The Case for Institutional Repositories: A SPARC Position Paper

So what's the scoop with Institutional Repositories? 

Link to The Case for Institutional Repositories: A SPARC Position Paper

Here at NOSM, we are currently working through the process of getting an institutional repository into place. And what the heck is an Institutional Repository you ask? Well, the link above is to a seminal paper on the topic that has been widely quoted. Warning: very dry reading. However, this whole topic is pretty important in this new era of online educational informatics.

If you want more crisp synopsis of what IRs are all about, check out Dspace at http://www.dspace.org/ - interesting how they call it the Dspace Federation - sounds like trekkie talk. But there is a reason for this - a federation of repositories that can all share and link to each other and search each others' material would make them all much more useful.

And, boy, do they need something to make them more useful. Now I am taking a bit of a devil's advocate stance here to try and provoke some discussion - I'm really not against them in principle and think that they are a wonderful idea...in principle. But I do wonder if they have been set up the right way. And why do I say this?

IRs have been around for several years now. In a recent survey conducted by the AAHSL, 42% of institutions had implemented an IR. Good. But the average number of users at each was in the single digits! Mostly they are being populated by librarians. Now I don't want to decry the valuable work of our colleagues - this is a very laudable role they are playing. But one of the points about IRs is that they should be a tool for self-archiving of material. To quote the SPARC article above:

"While institutional repositories necessitate that libraries-as their logical administrative proponents-facilitate development of university intellectual property policies, encourage faculty authors to retain the right to self-archive, and broaden both faculty and administration perspectives on these issues, they can be implemented without radically altering the status quo. "

So this self-archving is not happening. Now maybe the word hasn't gotten out yet...but I just came from Calgary, where they have been running Dspace for a while. Nice implementation. But despite quite intense "marketing", it has not exactly taken the place by storm.

Maybe there is too much focus on "archiving" - i.e. the tucking away of old dead stuff. Maybe we should focus more on using these for current and evolving material? Since many of us now work on multiple computers and from multiple locations, being able to get to our files from anywhere should be attractive. On the other hand, since these IRs are essentially open, who wants to air their dirty linen in public? It seems we've got a bit of a dissonance here. By the time we are happy to air it, it might be old hat.

Does the IR then become a graveyard of material that did not make it to publication? (Ooh, them's fightin' words, David - that should get a response if anyone actually reads this stuff). Another way to look at this is to examine the reward structures - what upside is there for faculty to post their stuff in a repository? This is touched upon quite eloquently in this article: http://www.dlib.org/dlib/november02/johnson/11john...

which is a commentary on the SPARC article. In this commentary, the author notes:

"The principal author benefits of online open access to their research pertain to enhanced professional visibility. This visibility and awareness is driven by both broader dissemination and increased use."

But where is the visibility? This reward structure might come into play when there is federated search across these repositories. Will they let Google and the searchbots in to play? and of course, if Promotions Committees start to consider such material then that might help.

To really be devil's advocate for a moment, maybe we need to take a look at the phenomenon being seen in the non-academic world. Look at how Flickr, Digg and YouTube have really taken off. There's a corpus or two that has content coming out of its ears. No problem with empty shelves there.

Admittedly, much of it is rubbish...but you can also find some stellar stuff there...and that is one of the points, you can find it. The one thing that they have done well is to make it brain-dead easy to post stuff. Login (which after the first time is automatic), give a brief description, some tags, and upload it. You're done. Couldn't be easier.

I think the really big difference here is this extremely easy way to upload material and simply index it.  

No these are not fully formed articles etc. Not the style at all. But some really useful objects in there nonetheless. And some sites like Flickr have looked at the IP issues and do encourage the use of Creative Commons.

Looking more at the ease of uploading an indexing, take a look at this nice little Captivate screencast (they call it a viewlet) created at UofC on how to upload a piece to their Dspace:

https://dspace.ucalgary.ca/bitstream/1880/43053/2/...

This is a very nice little screencast, which patiently takes you through the steps of uploading a viewlet. But there are about 127 steps...ok I exaggerate...but you get my drift. There are way too many steps to this for anybody but the most dedicated. It's gotta be simpler. You can see this with the single digit user numbers.