The SWORD is mightier than the pen

I am just relaxing in a hotel in Redmond, WA, US after two week’s very hard work instead of going downtown Seattle and shopping (which I hate unless it is for Obelisks). So the posts are going in all directions today. Incidentally a post seems to take about at least an hour which is why I don’t post every day. I am seriously considering audio blogs (podcasts) and would welcome advice.
Part of my environment now is “digital libraries” – at least I get to interact with librarians, informations scientists, etc. There is a real buzz of change in the air. One feature of that is the Institutional Repository (IR) – a digital place where people in an institution (usually a University) put digital things.   That’s as close as I can get to the communality of vision. Why are people doing this – why are they NOT doing this? What do they want to put in? I’ll probably blog about this later.
Caveat Lector is an intriguing blog from a librarian (Dorothea) an ex-publisher who describes herself as

Sure, I’m a geek. But I’m not a gadget geek or a Lib2 geek or even a web geek (more than incidentally). I’m not even really a markup geek any more; I sling XML now and then, but as a side requirement of my real work rather than the focus of my professional attention. What I am is a problem-solving geek. I have a problem with a technology (say, hm, I dunno, DSpace?), I beat the living daylights out of it with the nearest handy rock until either it does what I want or I decide that the problem needs a better tool than a rock and give up (complaining bitterly afterwards, of course).

She sounds like the sort of person that we desperately need – an XML-slinger who works in libraries and information science. We collaborate with Jeremy Frey (Southampton – CombeChem eCrystals, etc.) and yesterday that we agreed at the top of our shopping list we needed informations scientists embedded in chemistry (and other scientific) departments. That’s the role of the modern “library”.
Anyway there has been a lot of debate about repositories and how to get stuff into them. As a scientist I know that this requires two conditions

• There has to be an overwhelming motivation for the scientist (like losing their job if they don’t do it) and…
• … it has to be trivially easy

At present neither of these are true. Recently JISC has announced a project SWORD to help people (authors?, librarians?) reposit material. That is a great advance. Jim Downing who advises me on much of this thinks it’s important. And I am supportive of what JISC is doing with repositories (readers should allow for the fact that they are supporting us!).
So let Caveat Lector tell it her way…

Well, hot damn

I ask for middleware, and lo, there is middleware!

(No, I don’t think there’s a direct cause-and-effect relationship there. Even on my worst days I’m not that arrogant! Just shows that I’m not the only person with that particular train of thought.)
Bring on the SWORD, y’all. I’ll wield that baby, you betcha.

Dies Martis, 24 Aprili 2007

Repository middleware

I did a lot of IR marketing this week, despite my perfect awareness that IR marketing doesn’t work. For a tactic that doesn’t work, I did manage to come away with some contacts, and it appears that the IR made its way into some heads, and that’s all good.

But if marketing doesn’t work, what does?
Here’s the problem I’ve got: there’s a ton of material that’s IR-ready floating around, but I can’t get at it. My nose is mashed up to the window of other people’s hard drives, web servers, workflow silos, and collaboration tools. I want the stuff that comes out of those arenas. I just have no way to grab it.
Here’s the problem everybody else has got: they need the curation, preservation, and “put this important content somewhere safe (but otherwise out of my hair)” tools that an IR theoretically provides, but they don’t need the hassle of extra deposit steps. They need an “Archive It!” button. They just have no way to build one… if they even know about the IR to begin with.
I need middleware, and I need it badly.
I don’t think DSpace or EPrints developers should be directly considering building the kinds of tools that Peter Murray-Rust is talking about. We’re the wrong people for the job (we can’t even do versioning!), and the job is being done elsewhere by others anyway, because faculty want and need these tools, and IT is finally listening. (I have direct evidence of that from my own job, but I need to keep fairly quiet about it because work is ongoing. You’ll just have to trust me.)
What DSpace and EPrints developers should be considering is how to hook IRs up to the firehose of research products those other tools are producing. By my one-horse back-of-the-napkin calculations, that means an ingest API (no, not a command-line batch import tool, an API!) that is configurable enough to authorize certain tools for unmediated deposit and then prepopulate metadata fields with what those tools “know” about their content and the people who use them.
It’s a tall order, but I dearly hope it’s not impossible, because I want to get my IR’s ingest pipe connected to that firehose.

One caveat (small C). If you build it they may not come.  That’s the challenge. If they do come, and it’s a good SWORD, they’ll be less likely to go away. That’s why we have to work on all areas of encouraging people to capture and reposit digital artifacts.