Welcome to Orcmid's Lair, the playground for family connections, pastimes, and scholarly vocation -- the collected professional and recreational work of Dennis E. Hamilton

 
Cover Pages: Encoded Archival Description (EAD).  Ah, of course.  An EAD is MARC records brought into the XML age.  Well, the amount of tacit knowledge required to create a "correct" MARC record seemed mind-boggling to me when I checked way too long ago.  I would not be surprised that EAD doesn't get much closer with regard to coherent/consistent interchange and combined use of EADs from/among different sources.

When we worked on DEN and DMA in our parochial little world ten years ago (seems like lifetimes), we knew there was this problem around metadata and that we would not solve it.  Some of the bread crumbs we planted for metadata reconciliation were intended to let people know they had the problem, but I figured the bigger problems of content slackness (e.g., allowing anything a human reader would understand without thinking and that a computer will never figure out on its own) couldn't be addressed until the problem arose enough in practice for people to get it.  Or not.

XML with namespaces does provide an improved way to signal that maybe the content structure is not understood, even though the tag is spelled the same as a familiar word.  But since XML DTDs don't do anything about syntax and interpretation of content, we are still left to deal with the problem of malformed entries for recognizable tags.  Dorothea suggests that RELAX NG may provide the answer.  I don't know whether RELAX NG is promoted because XSD is too hard and inadequate or what.  This last problem of schema assessment is getting to be worrisome.

I am going to have to deal with this in the Situating XML nfoWare theme, so these markers are important to me.  I will fret over it later.
posted by Dennis at 5/2/2004 09:47:55 AM