Gerry


Support the principle .. Not sure about timescale but a stick does need to
be applied to get things happening ... And a carrot such as the data
available free TR API would help.
Also it would be good to encourage academics to record funding
acknowledgements consistently using the guidance developed by RIN, for
example at 
http://www.rin.ac.uk/our-work/research-funding-policy-and-guidance/acknowle
dgement-funders-journal-articles [Ideally we should have a list of codes
for the main funders - maintained by some central body - BIS Research Base
Funders Forum ? - as names and spellings can vary]

In terms of process, the funder acknowledgement is provided by the author
to the publisher -> WoS [Scopus] -> IR or CRIS -> RCUK
If what you suggests works - will it mean we don't also have to supply the
data via Research Outcomes, eVal, ROD - this would be another carrot :)
although would mean you only get publications rather than all the other
stuff requested in ROP.

Some detailed comments ...

1. We have a library subscription to WoS [and are also trialling Scopus at
the moment too]. We also have a subscription to TR INCites - mainly so
that we can import as much metadata as possible into our PURE CRIS [the
relevant metadata then being punted through to our IR if/when full-text is
available].  As you know, Pure allows users to create links between
various entities including between publications and projects [and
therefore funders and grant numbers]. These links can be made manually via
the interface and also, we are planning, reverse-engineered from WoS data
- assuming it is of good enough quality.

2. 
a.) Great if TR include the data in their web services 'lite' and Elsevier
are encouraged to do likewise. Although they[Elsevier] have told us they
will be including funding info next year they have not yet indicated
whether it will included in the Scopus API that comes with the standard
subscription.
b) OR facilitate this via Pure [as already happens] or other third-party
or in house CRIS  
c) data quality could be an issue given that sometimes
difficult/time-consuming to identify records belonging to an Institution
due to name variants on author affiliations. We have certainly spent
considerable time working with TR on this - in general we have found the
best approach is to encourage individual researchers to identify their own
records via the WoS API and bring them into Pure .. We retain the WoS UT
unique identifier so can update the records eg with citation counts.
d) OR in PURE or other CRIS ... This is our golden source as described
above.
e) Include Pure [and other CRIS that can support this] in your list of
'repositories' - it can expose dc data - but shouldn't we be thinking of
using CERIF rather than, or as an alternative to. arguing where it should
fit into DC??

Anna
________________________________________

Anna Clements
Data Architect, University of St Andrews
01334 462761
[log in to unmask]






On 03/08/2011 13:40, "Lawson, Gerald J." <[log in to unmask]> wrote:

>Colleagues,  
>
>1. I presume that all repositories on this list have a library
>subscription to Web of Science (?), and hence have access to (some)
>real-time metadata from WoS webservices
>(http://wokinfo.com/products_tools/products/related/webservices/)?  Are
>many Institutional Repositories using this to pre-populate their
>repositories? How successful has this been? Are there EPrints or DSpace
>plugins to facilitate this?
>
>2. Research Councils mandate that metadata on papers arising from our
>funding should be deposited in appropriate repositories.  This includes
>funder and grantnumber information.  Councils have not monitored
>compliance however - and many repositories do not have these fields - let
>alone have them fully populated. Since 2008 WoS has been 'mining' funder
>and grantnumber information from the acknowledgement sections of papers -
>but less than 30% of those providing a funder name also provide
>grantnumber - and without this is it difficult to match a multi-authored
>paper to a particular grantholder.  Much editing/checking of this data is
>therefore needed.  Would the following be workable - say from 1 Jan 2012
>onwards?
>
>a) Thompson Reuters are approached to include Funder and GrantNumber in
>their Web Services Lite package (they are currently only available in the
>premium version).
>b) IR packages provide plugins to facilitate access to the WoS/Scopus Web
>Services (Scopus will provide funder information 'from early 2012').
>c) Institutional Repositories are encouraged to use WoS/Scopus Web
>Services to complete any missing records  AND to add Funder and
>GrantNumber fields.
>d) Research Councils ask grantholders to ensure that data in these two
>fields in their Institutional Repositories (or Thematic Repositories) are
>CORRECT.
>e) An appropriate repository aggregation services provides a faceted
>search of all [UK-based?] repositories and allows searches by both Funder
>(dc.contributor.sponsor) and GrantNumber (dc.identifier.grantnumber).
>The DC fields are the current EThOS standard
>(http://ethostoolkit.cranfield.ac.uk/tiki-index.php?page_ref_id=47), but
>there is clearly a need to agree this with other standards agencies -
>e.g. Driver & Open-Aire. in the EU (what about the US?).
>
>This is tentative proposal focused on the UK - but comments from anywhere
>on the feasibility/timescale would be welcome.  For example the NSF/NIH
>Funding Dashboard (http://rd-dashboard.nitrd.gov/) makes heavy use of
>funder information from WoS - it would be interesting to know how
>mistakes/gaps in this can be corrected in the US?
>
>Gerry Lawson, NERC Research Information Systems, 01793-444417 (o)
>07740-068060 (m) [log in to unmask]
>________________________________________
>-- 
>This message (and any attachments) is for the recipient only. NERC
>is subject to the Freedom of Information Act 2000 and the contents
>of this email and any reply you make may be disclosed by NERC unless
>it is exempt from release under the Act. Any material supplied to
>NERC may be stored in an electronic records management system.