07 November 2008

Comments

We have had a few comments, just not through the blog! They fall into a few categories as follows;

1. this is a project whose time has come;

This has come from a number of quarters; from the library world and from people interested in learning object repositories as well as those running research repositories. Authority control has been around for a long time, but it seems the new context of digital repositories has led to the issue bubbling up for a rethink.

2. the timeline is very short;

Yes, indeed. Particularly with Christmas and the New Year in the middle, we recognise that we may have to cut our cloth to fit the timeline (sorry for the mixed metaphor). There may be a some flexibility with the March deadline, but we will see.

We are also focusing solely on personal names as an attempt to keep it as simple as possible.

3. will there be an operational relationship with People Australia?

We have already had a discussion about how this project might interact with People Australia and, without wanting to prejudice the outcome of the project, it does seem only sensible to build and use People Australia as the authority file for Australian researchers.

4. identifiers and vocabularies;

We have had mention of URL/URIs, People Australia persistent identifiers, ISNI, ISADN, OpenID and various commercial researcher numbers as identifiers. There are also many developments to do with schemas, DTDs, vocabularies, etc and sorting something reasonable out of all that will be a core part of the project.

2 comments:

Norman Paskin said...

You may already be aware of the following, which appear highly relevant to this work:

Existing work on Identifiers for Institutions, Public Identities, and Researchers – see recent webinar http://www.niso.org/news/events/2008/webinars/identifiers (If you have any trouble viewing this archive, please contact Karen Wetzel, NISO Standards Program Manager, at kwetzel@niso.org.). This covered an introduction of existing projects and standards, and details on three projects:
1. NISO Institutional Identifiers Working Group. http://www.niso.org/workrooms/i2 . Intending to "develop a standard for an institutional identifier that can be implemented in all library and publishing environments. The standard will include definition of the metadata required to be collected with the identifier and what uses can be made of that metadata". There are in fact already several used to various degrees (SAN, DUNS, ISIL etc). This is designed to overcome the lack of a common way of identifying the institution with its multiplicity of libraries, departments, campuses or offices. Internally some publisher already do this (the pioneer being Elsevier's internal Sales Information System); a similar shared solution is now available from Ringgold.
2. Researcher Id. www.researcherid.com. A Thomson product for Web of Science - not a formal standard at all; intended as a tool for disambiguating and de-duplicating author entries in Thomson's Web of Science.
3. ISNI. International Standard Name Identifier. http://www.iso.org/iso/iso_technical_committee.html?commid=48836.
Inspired by CISAC's (rights organisation) internal system for the music industry, and guided by the work on a party identifier done by Interparty (www.interparty.org); taking the Interparty work as a cue, it attempts to be a dumb number and intended use is to simply match entries in different databases and establish a re-usable link. Not yet finalised but at draft stage but not yet clear what business model can support it. Focus is on names of creators and characters (authors, pseudonyms, personas) but in principle extensible.

In Australia, the excellent work of the PILIN project (is worth noting re persistent identification infrastructure: https://www.pilin.net.au/

Stuart said...

Norman, thank you for your comment. We were aware of ResearcherID, ISNI and PILIN, but I for one didn't know about the NISO Institutional Identifiers work. So, I have a little more research to do ...