Monday, April 12, 2010

Plenary 1: Technology and Change, Richard Wallis

“Libraries have always been at the leading edge of technology” says Richard Wallis, Technology Evangelist from Talis, as he shows us some photos of very early OPAC systems, many times removed from the modern day library website. To further illustrate the changes of the past several decades he shows us a calculator from the 1970s, whose £39.95 price is the equivalent of £400 today, but whose functionality is now a free app on almost any electronic device you care to name.

The information industry is all about helping people to find things and linking students to the resources that they need. We need to rethink how we do this, bringing the information directly to the user, in the format that they want. There should be no need to bounce the user via resolvers and multiple URLs to a site that eventually proclaims “Here it is!”. It should just be delivered.

Getting users to the answers they need can be done via search (searching the OPAC or more likely Googling) but answers can also be computed (WolframAlpha) or navigated to. All of these methods rely on metadata. Librarians and metadata, Wallis points out, go back a long way, but there’s still a way to go. Library metadata for the most part is “built on principles that worked for physical stuff” and is more often than not only available from the library. This metadata can be made more useful.

And so to Linked Data, which, put simply, identifies things, links to those things and describes those things. The four principles of Linked Data are
  1. Use URIs as names for things
  2. Use HTTP URIs so people can look them up
  3. When someone looks up a URI provide useful info about the thing, in the right format (for human or machine consumption)
  4. Provide links to the URIs of other things to aid navigation.
To show some examples of Linked Data Wallis bravely attempts - and pulls off - a web demo. He shows us education.data.gov.uk, a semantic store of data about UK schools, and BBC Wildlife, which pulls in its descriptions of animals from Wikipedia.

What about libraries? Some are experimenting in Linked Data - the Library of Congress, National Library of Sweden, for example, so we’re seeing the start of library linking hubs. But we need to go further. Library linking hubs should link to non-library hubs (government data, the internet movie database, many other resources) and the library catalogue should become a set of links between concepts, become part of the Linked Data web.

The drivers for this evolution are not likely to be local cataloguers. Metadata for e-resources needs to be good and needs to delivered with the resource. Article level metadata can’t be catalogued, there’s just too much. But it is being aggregated (by CrossRef, for example). As this metadata gets better we’ll start to see non-library hubs linking to library data.

In summary, technology is evolving extremely quickly, and consumers are driving delivery methods - “get it to me on my device”. Education needs to link students to resources and search is only one way of doing this. Linked Data is powering the web but mostly outside of libraries, and libraries and publishers need to catch up.

“You can add great value to the web, but you need to be proactively of the web to do it. They won’t come just because you build it.”

See the slides here.

Labels: , , , ,

0 Comments:

Post a Comment

Links to this post:

Create a Link

<< Home