And the mapping band marches on

So, as it turns out, my metadata professor was right: a large chunk of what metadata professionals do is figuring out how to 'map' one scheme into much more contemporary, much more universal, metadata schemes. My project for today, which has been to develop a crosswalk map between a Lilly library scheme for what seems to be photographs into MODS so that all this data (and there is a lot of data here) can be exported into the Fedora opan access initiatve of which Indiana University has recently become a member. This has meant, among other things, determining which fields need to be added, which fields need to be cut, and which fields ought to be simplified.

I can't discuss this project in too much detail since I am currently in the midst of it. It will probably be the end of this week before I can say I've done the mapping completely. But, off the top of my head, I see at least two problems with the current scheme and its ability to integrate into MODS. Firstly, this being a scheme developed by archivists a bulk of the scheme (the first 80% to estimate) is devoted to pinpointing precisely who created the record itself, presumably to know where any inaccuracy in the record might have gone wrong. In MODS, however, there is very little space for this type of information and it just not satisfactory to toss 80% of the record into <recordInfo><recordOrigin>blahblahblah</recordOrigin></recordInfo>
and be done with it. Secondly, in MODS the assumption seems to be that one is dealing with a relatively self-contained entity that can be described completely by itself. However, archivists seem to work under the opposite assumption: no one thing can be described independently. Rather, every individual item in the archival repository relates very closely to the othe individual items on either side of it in some big box that itself relates very closely to the big boxes on either side of it. This concept of a record series is something MODS is lacking in functionality to handle, I think. I'm not sure what could be used to replace it. EAD is supposed to be universal; but, as is obvious to anyone who starts mucking about in the midst of it, it isn't really. It's meant to facilitate information/data transfer between a very specific subset of the information world: archivists. I suppose my real complaint in this rambling post is that if this any evidence there seems very little real collaboration between the different arenas of the information industry and this lack is only to the information professionals' peril as well as the alienation of their clients and customers.