sober evaluation of libraries’ repository endeavors

Dorothea Salo says some things I’ve been thinking too, but was scared to say in public, especially as an outsider to repository endeavors who maybe just lacked a full understanding to make my thoughts valid. But I trust Dorothea to know what she’s talking about.

What I think we need now is a triage-and-rescue operation for existing repositories in the United States. Eyes open going in: some existing IRs should be shut down, because their libraries and institutions cannot or will not support them at a viable level of function. This should not be viewed as a loss; it is a gain, because these sad limping excuses for repositories cause significant damage to the perceived viability of IRs generally. Arrangements will have to be made for the material inside them, but I don’t perceive that as a major barrier; willing hosts can be found, and the migration path is technically feasible.


I like this Ben O’Steen post as an illustration of the problem. A database? In DSpace? Fuhgeddaboudit. Digital librarians and IR managers everywhere need to read that post, though, because that is the future, passing right before our eyes. Not nice tractable Taylorist image-digitization projects that we’re already so good at dealing with. Nasty, complex, ad-hoc problems requiring innovative data-modeling and UI design as well as individualized project management. Are we ready for that? Are we hell. Do IRs make us ready? Do they hell. Walk up to Ben O’Steen’s problem with a cardboard box. I double-dog-dare you.


So that’s triage. The form I think rescue should take is, I hope, obvious by now: more weirdies, more devs, and more toolboxes. This means pragmatic staffing and resource-allocation roadmaps for library and IT administrators based on all the up-and-coming good repositories I mentioned above. This means admitting that DSpace is a Brooksian plan-to-throw-one-away lost cause, and folding it into Fedora. (Speaking of calling it as I see it…) This means no-nonsense service definitions that stop making labor invisible. We can do this! I know we can. Many of the pieces are in place already. But somebody has to step up and do it.

There’s actually two seperate but related things being critiqued here.

1) The way IR projects have been approached by libraries, and if those projects have been useful. What goals were defined as, if they were the right goals, if the right goals were met, if they were staffed appropriately, etc.

2) The open source software libraries have collectively put quite a bit of resources into developing, and if that software has actually turned out useful for the amount of resources put into it, if that software has turned into the right tools for the right goals.

PS: Anyone know who is this Brooks and in what context he plans to throw one of something away? That sailed right over my head.  Taylor, I know he was.


5 thoughts on “sober evaluation of libraries’ repository endeavors”

  1. I also have a very distant understanding of IRs, but have the general impression that at least many of them are really old-school silos that don’t integrate very well with how faculty actually work. My institution has one, for example, but I’ve only ever looked at it once, and that was enough to convince me not to go back.

  2. Bruce, that’s not unusual at all, more’s the pity. On behalf of repository managers everywhere, I apologize for not serving you better.

  3. Brooks = Fred Brooks, author of “The Mythical Man-Month”. IIRC, he wrote that when implementing software projects, you should plan to throw the first iteration away. That is: the first iteration is a learning process that flushes out implicit requirements and bad design/implementation decisions.

  4. Thanks Dorothea.

    For the record, I just heard from one of my institutional IR people, and I suggested BibApp as a nice alternative to the current DSpace-backed solution (though I gather they’re not exactly mutually-exclusive).

    OTOH, I’ve been thinking a lot about the general lack of integration not just of research and publishing, but in turn of teaching. Would love something like BibApp integrated into a more general CMS like Drupal, and to be able to tie this sort of functionality into, say, my department site (which I’m currently redoing).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s