Pete and all--

I agreed with many of Ann's comments. Our project team and some others from
the UIUC Graduate School of Library and Information Science (we have an
informal Metadata Roundtable that meets regularly) spent some time recently
applying the RSLP CLDT to a handful of the digital collections created
through the National Leadership Grant (NLG) program administered by the
Institute of Museum and Library Services (IMLS).  We also recently asked a
small group of NLG projects to test run our schema which included a partial
list from the RSLP schema (specifically the 'Type' and 'Content' sections).

In our own experiment, we had some trouble with the 'Type', 'Curatorial',
and 'Policy and/or Usage' sections. The 'type' section -- while I
understand it as it follows from Heaney's model
(http://www.ukoln.ac.uk/metadata/rslp/model/) --  was at times hard for us
to follow (we kept wanting to replace 'collection' in the descriptions with
'set' or some other word). We wondered where a portal (though that term can
be problematic enough!) would fit into this. 'Curatorial Environment', if
used, would need to be expanded (we had one collection created by a group
of scholars, independent of a library, and one that was created by a
non-profit educational institute). The inclusion of 'internet' in this
section was also problematic  - it didn't seem to us that 'internet' was
really a separate curatorial environment. And, as Ann pointed out, the
Policy and/or Usage section was rather jumbled -- though we very much liked
that some of these terms (special, subject, form, user, working) captured
the 'purpose' of the collection or how/why the collection was gathered
together.  We're trying to capture some of this information in our
collection description schema through an element for a 'collection
development policy statement'.

In our very small sample 'test' of the schema all of the respondents used
the type 'collection' and one or more of the content types (so
collection.image, collection.text, etc).

All that said, running through the CLDT was a very useful exercise for us;
it forced us to focus on what we really wanted to know when we said 'type
of collection'.  In the end, I think that Pete's proposal to adopt a
vocabulary that uses the 'content' types is a good one -- that's basically
where we ended up. We decided to focus on the type of the contents of the
collection (though we are veering from the DC type list).

I'm not sure how clear all that was, but wanted to share our experiences!

Sarah
-----------------------------------------------------------------------------------------------
Sarah L. Shreeves
Visiting Project Coordinator, IMLS Digital Collections and Content
University of Illinois Library at Urbana-Champaign
Phone: 217-244-7809
Fax: 217-244-7764
Email: [log in to unmask]
Web: http://imlsdcc.grainger.uiuc.edu

At 07:06 PM 8/1/2003 +0100, Pete Johnston wrote:

>OK.
>
>I can't argue that type vocabularies are difficult, and as I say, I'm
>not 100% happy with CLDT.
>
>However I do think that for an effective AP for CLD, we probably do need
>to specify a suitable type scheme.
>
>What if we adopted a vocabulary that used only the "content" types from
>CLDT? i.e. those "corresponding to" (but not the same as!) the DCMI Type
>Vocabulary terms (probably excluding a Collection of Collections and a
>Collection of Services) :
>
>i.e. a term to indicate that a collection is a Collection of Text items,
>a term to indicate that a collection is a Collection of Image items,
>a term to indicate that a collection is a Collection of Sound items,
>etc etc
>
>Pete