Open access: next steps

Robert Terry

Senior Policy Adviser

The Wellcome Trust

r.terry@wellcome.ac.uk

Funded by the Wellcome Trust

Why open access matters to us...

Why should open access publication be important to research funders?• Just funding the research is a job only part done – a

fundamental part of their mission is to ensure the widest possible dissemination and unrestricted access to that research.

• It’s all about improving access – improving research

• Web developments have created a new publishing model - not fully realised whilst access mediated through subscriptions and bundle deals. 90% of NHS-funded research available online full text 30% immediately available to public

Only 40% immediately available to NHS staff

Submission to the House of Commons Science and Technology Committee's Inquiry into Scientific Publications “How accessible is NHS-funded research to the general public and to the NHS's own researchers? Matthew Cockerill Ph.D., Technical Director, BioMed Central Ltd. http://www.biomedcentral.com/openaccess/inquiry/refersubmission.pdf

Shouldn’t those who pay for the research be able to read it?

•Over 90% of research funded in UK universities is public money from government, research councils and charities (17%)*

*Investing in Innovation - A Strategy for science, engineering and technology – July 2002 DTI

http://www.wellcome.ac.uk/publications

Why don’t researchersknow or care?

Free

Publishers

LibrariesResearchers

Shareholders & Societies

Gov / ngo funding

£ Profit

Free

£

£ £

Funders mission?

No money for peer review or to

author

Two routes to open access – both need to be supported

•publish in an open access format

•publish in any journal but deposit a copy in an open access repository

Publish in an open access format (journal?)

•Immediate global availability •Increased usage (citation) improved efficiency •Affordable •Archived for long term storage and access•Long term solution

•Few titles but growing •Not as well known – impact but growing•Resistance to the author pays model from authors (passive),

learned societies, commercial sector.

Will OA publishing cost more?

Cost element Proportion of

costs

1. Refereeing 22%

2. Editorial and typesetting (i.e. from

acceptance to first copy)

33%

3. Subscription management 7%

4. Physical production and distribution

(including postage)

23%

5. Sales and marketing 13%

6. Promotion to authors 2%

Total 100%

Whole system savings of 30% ?

Estimated costs per article:

$2,750 subscription

$1,950 open access

Submission fee $175 publication drops to $550

Charges of $10,000++ include contribution of funds to overheads, surplus or profit

Journals with> 30 papers1995 - 1999*

Commercial33%

Society43%

University Press24%

Elsevier 10%Portland Press 5%CUP 5%Blackwell 4% OUP 4% Nature 3%

Total Trust papers n=16,646 in 1292 journals*Source: ROD

What will it cost funders? Trust estimates: 1 – 2% of research budget

Publish in any journal but deposit a copy in an open access repository (e.g. UK PMC)

Institution and/ or Central (subject based) – both valid but offer users different solutions

What do funders want:

•Immediate access - but can work with 6 months delay to allow market to adapt.

•Long-term digital archive – accurate, future-proof preservation

•Searchable – ‘under one roof’ subject based

•Build on existing research practice e.g. Medline

•Funders attribution, additional features e.g public engagement

•Links with other databases e.g. genes, proteins

•Strategy, evaluation and impact

NLM Master Database

NLMBackend Processing

Journal Setup(Manual Process)

- attributes- scheduling

PMC XMLWEB Images (Jpeg, Gif)

PDFsScanned Pages (TIFFs)

Supplemental Data

Inter-Article Links- XML

pPMC CollectorDistribution Files (zip)- article source files- journal setup files

pPMC Loader

Portable PMC DatabaseField Index

Search Pages and Results

NLM PMC

Portable PMC (pPMC)

TOC

Front Pages

pPMC Render

Article

Deta

il

Search Web ServiceSource: David Lipman, Director, National Centre for Biotechnology Information, NLM, USA

Portable PMC mirror

How will UK PMC work

Source: David Lipman, Director, National Centre for Biotechnology Information, NLM, USA

UKPMC – quality, consistency, adaptability

There are three types of errors that PubMed Central deal with:

1. Structural Errors do not conform to the ruleset (DTD) that they were written for e.g. XML tags are wrong: <surname>Jones</snm>

2. Content Errors formula, tables, paragraphs, special characters (Greek characters or symbols) are not correct.

3. Consistency Errors tagged in one style suddenly switches e.g. For the first 5 years of content, Journal X has been tagging dates like:<date>10-12-2004</date> (m-d-y)

Then, this date appears in content:<date>14-12-2004</date> (this must be d-m-y)

Source: David Lipman, Director, National Centre for Biotechnology Information, NLM, USA

Link to imaging agent in PubChem through MeSH

Source: David Lipman, Director, National Centre for Biotechnology Information, NLM, USA

Links between sequence and related proteins

Funder initiatives• Leadership - demonstrate engagement with issues, join with other

research funders, raise awareness in research community

• Fund - cost of publication (marginal to research costs)

• Copyright - encourage author retention use of Creative Commons

• Repository - establish open access repositories and self-archiving UKPMC

• Evaluation - recognise intrinsic value of content of paper rather than title of journal

• Digitization - of existing titles

Dissemination costs are research costs

Greater accessibility = greater impact of research

The future?•More of the same? – unlikely

Increased use of repositories and self-archiving – likely: NIH policy still under discussion Scottish Science Information Strategy Working Group (http://scurl.ac.uk/WG/SSISWGOA/declaration.htm)

Further debate in HoC Scientific Publications: Free for all?

•More support from funders? - very likely

e.g. funders group for UKPMC, Howard Hughes, Max Planck, CNRS,

NIH.

•Research Councils UK – under discussion…….

Once deposition working, add a peer review

element - questions the concept of the journal?