Making It Work

There has been a lot of thinking and work going on among clever people with regard to Google Scholar and Google Desktop.

Art Rhyno, at the University of Windsor Library, and Peter Binkley, at the University of Alberta Libraries, have been tinkering with Google Desktop and Google Scholar with most interesting and promising results.

Art has blogged his Google Desktop tinkerings on his own blog, LibraryCog which is the best place to read the details--and I confess I skim over the techie parts but I am jazzed about the results.

In a nutshell, he started musing on what it would mean to have catalogue searching carried out by "third parties" (say, Google Desktop), and he's come up with a proof of concept, using Google Desktop against the library catalogue at Windsor. I know Art so I emailed him and said, so, if Google Desktop might be able to search against OPACs does this make the Open WordCat program redundant? Why would records need to go to Yahoo! and Google if bib records can be crawled from point of origin? And this is what Art replied to me (posted with his permission).

"I am impressed by the efficiency of Google Desktop though I don't know if what I am doing with our collection will go any further than our public stations. It's a bit like the "teachable moment" that we talk about in Information Literacy, how to make library access seamless at the point of information need. The advantage of Google Desktop over fighting it out with the billions of documents that Google indexes through the web is that it reserves some prime real estate for displaying the results.

But I know that anything that relies on software to be installed locally is going to be problematic. I think OCLC's approach of working with Google is still the best. I would love to see some dialogue on how to inject rights management into this process. Google's business model depends a lot on instant gratification, exposing content that can't be accessed immediately treads on to some tough ground and I am sure the folks at Google understand this. Even going outside of the Google cache is a bit tough because it shakes the fault tolerance of Google's infrastructure, but I am sure many publishers get nervous about caching licensed content.

We are revamping our lookup bookmarklet to augment URLs in Google Scholar which are listed in our proxy tables. The idea is to automatically give patrons some chance to access content that has already been licensed for them, but this is still a pretty weak measure. Even building a toolbar, which some of my colleagues are very keen on, seems to me to be competing in a space where there are powerful forces that can squeeze us out. Still, Google has some facility now for creating profiles, and I think the hooks for an infrastructure for rights management already exist, what's missing is a forum plugging the two together, though maybe this is already happening.

To be honest, at least some of my interest in Google Desktop comes from a game that I worked on with my kids a few years ago. We named it StackQuest and the idea was that creatures chased you around library bookstacks and the lead characters had to find books on certain topics to throw at the creatures in order to escape. I have always been a big fan of Don Norman
and I loved that the Environmental Scan looked at gaming. I think we tend to badly underestimate how powerful the notion of fun can be in learning. Interestingly, I have discovered that games have a long history in [Canadian] indigenous cultures for teaching."

One of the reasons I asked Art if I could post his comments to "It's All Good" is because of the last paragraph--our faithful readers know how interested we are in gaming.

And on to Peter Binkley. What about Google Scholar, rights management, and the appropriate copy problem, Peter? No problem...he's been working on that.

Yesterday, on the Web4Lib listserv, Peter announced that "As a proof-of-concept I've built a Firefox extension that adds OpenURL links to the results lists in Google Scholar. It can be downloaded here: http://www.ualberta.ca/~pbinkley/gso/ . The functionality is pretty basic but it shows what might be possible."

If you don't belong to the Web4Lib listserv, you might want to join if you're interested in keeping track what Peter, Art and others are doing to develop practical, usable ways of integrating library metadata into the shared space of the web.

This really is All Good.