DIY LOCKSS?

When I was but grasshopper-knee tall, my father the anthropologist took me to his university's library to help him locate and photocopy articles in his area of study for his files. He had two or three file cabinets full of such copies. (He may still.)

I have similar file cabinets, two of them: my del.icio.us account and my Zotero library. The del.icio.us account consists merely of links. The Zotero library, on the other hand, includes the actual digital object(s) as often as I can manage it (even at a major research university like MPOW, I cannot always lay eyes on everything I want to read). Zotero is capable of holding onto those items for me, and even backing them up in "the cloud" (actually my university-provided, passworded WebDAV space) without setting them free on the open Web in ways that would clearly and obviously violate copyright.

Now let us consider LOCKSS, and particularly the variant known as "Controlled LOCKSS" or CLOCKSS. Without wandering into the techie weeds, these programs do some elementary digital preservation on the e-journal literature by reproducing it widely in a geographically-distributed fashion, and coming up with policies for the opening of parts of the dark archive thus created in case of a crisis that removes the normal methods of access.

The thought occurs that Zotero, Mendeley, and similar bibliographic managers are a sort of do-it-yourself LOCKSS system. Metadata? Check. Digital items? Check. Reproduced widely in a series of dark archives? Check. All we're missing is the crisis-policy piece.

I don't wish a huge e-journal database loss on anyone, believe me; we would all be the poorer. I do feel just a tiny bit relieved, though, about this particular emergent effect of the widespread use of popular bibliographic managers.

This adds another soupçon of urgency to the movement for open data, to my mind. Open data will often (storage space willing) be reproduced by other researchers wishing to work with them. All by itself, organically, that phenomenon helps insure those data against loss, destruction, falsification, and other evils.