My impression is that very very few institutions are using the strictly
recursive model for components, both because it is more difficult to manage
the authoring process and because, as Jackie points out, it really doesn't
seem to matter very much if you use the level attribute to convey the
conceptual level of the component. I use the word conceptual advisedly
because the concept of a series, like that of beauty, lies in the beholder.
Moreover, there is a theoretical problem with naming conventions. Holmes
defined five levels of hierarchy I believe but many collections, and
therefore inventories, contain many more levels of organization. Some of
these are simply filing units. What does one call the third level below
the series that is still not to the file level? Is it a
sub-sub-sub-series? What's the value in trying to determine the answer?
I wonder if we are not more hung up on archival theory and terminology than
on the concept that to me users need grasp- that there is a hierarchy of
organization and arrangement that they need/are able to navigate. I doubt
that the public cares much about what the mileposts are named.
Having said that, it is certainly possible to create stylesheets that
produce displays that are driven more by the level of a component than its
enumeration. I have used two different approaches with XSLT stylesheets.
One has been to create, within a single stylesheet, multiple options for
display depending on level of the c01 component. Is it a sub-collection, a
series, or a file? We see these three variations in our archives depending
on the relative size and complexity of the collection. Our basic
stylesheet offers three options for presentation based on which of these
situations is present. However, we were motivated to create different
displays simply by presentational concerns, such as what should be indented
and how much.
I am also now creating another stylesheet in cooperation with an archives
that shall remain unnamed that wants the display of components to be
strictly relative to component level. You need to understand that they
have very little hierarchy- no more than four levels- and do not use
indention to express hierarchies but rather distinguish series, from
sub-series, from files, from items simply by type face, size, and emphasis.
It will be interesting to see what reaction users have to this approach.
Michael J. Fox
Acting Assistant Director for Library and Archives
Minnesota Historical Society
345 Kellogg Blvd West
St. Paul, MN 55102-2409
[log in to unmask]
> From: Jackie Dooley[SMTP:[log in to unmask]]
> Sent: Friday, December 08, 2000 11:11 AM
> To: [log in to unmask]
> Subject: Re: Series list and descriptions
> Susan von Salis wrote:
> "... At Harvard we tend to use recurive <c>'s; then it doesn't matter that
> different finding aids have different "levels" (e.g. in one finding aid a
> folder description might be a <c02> and in another it might be a <c05>,
> depending on how many series and subseries there are) ..."
> I would argue that it already "doesn't matter" that different finding aids
> have the same intellectual level encoded at different <c0x> levels. The
> Online Archive of California stylesheets, for example, simply don't
> specify particular numbered <c>s to effect display of specific
> intellectual levels. That's what the LEVEL attribute is for in a <c0x>
> (e.g., <c02 "level"=subseries>).
> Jackie M. Dooley, Head of Special Collections and University Archives
> UCI Libraries, P.O. Box 19557, Univ. of California, Irvine, CA 92623-9557
> Internet: [log in to unmask] Phone: 949/824-4935 Fax: 949/824-2472