Today I’m giving a short presentation on “the researcher’s perspective” on open access institutional research archives, like BORA at my university. The presentation is part of the Institutional Respository Workshop that’s being held in connection with the dSpace User Group Meeting being hosted by our university library this week.

I know they want me to talk quite specifically about my experiences putting my publications into BORA, but of course I want to talk about blogs and openness and the information scavenging we do online that doesn’t always fit with a database-model of publication. And I want to talk about the difference between my own, personalised publication archive, shaped completely by me and even with my face grinning out at you, and the impersonal results BORA give you if you search for publications by “Jill Walker”. If we find and use knowledge through our social networks, the individuality and the “face” of publication archives is probably as important an accessory to it as the body language and voice and style and dress of a presenter at a conference, or as the cover and quality of print and publisher of a book. Of course those things don’t change what the person is saying – or not exactly, Romeo and Juliet is the same whether I read it in a scruffy old paperback or a leatherbound, giltedged volume. But the context for my understanding is different.

Of course, having a repository is wonderful, and I love BORA and its staff. They’ve even helped me track down stuff about rights and dealt directly with journals on a couple of occasions to ensure that putting it online would be OK. I wish everything anyone at our university published was automatically routed to BORA, even if I suppose maybe books might have to have restricted access for a while if hte publishers are to have any interest in actually publishing them.

But I want a system where I can register stuff once and automatically fetch out info, links and contextual links for my own, personalised publication page. That’s the page I really care about.

[Added after my talk, as they wanted some more specific links for posterity]
[And sadly, the net was down just as I gave my talk – so I had to explain folksonomies by interpretative dance…]

An institutional repository is a centralised database – not a knowledge community. Blogs and individual researcher websites are independent nodes in a network that allows browsing and context.

Who organises information?
1. Professionals classify it (librarians etc)
2. Authors classify it (e.g. html metadata on homepages and websites, keywords submitted by authors when uploading to BORA)
3. Users classify it – “folksonomy” – no centralised

How do you make a repository that researchers want to contribute to? I’m probably an ideal user of an institutional repository, being internet savvy and already having put most of my publications online. Yet it took me ages to start using it. Legal issues were the main hindrance – the copyright stuff involved me hunting out old copyright forms I’d signed and trying to figure out what I was allowed to do (is an institutional repository “self-archiving”?) was painful. Given my articles already are online using BORA hasn’t given me any short-term benefits. For presentation purposes it’s clearly wonderful but my articles are already online and I get a lot more feedback and citations from my own “self-archiving” than through BORA.

Here’s an example of a document repository that works wonderfully: Flickr. Flickr is an online photo sharing site. People upload their own photos, and the system provides them with so many cool ways of using their photos through this system that there’s a lot of incentive to do so. You can see your friends’ photos, you get comments, you can find other photos of the same things, you can more easily find your own photos. You can have your most recent photos show up on your website (through RSS) or you can write programs to do cool stuff with your own or with other peoples’ photos.

People also contribute wildly to sites like CiteULike, which is like an exoskeleton rather than a repository. CiteULike allows you to bookmark academic articles, add tags to them and then you can see who else has those articles bookmarked, which other articles they’ve also bookmarked, you can generate a watchlist to show you everything that so-and-so bookmarked and everything tagged “hypertext” and so on.

Adding social features like these to institutional repositories could make them useful for researchers to actually use, and not simply a smart way of preserving research publications.

My wishlist:

1. I want the data I enter to be usable by other web applications – an open API. For instance, Flickr’s open API means that I could write a program that uses my
2. I want RSS feeds that I can put in my website, the way I can put my most recently bookmarked articles from CiteULike there. (Look down in the left menu to see an example) (This is in the version currently in beta, I’m told.)
3. I want to be able to click a keyword on an article and see other articles not just in my own institution’s repository but across repositories.
4. I want the University to be the bad guy and insist that everything I publish go into the Institutional Repository. That way I can blame the university when I have to argue with the journals and publishers.
5. I want the University to make a contract I can send the publishers to simplify this process for me.
6. I only want to deal with one system! I don’t want to register my publication in FRIDA and in BORA (they’re going to be combined, I’m promised)