Back in Europe

My first day in Britain is spent at the meeting of the researchers' working group for the International Internet Preservation Consortium (IIPC). This is within a couple of hours of stepping off the plane from Brisbane - and I can now also claim the privilege of having showered next door to the boardroom of the British Library, which is where we have now convened (the boardroom, not the shower).

King George III's library
inside the British Library.
Several stories of books!

Steve Schneider from SUNY Utica presents on the Web Archivist system to support collaborative Web archiving projects. There are important ethical issues to be considered here, and Steve identifies some important tasks: identify, acquire (objects and rights), verify, catalogue, and provide discovery aids. Who does all of this, and who does it well? Libraries and scholars con contribute to each task, or could both do everything themselves (and then perhaps miss out important things) - so Steve stresses the importance of collaboration.

But who, if anyone, directs this collaboration? Scholar-directed projects could be treated by libraries as 'collections' (and libraries have much experience in processing donated or deposited collections). Library-directed collections are driven (obviously) by the librarians, and then made available to a more general audience; libraries assume responsibility for the collection, but not for the analysis.

In collaborations there would be a mixture of responsibilities; advantages then are funding (multiple sources, mutual leveraging), professional development (mutual learning, broader distribution of outcomes), and systems development (sharing of costs and expertise). On the other hand there is more time spent organising the cooperation, of course (high interaction costs), and a lack of total control by any one partner.

Web Archivist has various projects: the 2002 US Election Web Archive and the September 11 Web Archive. Its underlying architecture involves managing systems for collections, crawl, metadata, and access. There is plenty of detail on the processes involved which will be available in Steve's full paper on this. Most importantly, with a system (which is entirely Web-based) like this institutional collaboration is enhanced; data streams between organisations and collaborators are largely seamless; and XML is used as a fundamental standard. Steve argues forcefully for collaboration amongst the partners involved.

The issue of collaboration costs is raised further in discussion. Especially initially these are high as partners work out how to work together; it is hoped that the costs will go down both as experience grows, and as past experiences are shared and collaborative tasks are standardised so that in future projects they are easier to understand and undertake.

Identify at Web Archivist is an example of a public identification system through which scholars and librarians could collaboratively and interoperatively identify worthwhile resources.