The way we used <origination> and <controlaccess> in tagging up the
Bentley finding aids was as follows:
<archdesc level="recordgrp"> (or level="collection")
...in order to record the creator of the records/collection. I agree that
it would be very useful to allow <controlaccess> inside of the
<origination> tag. Also, the Bentley decided that the name from the
<origiation> tag would reappear in all cases as one of the <controlaccess>
terms grouped further down in the f.a.
We decided to group all controlled access points for the
records/collection after the <bioghist>, <scopecontent>, and
<organization> information. I think we managed to work in the groupings
to which you refer by using <controlaccess> embedded within itself. We
used 3 groupings: contributors (would that be equal to the provenance
grouping of controlled access terms that your referred to?), subjects, and
form/genre. So ...
<controlaccess> (label "Controlled Access Terms" supplied by the system"
<controlaccess><head>Form and Genre Terms:</head>
We only used whichever of these three groupings were appropriate, given
that we were converting legacy data and not analyzing collections/records
for the inclusion of additional controlled access terms. Also, we chose
to use the <head> tags here because all three "groupings" might not
necessarily appear predictably in every finding aid, so the system could
not supply a label for each "grouping".
Would that approach address the concerns you raised? Just wondering,
mainly for my own interest, since I'm grappling with ISAD(G) and EAD right
________ [Bill Landis] ___________________________ [[log in to unmask]]
Graduate School of Education & Information Studies
University of California, Los Angeles