The DSpace Community held its User Group conference in Rome this year that was every bit as interesting and entertaining as last year’s DSUG in Bergen. The conference was held at the UN Food and Agriculture Organisation (FAO) headquarters in the centre of Rome. The 8th Floor terrace restaurant at FAO must have one of the best views imaginable, overlooking the Forum, the Circus Maximus, and the Colosseum, with the winged chariots of the Victor Emmanuel monument and the pale blue-grey dome of St. Peter’s beyond.
DSUG can be a slightly tense affair, since it must address the expectations of both technical and non-technical staff engaged with repositories. Interesting as I found Federico Meschini’s demonstration of managing the DSpace code base using Eclipse and Maven, I think many librarians attending found this “basic” tutorial beyond them. The slightly aspirational tendencies of several discussions can also be a little frustrating. It’s interesting to know what’s planned for version 1.6 and 2.0 of DSpace, but since 1.5 has yet to be released, this can seem like pie in the sky.
Stefania had asked me to contribute a short presentation on SAS-Space for a session of case-studies of DSpace instances. The brief case-study I presented was rather less than earth-shattering, and my plans to show a little more of it than the front page were scuppered by the FAO firewall. But there’s only so much you can say in 10 minutes: I hope I was at least competent. What seemed to provoke the most interest, in later discussion with other delegates, was the fact that we haven’t restricted repository content to peer-reviewed articles, theses, and dissertations. This probably never seemed at all surprising to me, after so many years of association with a dataset archive. Things like the Masefield bibliography, London Book Trades database and 19th Century Francophone Music Criticism transcriptions are equally valuable scholarly works, in my opinion, and worthy of preserving and publishing in an Institutional Repository.
Interest was also expressed in the idea of using a wiki to manage the documentation associated with the SAS-Space Repository - policy, user guide, and such like. I’m not sure I didn’t borrow the idea from another IR, but, all the same, it still seems like a good idea to me! (Down with websites and CMS!) What I’d most like to do is integrate the wiki content a bit better with the DSpace pages/templates. I’ve been unsure how this might easily be done, but now I’ve seen Manakin…
Scott Philips’s demonstration of Manakin was particularly interesting to me, as Manakin uses an XML/XSLT approach to rendering repository metadata that has a lot of similarities to the approach I developed a few years ago for generating NDAD catalogues (albeit with a Perl backend, where Manakin, of course, uses Java).
Manakin offers a three-tier abstraction model:
- Aspect Tier: This is the Java/Cocoon backend that assembles a METS-based XML document from the repository database.
- Theme Tier: XSLT is used to transform the raw XML into an XHTML document
- Style Tier: CSS styles the XHTML document
This means that no intervention in the Java source is necessary for any modifications that can be addressed in the Theme or Style tiers. For example, adding the logo of the institution to which the item belongs to the Search Results view can be achieved by adding some simple templates to the default XSLT templates. The Texas Digital Library is running entirely with a Manakin-based UI, and very nice it is too.
One other particularly interesting presentation and discussion I hope to follow up is Federico Meschini’s work on unifying a DSpace repository with a Learning Object repository. This involves establishing mappings between Dublin Core and LOM metadata schemas - something that may be of interest to us for work the ongoing SPELOS report.
I was also interested to find out more about CASPUR, with which Federico and the conference organisers are associated: it is what remains of the consortium that operated supercomputers for the Italian academic sector. Sound familiar?
I enjoyed the chance to chat to Robert Tansley, one of the people behind both Eprints and Dspace, who’s obviously a very sharp and clever guy. Rob’s now moved from MIT to Google, where he’s working on hush-hush research projects, but continues with DSpace/Google Scholar-related work as part of his side projects. We both agreed that, thanks in part to work on DSpace, Eprints, and their ilk, and the many standards and developments they embody and build on, the technical problems of managing and preserving most common types of digital documents seem largely solved. Rob talked of his interest in new approaches to visualising data, relationships and the like (notable Gapminder)and seemed genuinely interested when I suggested that the NDAD dataset archive might contain some material worth experimenting on. I’ll be dropping him a line, and a link, in the near future.
There were plenty of opportunities to chat informally with other repository users and developers, during lunch and the Google-sponsored evening reception, and at the conference dinner. At €150 for the whole conference package, it’s hard to imagine better value: Imma, Paola, Stefania and others did a fantastic job. The organisers also designated a unique tag, dspacerome2007, for sharing online resources from the course, including photos in Flickr and links in del.icio.us (just as we had done for photos of the Bergen meeting last year). And I’d barely even touched down when I found I’d been blogged in Italian…
 |
 |
 |
 |
 |
