The Software Requirements Specification document is now available to download from the Names website documents page. The document catalogues and describes each of the recognized functional and non-functional requirements that will be placed on both the Names prototype software and a future operational name authority service. It also provides several software use case scenarios from which many of the requirements have been derived.
Throughout the process of developing this document we’ve taken into account individual requirements and feedback from a number of sources and the result is a comprehensive analysis of the conditions required to create the most flexible solution possible, and also constraints within which it must work.
Due to the investigative nature of the project and the iterative approach to development these requirements may alter as we make further progress, however the fundamental underlying functionality of the software described here should remain unchanged. Please do comment on the requirements if you would like to feed into the development of the Names prototype.
The Names project was represented at two JISC-organised meetings last week.
Federated Access: Future Directions Day
The first, on 30 June, was the Federated Access: Future Directions Day in Birmingham, organised by JISC’s Access Management team. Names has relevance to two of JISC’s funding areas: repositories and identity management. The latter is mostly focused on helping institutions to implement the Shibboleth access management system, but this meeting was set up to look at identity management in its widest sense.
Not being an expert on federated access systems, I was concerned that my talk would not be directly relevant to the group. Well, my talks, in fact, as the parallel session speakers had to do their talks twice, both before and after lunch. However, there were some good questions about the project in each session. Slides (via SlideShare) are already available from the event page and audio (eek) will follow soon.
The meeting was called to look at funding directions for a £2.5 million wodge of funding that JISC will be spending in the area of access management and identity in the next three years. Our break-out group was led by Nicole Harris, who laid out the various suggested funding priorities on the wall of the conference room and invited the group to arrange them in order of priority (rather in the style of the cool wall on BBCTV’s Top Gear). One of the categories was ‘Domain Integration: Repositories/Author identities’, which (I was pleased to see) ended up on the cool side of the wall, together with other cross-domain initiatives. I didn’t put it there, I hasten to add.
The JISC Access Management team itself will cease to exist soon and this area of work will be taken forward by the Information Environment and e-Research teams. I forgot to take any pictures during the event, but I did like the name of this pub that was near the conference venue, so that will have to do instead.
Repositories Infrastructure Workshop
The second meeting was a workshop held in London on 3 July, called to look at the architecture required to support repository functions. Paul Walk gave a presentation in the morning about how such an architecture might look – links to his slides are available from his blog entry about the day. As Paul points out, there was a fair amount of discussion about repository content being in subject rather than institutional repositories and there was a lot of focus on making the process of depositing materials in one or more repositories as painless as possible. A name authority service was mentioned as being a useful part of the infrastructure on a number of occasions. There is a pretty full transcript of the discussion on JISC’s wiki.
I was pleased to meet Brian Rea of the National Center for Text Mining (NaCTeM) at the event. There is scope for using text-mining techniques in a future phase of the Names project, particularly when it comes to enhancing existing name authority records (for example expanding first initials into full first names), so it was interesting to be able to talk to Brian about that. I also caught up with Phil Vaughan, who is now working for the British Library as the Programme Manager for UK PubMed Central, and talked to him about possibly using information from the grantees data in UKPMC for the Names prototype.
There is a ‘crowdsourcing’ website that is being used to gather suggestions, comments and votes for aspects of the repository architecture work. Another meeting is being held on 21 July to take this area forward.