> I agree with Richard Gartner's comment about tables (I'd quote it
> here, but my deletion compulsion got the better of me).
> We will use <did> instead of tables for our inventories because we
> care less about replicating the format of printed finding aids than
> about tagging the content, intellectually.
> -Kim Brookes
> Radcliffe College
The encoded examples I have seen so far seem to follow a tabular form, which demonstrates well the
hierarchical strengths inherent in the EAD - with rapid successions of containers but a fairly summary
treatment at item level.
Our existing finding aids tend to cover everything at item level - which means that you have several
hundred pages of item level description with perhaps only half a dozen points where a change in level
occurs (and not a table in sight). Thus an enormous amount of tagging will be required at item level.
As this will be the largest part of the encoding process, would it not be useful to devote some
thought to item level coding not in tabular form - perhaps a guideline model or two for minimum
compliance at item level? While we all want to put as much coding as possible in, it would be rather
pleasant occasionally to skim over some of the more arid tracts. What would be the least number of
tags required to distinguish an item (reference number, descriptive narrative and when possible, a
On the question of size, at what point do EAD finding aids become unmanageable? The eadgrp dtd
is expressly not for linking multiple accessions to the same collection, but there does seem to be a
need for such a super wrapper to allow large groups of large accessions to be dealt with in
manageable amounts. Several of our collections are simply too large or complex to have a single
finding aid: while there are occasions when it would be useful to include all parts of the collection
under a common banner, most users are only interested in one part thereof. Given that the SGML
tagging can almost double the size of file, how large are these finding aids going to get?
Durham University Library