Versions of academic papers : current practice and attitudes of economics researchers towards creating and storing digital versions Frances Shipsey, VERSIONS.

Versions of academic papers : current practice and attitudes of economics

researchers towards creating and storing digital versions

Frances Shipsey, VERSIONS Project, London School of Economics and Political Science

2nd International Digital Curation Conference, 21-22 November 2006, Glasgow

21 November 2006 / 2

Versions and digital curation• What has the question of version control or version identification to

do with digital repositories and digital curation?• How many versions are authors needing to manage?• Have authors kept their papers?

– If not, what are the reasons for loss?• If they have kept their papers, can they find them again when needed (eg for

deposit in a repository)?• What are authors’ views about responsibility for the long term storage of their

research outputs?• How do authors deal with citing papers in the face of change, in a discipline

with a strong pre-print tradition?

• The VERSIONS Project asked authors about these issues• Survey in summer 2006 – 464 responses from researchers, (75% of which

were economics researchers)

21 November 2006 / 3

Pre-print culture in economics – wide dissemination through different channels

57

408363

273

153

66

19 6 1

0

50

100

150

200

250

300

350

400

450

1 only

2 or more

3 or more

4 or more

5 or more

6 or more

7 or more

8 or more

9 or more

VERSIONS survey of researchers Q4. Number of different types of research output produced in a typical research project

21 November 2006 / 4

Which versions do researchers keep?

Which revisions do you personally keep / plan to keep stored in electronic form (eg on your computer or network drive)

36%

54%

8%

0%

2%

Keep all revisions

Keep major revisions but not all

Keep the latest revision that Iworked on only

Do not keep a personal copy

Other (please specify)

VERSIONS survey of researchers Q5. Thinking about revisions you make to your research outputs during their preparation, which revisions do you personally keep / plan to keep stored in electronic form (eg on your computer or network drive) at the end of the process?

21 November 2006 / 5

Permanent storage by authors of multiple versions of their journal articles

VERSIONS survey of researchers Q7: ‘Which of the following versions of a paper, that you have written for publication in a refereed journal, would you personally keep (eg on your own computer or network drive)?’

Revision stage Percentage of respondents who keep this stage permanently

Number of respondents who keep this stage permanently

Early draft version(s) before circulating to anyone, other than co-authors

39.9% 185

Draft version circulated to colleagues or peers for feedback before submission

53.9% 250

Version submitted to a journal for peer review 78.9% 366

Final author version produced by yourself/co-authors – agreed with the journal following referee comments

90.7% 421

Proof copy (publisher-produced version) 62.5% 290

Final published version (publisher-produced PDF) 91.8% 426

21 November 2006 / 6

Easily accessible final author versions

0

50

100

150

200

250

300

Easily accessible final author versions

All

Most

Some

None

Don't know

Don't produce

VERSIONS survey of researchers Q8. Thinking about storing your academic papers in the long term and focussing on 'final author versions' of your papers, do you have an easily accessible copy of these among your personal files (electronic or paper)?

58% have all, and 36% have most of these versions easily accessible

21 November 2006 / 7

Reasons for not having easy access to own final author versions

Reason Number of respondents

% of total respondents to this question (156) who gave this reason

% of all survey respondents

I do not have electronic copies before a certain date 90 57.7% 19.4%

I do not have copies produced while I was at a previous university/institution

19 12.2% 4.1%

I do not have copies of papers that I co-authored, the principal / lead investigator has this version

26 16.7% 5.6%

Changes to the manuscript are made iteratively between myself and the publisher in the later stages so I would have to assemble such a version

24 15.4% 5.2%

Loss or damage to my computer 29 18.6% 6.3%

Papers are stored electronically but would be difficult to retrieve from various servers

30 19.2% 6.5%

Loss or damage to paper files 11 7.1% 2.4%

I have discarded print copies of older papers before a certain date and do not have electronic versions

17 10.9% 3.7%

Other 19 12.2% 4.1%

21 November 2006 / 8

Responsibility for secure storage of different versions

Authors/ Co-authors

Authors' universities (including libraries)

Publishers Subject repositories

Early draft version(s) before circulation

181 13 9 14

Draft version circulated for feedback

242 40 6 47

Submitted version 340 46 116 47

Final accepted version

349 81 168 62

Proof (publisher's) 156 47 243 33

Published version 239 305 434 257

None of these 18 102 7 35

Don't know 9 16 9 135

21 November 2006 / 9

Citing versions in the face of change

Q24. If you read an earlier version of a paper that has been published in a journal, how do you prefer to cite it?

339

33

58

22

12

0 50 100 150 200 250 300 350 400

Cite the published version only

Cite both the published version and theearlier author version that I have read

Cite the earlier author version that I haveread

Do not cite any version of the paper if I havenot read the published version

Don't know

21 November 2006 / 10

Free text comments on citing earlier versions

• A large number of free text comments were made which clarified respondents views on a range of issues, eg: • The wish to cite a reliable version that will remain accessible• The need to take the time to check published version before citing it

to make sure the relevant paragraph was not cut or that the argument was not substantially changed

• Citation of both versions if differences in the content merit this• Actively seeking to provide a citation to an open access copy• Actively seeking to cite published journal article version out of

courtesy to the cited author

21 November 2006 / 11

Other projects and initiatives on versions

Ongoing standards development work - • NISO/ALPSP Working Group on Versions of Journal

Articles

Two JISC activities during 2006 - • RIVER – Scoping Study on Repository Version

Identification• JISC Eprints Application Profile Working Group

21 November 2006 / 12

NISO/ALPSP Working Group on Versions of Journal Articles

• Publisher-led group, with larger review group made up of publishers, librarians and other stakeholders

• Detailed use cases developed on the workflow from submission to publication and beyond

• Draft documents including Terms and Definitions for versions (March 2006)

– Author’s Original– Accepted Manuscript– Proof– Version of Record– Updated Version of Record

http://www.niso.org/committees/Journal_versioning/JournalVer_comm.html

21 November 2006 / 13

RIVER – Scoping Study on Repository Version Identification (RIVER)

• Led by Rightscom Ltd with partners London School of Economics and Political Science Library, University of Oxford Computing Services

• Report to JISC Scholarly Communications Group, March 2006 at: http://www.jisc.ac.uk/uploaded_documents/RIVER%20Final%20Report.pdf

• Recommendations to JISC for further work:• More detailed survey into development plans for repositories and awareness

of versioning• Research definitive sets of version identification requirements• Produce a more robust set of taxonomies from tentative and draft versions• Develop framework policies for use by institutions and for interoperability

21 November 2006 / 14

RIVER

• Defined two broad classes of requirement for version identification:• Collocation• Disambiguation

– ‘Identifying that two digital objects which happen to share certain attributes […] have no contextually meaningful relationship’

– ‘Understanding the meaning of the relationship between two digital objects where one exists [without inspecting and comparing the objects themselves]’

• Defined a tentative typology of ‘versions’ covering both time-based version relationships and others

21 November 2006 / 15

JISC Eprints Application Profile Working Group

• Carried out within JISC Digital Repositories Programme• Approach based on FRBR (work – expression –

manifestation – item) and the DCMI Abstract Model• Provides more detail and structure than simple Dublin

Core• Deals with versions very well• Work carried out June-August 2006 with follow up to

take place through a DC Task Group• http://www.ukoln.ac.uk/repositories/digirep/index/Eprints_Application_Profile

21 November 2006 / 16

To finish … an illustration

• Version identification is not a new problem, nor a problem limited to curation of digital objects only• Higher criticism - focuses on the sources of a document and tries to

determine the authorship, date and place of composition of the text http://en.wikipedia.org/wiki/Higher_criticism

• Textual criticism – concerned with the identification and removal of errors from texts http://en.wikipedia.org/wiki/Textual_criticism

• Ulysses family tree – illustrates both of these

21 November 2006 / 17Reproduced by permission of the National Library of Ireland and X Communications

21 November 2006 / 18

Looking forward – some issues

• Inform academic authors about the need for personal information management strategies – keep everything vs keeping milestone versions– Idea of granularity reduction – could a tool be developed that

could support authors– (Kjetil Nørväg (2006), Granularity reduction in temporal document databases,

Information systems 31 (2), 134-147. DOI: 10.1016/j.is.2004.10.002) • Institutional policies on digital preservation to consider which

individuals or institutions (if any) should take responsibility for long term storage of multiple versions

• The wish to cite multiple versions of a work may increase - – FRBR-isation of repositories and search services could help– International Standard Text Code available as a Draft

International Standard (DIS 21047) if this progresses further it could be a very interesting way to maintain relationships between versions into the future http://www.collectionscanada.ca/iso/tc46sc9/wg3.htm

21 November 2006 / 19

The VERSIONS Project

• VERSIONS : Versions of Eprints – user Requirements Study and Investigation of the Need for Standards• www.lse.ac.uk/versions• [email protected]

• Funder: Joint Information Systems Committee• Partners: London School of Economics and Political Science (lead)

and Nereus (associate partner)• Survey of researchers conducted in summer 2006• Project runs until February 2007

• Thank you