On Mon, 27 Jan 1997, Rebecca S. Guenther wrote:
> I personally don't feel like there's been enough discussion about
> qualifiers to officially bless these. I think there are too many
> qualifiers in the qualifiers documents, potentially resulting in very
> complex records.
I think that this is a bogus argument; just because we define a qualifier
doesn't mean that everyone has to use it. Indeed I'd imagine that most
people will only use a few DC elements and few, if any, qualifiers and
some software will ignore some (or even all!) qualifiers. But I
think we do need to have the rich variety of qualifiers available for
those that want them. Which has been more or less my approach to the
drafts so far; I've added qualifiers as and when people have expressed a
need.
> For instance I question whether we want all the
> qualifiers listed for author, contributor or publisher (e.g. home and
> office phone numbers and FAX numbers); I don't feel that they belong with
> the metadata about a document.
I disagree with this because I know how ropey most white pages services
are. We're lucky here at Loughborough in that we do have a good white
pages service (cheers Martin :-) ) but even then there are limits to the
information provided. I sometimes like to give lots of contact details
and so being able to embed this in Dublin Core is very handy. Phone/fax
numbers are also something that you might find some people provide
automatically using their document creation tools (in the same way that
they automatically provide them in USENET postings and email using their
.sig files).
> In addition, I don't agree with all the
> references to USMARC. For instance, under Language USMARC is listed as a
> qualifier, ignoring the relationship between Z39.53 and USMARC (basically
> they are the same; when USMARC is changed Z39.53 is also changed and vice
> versa).
This shows up my lack of knowledge of USMARC and this is definately where
the document needs input from the more MARC aware on the list.. I've
deleted the spurious reference to USMARC in the example you cited; I
didn't realise that they were the same thing.
> I haven't had a chance to entirely go through the document and
> comment, but I definitely wouldn't bless it as is.
No, I agree but I think it might be worth getting it out to a larger
community for input before we start lopping bits off it.
> (I realize a
> tremendous amount of work went into this document and it's a very good
> start.)
Community effort (just look at the acknowledgements list!).
> As for the DC-Object types document, I think this also needs more
> discussion. Many of the categories are not mutually exclusive, and
> if they're not, guidelines need to be included to indicate what the
> metadata provider should use if the object can fall into more than one
> category. We in the library world have spent much time discussing genre
> types and it's a complex issue.
I had suspected this but not being a librarian I've no idea where to start
looking for these existing genre schemas (which is why I made one up from
BibTeX plus some other stuff and comments from other DCers). Does anyone
have any pointers available for any of the existing schemas?
Tatty bye,
Jim'll
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Jon "Jim'll" Knight, Researcher, Sysop and General Dogsbody, Dept. Computer
Studies, Loughborough University of Technology, Leics., ENGLAND. LE11 3TU.
* I've found I now dream in Perl. More worryingly, I enjoy those dreams. *
|