  Impact considered for funders, researchers, academic institutions, publishers, learned societies, etc.  This thought experiment is an update of the following work.  Miho Funamori, “Thought Experiment on the Impact of Plan S on non-Plan S countries and Japan,” 2019 IIAI International Conference on Advanced Applied Informatics (IIAIAAI)

55 The Plan S Principles OA publication fees are covered by the 04 Funders or research institutions, not by individual researchers

Impact on non-Plan S countries: BAD  Non-Plan S researchers have to fund the APC by themselves.  As number of OA journals requiring APCs will rise with Plan S, this will result in:  Less annual publication from non-Plan S countries  Less visibility and reduced research competitiveness in non-Plan S countries 56 Source: Plan S, “Principles and Implementation,” (2019.5) https://www.coalition-s.org/principles-and-implementation/ How many articles can you publish with average APC US$1000-2000?

$1500-2000

$3000 $2000-3000 $1000-1500 -5000 APC無し $1000-2000 $750-1000 APC無し $500-1000 $500 -750

APC無し $5000- $1500- $3000 -5000 APC無し $1000-1500 $1000-2000

$2000－3000 $300-500 $500-800

Source: Open Science, “How Much Do Top Publishers Charge for Open Access?” (2017.4.20) 57 https://openscience.com/how-much-do-top-publishers-charge-for-open-access/ The annual research budget of a Japanese researcher (FY2015)

Researchers with annual Research budget less than US$4000

-$900 $900-$2700 $2700-$4500 ALL $4500-$9000 $9000-$18000 $18000-$27000 $27000- Other (Unknown)

 If more than half of Japanese researchers rely on less than $4000 （50万円） for annual research budget, the strong reliance on APCs $1000-2000 can be damaging for number of research outputs.

Source: 第8期研究費部会（第8回）配付資料「個人研究費等の実態に関するアンケート」について（調査結果の概要）」 http://www.mext.go.jp/b_menu/shingi/gijyutu/gijyutu4/037/shiryo/__icsFiles/afieldfile/2016/08/16/1375827_04.pdf 58 Compact for Open-Access Publishing Equity（COPE） ―University subsidizes APCs

COPE compatible COPE Signatories OA funds Cornell University Brandeis Univ Dartmouth College Carnegie Mellon Univ Harvard University Colorado State Univ MIT ETH Zurich UC Berkeley George Mason Univ University of Ottawa Indiana Univ-Purdue Univ Indianapolis Columbia University Johns Hopkins Univ Memorial Sloan-Kettering Cancer Center Lund Univ Universitat de Barcelona Northern Illinois Univ Duke University Southern Illinois Univ Carbondale University of Calgary Tufts Univ Simon Fraser University Univ of Bielefeld CERN Univ of California, Davis Karlsruhe Institute of Technology Univ of California, Irvine University of Utah Univ of California, Merced No signatories University of Pittsburgh Univ of California, San Diego renewed University of Tennessee Univ of California, San Francisco Texas A&M University Univ of California, Santa Barbara after 2014 Emory University Univ of California, Santa Cruz University of Rhode Island Univ of Colorado Univ of Florida Univ of Illinois at Chicago Univ of Iowa Univ of Kansas Univ of Manitoba Univ of Minnesota Univ of North Carolina at Chapel Hill Univ of North Carolina at Charlotte Univ of Oklahoma Source: Compact for Open-Access Publishing Equity Univ of Oregon 59 http://www.oacompact.org/ Univ of Tromsø Univ of Wisconsin - Madison OA block grant by UKRI Japan publishes ca 16,000 articles annually.  OA block grant provided by UKRI So, an estimate of 2.8 to universities, and redistributed billion yen is spent for OA publishing. to researchers by university. Since 30% are OA, Japan will need 9 billion yen for such OA block grant provided in FY 2016/17 block grant. Total Full OA Hybrid JUSTICE estimate Top 10 UK universities F.Y.2016 by OA block grant provided Block grant ￡ provided 14M UCL 2.4億円 (￡1.63M) Cambridge U 1.9億円 (￡1.27M) Number of APC- 億円 (￡1.08M) 10,000 2500 7500 Manchester U 1.6 funded articles Oxford U 1.3億円 (￡0.91M) Edinburgh U 1.3億円 (￡9.88M) Average APC ￡1988 ￡1654 ￡2101 Sheffield U 0.9億円 (￡0.62M) Glasgow U 0.9億円 (￡0.61M) 億円￡ Total APCs ￡4M * ￡16M * Warwick U 0.7 ( 0.50M) ￡18M Leeds U 0.7億円 (￡0.48M) provided （* Total spent） Bristol U 0.7億円 (￡0.48M)

Source: RCUK Open Access Block Grant analysis August 2013-July 2017 Source: UKRI, “2016-2017 block grant awards” https://www.ukri.org/files/funding/oa/rcuk-apc-returns-analysis-2016-and-2017-pdf/ https://www.ukri.org/files/legacy/oadocs/open-access-block-grants-2016-17-pdf/60 How to establish funding

schema for APCs Offset agreement!

1. Turning subscription budget to APCs  For many universities, the subscription budget is bigger than the total APCs spent. Thus, this is feasible.  However, for transitional period, some additional budget for APCs may be needed.  Also, for big research-intensive universities which produce many articles, turning subscription to APCs is not enough. 2. Acquiring grants from funding agencies  Funders could provide grants based on research grants allocated or number of published articles.  Funders could also support bigger research universities which need excess money for the many research publications.

61 Yomiuri Shimbun（2019.2.13） “APCs one of reasons of research strength decline in Japan”

Juichi YAMAGIWA  President, Kyoto University (2014-)  President, Science Council of Japan (2017-)  Former President, Japan Association of National Universities (2017-19)

62 The Plan S Principles Funders may contribute to financially 08 supporting transformative arrangements of hybrid journals.

Impact on non-Plan S countries: BAD? GOOD?

 Publishers might be reluctant to make transformative agreement with non-Plan S countries.  Might affect the visibility of non-Plan S publications during transformation.  However, if hybrid journals become OA journals, subscription is no longer needed.  Non-Plan S will need country-level coordination to make transformative agreement. 63 Source: Plan S, “Principles and Implementation,” (2019.5) https://www.coalition-s.org/principles-and-implementation/  Research-intensive Publish & Read Agreement university will pay more. towards transition to full OA  Teaching-intensive university will pay more.

Pay-walled Articles Full OA

APCs paid

Open Access Rate by author Double Dipping OA

Subscription-based Publish & Read APC-based Contract at Institution-level Contract Agreement Contract Transformative agreement 64 towards full OA Registry for transformative agreement

Plan S recommended

Source: ESAC, “Transformative Agreement, Agreement Registry“ 65 https://esac-initiative.org/ Publishers positive for transformative agreement (TfA)

 Some publishers are positive for TfA  Strategy to lock in authors by transforming to OAJ at early stage?  Wiley, Springer-Nature, Cambridge Univ. Press, Oxford Univ. Press, RSC, ACS, SAGE, de Gruyter, Thieme, IWA Publishing, Karger etc. Please publish with us!  TfA mostly at country-level  Country： Austria, Germany, Netherland, Norway, Switzerland, Hungary, Greek, Slovenia, Spain  Institution：Max Planck, Delft University of Technology, California Digital Library, Iowa State University

66 JUSTICE OA2020 Roadmap

• Data Analysis（ subscription fees, publication output, APC expenditure） • Pilot towards OA publishing model (TfA) • Flipping the subscription cost & grant • Consensus & Cooperation（Collaboration with univ. assoc., boycott publishers? ） • Supplementary Actions （出典）JUSTICE「購読モデルから OA 出版モデルへの転換をめざして～JUSTICE の OA2020 ロードマップ～」（2019.3.5） https://www.nii.ac.jp/content/justice/overview/JUSTICE_OA2020roadmap-JP.pdf 67 Should we proceed with transformative agreement?

Usually Publisher small publishers

＆ Elsevier Negative NO! With TfA

Number of Published articles

Positive Small NO! With TfA number Wait until journal becomes full OA

Publisher eager to Large lock in authors number YES! Cost cut because TfA is

cost-neutral 68 Various reactions towards Plan S…Funding agencies (2)

 European countries increases  11->21 agencies  China joins  Big player!  Zambia, Jordan joins  Plan S deal for developing countires  US  Doesn’t join!

69 大学間の連携が Coordination among 求められている！ universities needed

 Transformative agreement (TfA) is cost- neutral  If TfA is made only at institutional level, publishers will go bankrupt  Institutions with small number of published articles don’t go into contract and become freerider  Thus, TfA should be made at country-level.  The Japanese consortium (JUSTICE) is not a legal entity, and each institution makes its own contract ⇒ Need univ. assoc. to coordinate the universities 70 The Plan S Principles

The Funders commit that when assessing 10 research outputs during funding decisions they will value the intrinsic merit of the work

Impact on non-Plan S countries: GOOD

 Since the impact factor and other journal metrics were in favor of western countries, non-Plan S countries might be rated better with DORA principles.

71 Source: Plan S, “Principles and Implementation,” (2019.5) https://www.coalition-s.org/principles-and-implementation/ OA gains more citations!

OpCit project carries many related information

Source: Times Higher Education, “Open access papers ‘gain more traffic and citations’” (2014.6.30) http://www.mext.go.jp/b_menu/shingi/gijyutu/gijyutu4/037/shiryo/__icsFiles/afieldfile/2016/08/16/1375827_04.pdf 72 APC-IF-OA-research budget considered to determine publication venue in life sciences

Which journal is a bargain?

Let’s invite someone as co-author who has funds for APC

（出典）第3回 SPARC Japan セミナー2018 「オープンアクセスへのロードマップ: The Road to OA2020」大隅典子（東北大学副学長 / 附属図書館長/ 医学部・医学系研究科教授）「日本におけるOAの推進を阻むもの：一（いち）生命科学者より」講演資料より 73 Factors affecting publication venue in OA age

 Journal impact factor?  Whether journal is OA

The faster article is OA, The faster you get  Amount of APC user comments!  Swiftness of publication  Swiftness of user comments ※ Publishing first on a preprint server proves to be most effective!

74 Summary of the thought experiment: Impact of Plan S on non-Plan S countries

 Non-Plan S countries are indirectly affected by Plan S by being forced to adapt to the world standard that Plan S sets forth.  As many non-Plan S countries lack support for this transition from their respective funding agencies, they will be seriously disadvantaged to adapt to the new standards.  Cost transparency and APC capping  Machine-readable JATS standards  Transformative agreement  Countries from Global South or non-English speaking countries might get isolated from these moves; however, might also be advantaged to get access to

scholarly contents. 75 PROPOSAL: Future of 5 Open Access Publishing The transformation of acad. journals by state of OA

2016 Subscription journal Hybrid journal OAJ

Jap. article OA rate 30％ Plan S takes effect ＆ TfA period ends $ 30 mil APC

2024 Subscription journal OA journal

Jap. article OA rate 60％

50 yrs OA journal after

Jap. article OA rate 100％ APC 100% flowing out of country 77 How many acad. contents, i.e. APCs, can we keep in Japan?

If articles are published Let’s use If certain portion of contents 100% in overseas OAJs Jap. OA can be published in Japan platforms！ APC-free OA platforms OA journals with APCs APC 100% OA journals flowing out of country with APCs

78 Next Generation Repositories proposed by COAR

Contents does not need to be on the same platform as the publishing arm.

COAR, “Next Generation Repositories” https://www.coar-repositories.org/activities/advocacy-79 leadership/working-group-next-generation-repositories/ Proposal) EU provides the e-infrastructure where articles can be submitted, peer- reviewed, published, made available OA.

European Open Access Platform

Government supporting the E-infra, excluding publishers https://blogs.openaire.eu/?p=1961 80 http://blogs.lse.ac.uk/impactofsocialsciences/2017/04/10/rather-than-simply-moving-from-paying-to-read-to-paying-to-publish-its-time-for-a-european-open-access-platform/ Could we use Episciences on our institutional repository cloud service JAIRO-cloud?

Episciences.org marries in an innovative way the two avenues of free access: the golden path through the hosting of open access journals (epi- journals) and the green way since the process of submitting articles is done by filing in an archive opened

81 JAIRO Cloud Episciences https://www.episciences.org/ In combination with NII Research Data Cloud, this will enable future publishing also with research data

Access Control Metadata Mng Data Depositor Institutional Research Data Mng Exp/Store Archive Research Data Management System Research Data Repository

Call drafted based on the belief that OA publishing should NOT be pursued by single solution, namely the gold OA way which requires APCs.

83 https://jussieucall.org/jussieu-call/ San Francisco Declaration on Research Assessment (DORA)

 The need to eliminate the use of journal-based metrics;  The need to assess research on its own merits;  The need to capitalize on the opportunities provided by online publication Source: San Francisco Declaration on Research Assessment (DORA) 84 https://sfdora.org/ Challenges for “Overlay journals with distributed institutional repositories ”  Academics, especially the editorial of journals, tend to favor shiny toys provided by commercial journals.  Need to raise more awareness on the problem that arises by using commercial publishers.  Need to work on business models that works for academic societies.  Need to change the workflow of researchers.  To publish its work first on institutional repository or preprint server.  Need the collaboration of librarians.  Checking the uploaded work and metadata.