OER and linking up resources to users

Reading some Tweets yesterday from David Davies and others has made me want to write at a bit more length than 140 characters about opening up resources to users. David blogged about a post by Tony Hirst on opening up library resources to searchers.

This has made me think again about how we structure collections of information, whether it be books on physical shelves or electronic articles in a bibliographic database. So much of my job in teaching students and researchers how to find information is spent describing how databases/search engines/libraries/indexes etc. are put together. Librarians like me come up with clever analogies to help demystify databases and the necessary search strategies ("imagine you're arranging a dinner party, you have to go shopping, now we need to put together a shopping list" etc.). The reason we spend so long training students how to do this is that, in many cases, databases aren't built to help people find information, they are built as repositories.

Some databases are better than others, but I think we should glean something from the success of Google and Wikipedia about usability and how users instinctively look for stuff. I don't want to turn any of my students or researchers into librarians and it can be frustrating sometimes, having to go into great detail about MeSH searching and Boolean operators when all they want is to find the perfect article.

I think that we need to concentrate on pushing content out to users, to places where we know they will be looking. At Birmingham we've configured our e-resources so that if a student is on campus or off campus and signed in to our eLibrary, then any search they do on Google Scholar will display our e-journal holdings (along with all the open web results) with a branded link. This way users are guided back to the (very expensive) resources that they are entitled to.

I go back to my shopping analogy again. In my local Tesco there are three locations where I will find pistachio nuts: the baking aisle, the wholefoods aisle and the snack aisle (all with different prices but that's another matter). What Tesco's are doing is placing the product where they know shoppers will go, rather than forcing the shopper to behave in an unnatural way. This is what we must do with information. Don't try to turn every student into an expert searcher, place the information at their point of need.

This is why I think that really thoughtful metadata can be very helpful. It's all very well having a sophisticated taxonomy with technical medical terms but if your users don't think or search in that way, they'll never be linked up with what they need. Maybe this means more emphasis on user collaboration/consultation when building databases (without it going all Homer-Simpson-designs-a-car). Or maybe the semantic web will solve all of these problems!?

I think we need to ask ourselves what we really want our databases/repositories to be: carefully structured but inpenetrable warehouses or open, usable, welcoming goldmines.