AustLit
AustLit is often given as an example of how FRBR can be put to very good use, but if you’re like me and not living in Australia then you probably won’t get a chance to see it in action. However, you can see some samples of how it works, and read lots about it. It looks really useful and will be of great interest to anyone implementing FRBR.
I assumed AustLit was open to all, but it’s not, it’s only available to subscribers. I wanted to see how they handled Sean McMullen’s Souls in the Great Machine (1999), which is a work that is a blending of two earlier works. (You should read it: it’s set in an Australia one thousand years in the future where librarians fight duels to the death.) Here is AusLit’s entry for Sean McMullen, but non-subscribers can’t see any of the interesting information behind it.
Nevertheless, they have some sample pages up for viewing. Their basic search is complicated enough to be anyone else’s advanced search, and their advanced search lets you build a customized search. They have a page up about Patrick White’s Voss, and it shows three related works (including an opera) and fourteen expressions, some with multiple manifestations. (You may wonder why the English version of the novel isn’t listed first, it being published first, and so did I.) They stop at the manifestation level; item information, which is to say library holdings, can be found elsewhere.
One great thing about AustLit is the amount of documentation they’ve put online. Don’t miss AustLit Data Models, this development site with things they were thinking about in 2000, and a huge manual about what a work is and how to handle one.
Marie-Louise Ayres, who was in charge of the project, has presented papers about it, such as Case Studies in Implementing FRBR: AustLit and MusicAustralia (2004, in Word) and AustLit: A Gateway on Steroids (2001?, with Kent Fitch, Annette Scarvell, and Kerry Kilner). That second one has this nice statement:
One of our major worries in adopting the FRBR model was that it could prove too expensive to create and maintain FRBR records. This has certainly not proved to be the case. Educating our staff about the FRBR model certainly took a lot of work, especially because practical implementation raised many issues. But once they were familiar with the model, they loved the fact that it allowed them to represent works in a rich context. They also thoroughly enjoy the maintenance interface which gives them many choices about how to describe works and authors, and gives instant satisfaction: create or edit the record, update it, see it in the browser immediately. We also have a very effective review interface, which allows our two Content Managers to review work and provide timely feedback by email. From a management point of view, I am absolutely thrilled about the productivity of the staff, and our earlier fears about the expense of both implementing and using the model have been emphatically put to rest.