Replies to a few points for clarfication:

On Fri, 1 Aug 2003 Pete Johnston wrote:

> Ann said:
>
> > Also is it
> > sensible for Title to be repeatable, especially given that
> > sub-title goes with the main title. Would it be better to
> > have a single mandatory main title, plus multiple optional
> > alternative ones?
>
> Do we want to allow for multiple occurrences of (main) Title to
> provide the value in different languages? Or do we consider a single
> instance of a collection description record to be monolingual?
>
> If we do want to permit multiple main Titles in different languages
> but not multiple Titles in any one language, then we could specify
> that it is repeatable only for that purpose?
>
Ah...I forgotten the multiple language reason for multiple titles.
Your suggestion seems reasonable.

> > Description. I question whether it should be mandatory.
>
> Oh! I guess I'd been working on the basis that a free-text Description
> would be a required part of a CLD! I must admit it feels pretty "core"
> to me, but I suppose it doesn't _have_ to be....
>
> What do others think?
>
It is probably and academic point. I'm sure every collection will
have a description. But there could be cases where it's just
repeating the title.

> > Also are multiple descriptions sensible? Why not just put it all in
> > one description?
>
> Again, I guess the language consideration may apply? But yes, if this
> is not a consideration, then Description probably does not need not be
> repeatable.
>
Again ... I'd forgotten language. You do need repeatable
descriptions for that purpose. But I think it would be better to
discourage multiple descriptions just to indicate paragraphs, lists,
etc.

> > Language. Should free text be allowed as well as controlled
> > vocabularies?
>
> I think I'd have a strong preference for retaining Language with the
> use of a scheme for retrieval purposes. I suppose we could also permit
> Language with free text value in addition to that?
>
> Was your reason for this to cover languages not covered by the
> schemes, or did you want to provide additional human-readable
> information (like "90% of the materials are in English, and the
> remaining 10% are a mix of French, Spanish and Italian")?
>
I think we're talking at cross-purposes here. Your definition of
Language does allow free text. I was questioning whether that
should be allowed. I would think it best to require a scheme to be
used. Additional human-readable information sounds like it should
go somehere else (description?). But maybe there are obscure
and/or ancient langauges not covered by the schemes?

> > Place/spatial. Would it be better called 'spatial' - place
> > implies a location (where collection is) to me rather than coverage.
>
> I must admit that I'm from the school that finds the use of these
> adjectival forms as the human-readable _labels_ a bit peculiar in a
> context like this where these properties are used as "stand-alone"
> attributes of a resource.
>
> I think of myself reading out a record.... The listeners are with me
> when I say, "The Identifier of my collection is 'GB 248 001', and the
> Title of my collection is 'The Frederick Bower Papers'", but when I
> say "The Spatial of my collection is 1914-1918", they look at me as if
> I'm mad. ;-)
>
> Could we use a label of "Spatial Coverage" for the property
> dcterms:spatial in this AP? Or do you feel that conflicts with
> established practice elsewhere?
>
Sounds good. Especially as spatial is a refinement of coverage.

        Ann




--------------------------------------------------------------------------
Ann Apps. Senior Analyst - Research & Development, MIMAS,
     University of Manchester, Oxford Road, Manchester, M13 9PL, UK
Tel: +44 (0) 161 275 6039    Fax: +44 (0) 0161 275 6040
Email: [log in to unmask]  WWW: http://epub.mimas.ac.uk/ann.html
--------------------------------------------------------------------------
NISO's OpenURL: the standard that puts thinking back in linking