Dan Needham, the developer working on the Names Project, attended the Web Services and Repositories workshop that was organised by the EThOS project and held at the British Library on 2nd June.
He gave a presentation [PowerPoint format, 205KB] on the project and the aims behind the web services for the Names prototype that he’s been working on and recently testing with colleagues from Cranfield University.
UPDATE: the audio from Dan’s presentation and all the other materials from the day are now available on the EThoS site.
The Coalition for Networked Information (CNI) convened a workshop in Washington, D.C. yesterday entitled ‘Authors, Identity Management and the Scholarly Communication System’. Attendees included publishers, university repository-builders, and authority control experts from the Library of Congress. A number of us gave presentations about our activities and there was time for a decent amount of discussion about the various issues faced by the different constituencies that were represented.
The meeting was very timely for the Names project, as I was putting the finishing touches to the Requirements Report and was able to incorporate some of the conclusions of the meeting (as I interpreted them, anyway) into that document. These included the following observations:
- There is unlikely to be one master list of author identifiers. Consequently, systems should:
- maintain information on other known identifiers for an individual
- support queries on names (and other metadata, if available) via a Web Service and return responses, to include some or all of: author names, affiliations, dates, article titles and other known identifiers. As yet there is no standard defined for this query/response
- Author identifiers need to be URIs that are resolvable as URLs – no one identifier scheme has been agreed upon for use by all name authority systems
- There are two points of vulnerability in systems that rely on author input:
- At the point of data entry (this relies on registration mechanisms of the publisher/repository – is this person really the author of this material?)
- Erroneous assignments of authorship (accidental or deliberate) by creators within the system – there need to be mechanisms for resolving disputed ownership of materials
- Systems should support UTF-8 encoding to allow for non-Roman characters and names
- It is better to link identities across systems than to merge identities – easier to dismantle if the individuals turn out not to be the same person
- Availability of the data may be an issue, in relation to privacy. Much may already be in the public domain, but bringing information about the affiliations and works of an individual into one public place may open the data to uses (perhaps commercial) that have not heretofore been possible. Thus it may be best to share the data only with trusted partners and to support queries, rather than to make the entirety available.
CNI will be producing a report of the meeting, which I’m sure will be more comprehensive than these notes of mine.
This Flickr image from AlbinoFlea shows the vertiginously long escalator at the north exit of the Dupont Circle Metro station in Washington. It wasn’t working when I arrived on Sunday night – walking up it with my bags was not much fun. Maybe it’s a sly attempt by the city authorities to improve the fitness levels of Washington’s citizens.