Caveat Lector

This is interesting, and also enormously cool. The granddaddy disciplinary repository of them all, arXiv, can accept SWORD deposits! Thanks, arXiv!

I entirely agree with all of Dr. Ginsparg’s comments as stated, so I won’t restate them. I do note that my characterization of Dr. Harnad’s single-deposit vision as impractical holds up nicely; arXiv is prepared to accept deposits from IRs, but it is not prepared to send IRs its own deposits. Ginsparg’s comments don’t look to me to offer much hope of that happening, either.

I offer, through my usual Achaea University personae, some scenarios that might present a “natural functional advantage” to the IR-plus-SWORD-plus-disciplinary-archive route:

• Achaea University adopts a Harvard-style open-access mandate. If she wants her articles in arXiv as well, Dr. Troia must rather annoyingly dual-deposit… unless Achaea’s IR implements a deposit pipeline to arXiv, in which case the most she has to do is tick a ticky-box (and I can imagine ways to abstract away the ticky-box).
• NSF open-access mandates become a reality, and Dr. Troia has to send her work to a PubMedCentral analogue. Again, if she also wants it in arXiv, she has to dual-deposit, unless her IR gives her a hand. (Multiply this by the number of OA mandates any given research project might be under. I can easily imagine projects funded by multiple federal agencies, each with its own mandate, on top of campus mandates.)
• Achaea University builds a research-reporting system. As under the Harvard-mandate scenario, Dr. Troia is typing metadata at least twice unless the system can send out to arXiv—and that’s a completely separate question from whether Achaea has an IR at all.

Libraries and repository-rats, the ball is in our court now. We need to start working on showing that the above scenarios, in addition to other possibilities, are workable and useful. I know that once the 1.0 release push is past, I will be talking to the other BibApp folks about this. (I’m actually rather gleeful about all this on my own account, because our first BibApp pilot was with the local Department of Engineering Physics, so there are real possibilities for test projects there!)

Repository software developers: our repositories need to be SWORD deposit tools as well as SWORD accepters. I’m not sure how that plays out in practice. (Apologies to the late Isaac Asimov for the title of this post, incidentally.) The metadata-crosswalking problem Ginsparg points out will have to be solved as well, though the technical underpinning there needn’t be too frightening, I don’t think.

We will, unfortunately, need to give serious thought to our grotesque metadata, why it’s happening and what to do about it; Xia and Ginsparg aren’t wrong. The BibApp has some fairly useful controls on this, but it’s an ugly, ugly problem; it has been since long before Lagoze et al. lamented it in 2006 at JCDL. As I pointed out in the name-authority article, another piece of this problem is propagating corrected metadata in some standardized manner, because corrections do happen at various parts of the process, and it’s counterproductive not to propagate them. OAI and JISC, this is to your address, I believe.

But on the whole, this is a road forward! I couldn’t be more pleased.