While this is not directly an EAD question, it is one that implementors
may be in a good position to consider or answer.
In the days before online handlists, when retrieval was solely based
upon a collection level description (such as those in Evans and Weber
_MARC for archives and manuscripts. A compendium of practice_), the
indexing of that collection level description was the only opportunity
to register items within the collection in a way that might enable them
to be found by searchers. This leads, for example, to long lists of
people occurring within the collection who the cataloguer has identified
to be of note, although they may only appear a few times in a large
collection.
Now that there is the potential to place online multilevel descriptions,
with indexing at the relevant level of the collection, should collection
level indexing still be exhaustive, or should the index terms be
attached at a more relevant point within the list (such as series)? I
realise this question will have different implications for those
participating in an ongoing union catalogue such as exists in the USA,
where there is the need to maintain an existing standard, but here in
the UK, where we have no existing pool of collection level descriptions,
it may not be necessary. Should collection level indexes only cover the
broad topics of the collection as a whole, or should duplication of all
index terms used within the collection occur at the top level? Or some
balance between the two? While this question may depend upon not
breaking with existing practice (in the USA etc), and to an even greater
extent upon the retrieval systems to be used with the finding aids, what
would be your opinion if starting from scratch?
--
Richard Higgins
Durham University Library
|