I am very torn about this, because while I set great store in the value
of well-formed metadata, the notion that "DC" will develop more rules
about the formation of content seems counter-productive.
As others have noted, the question rapidly becomes "whose rules?" This is
explicitly an international, interdisciplinary environment.
We librarians should not assume that "our" rules will carry the day.
For many of us, DC will be a by-product of metadata created for other
use, in other formats. If DC becomes too prescriptive and we cannot
generate "good" DC algorithmically from that metadata, we will have to
choose between following our community's rules and DC's rules. Assuming
that we are not in a position to simply abandon our existing practices
and infrastructures, we would then have to choose between creating
"bad" DC or no DC at all. Who would that serve?
I would rather we put our effort into encouraging real semantic
interoperability (e.g. don't just stuff any old thing into a Publisher
element, dammit), and gradually work toward increased authority control
as the entire "DC" world gains experience with shared resource description
and discovery.
--Robin
Robin Wendler ........................ work (617) 495-3724
Office for Information Systems ....... fax (617) 495-0491
Harvard University Library ........... [log in to unmask]
Cambridge, MA, USA 02138 .............
On Fri, 23 Jul 1999, James Weinheimer wrote:
> Are we talking past each other here? AACR2 and any other genuinely useful cataloging rules are all
> far too complex for the average user. We should never expect that Joe Q. Public is going to follow
> any useful rules at all, when--and if--he decides to catalog his personal web page.
>
> But when institutions and corporations, scholars and publishers work with their pages, this is
> another matter. I think we can expect something more from them--after all, they've probably gone to
> greater trouble and expense for their pages, vs. some young boy's "Pam Anderson" page. I'm sure they
> really want people to find their work.
>
> I want users to be able to find the (yes!) more important pages without having to wade through a lot
> of garbage. And I think, if we demonstrate it clearly, it's in the interests of the institutions,
> corporations, etc. to follow certain rules and forms. I think they're anxious for someone to take a
> leading role, and this is where we can step in.
>
> At the same time, I believe that it's unfair to let people think that merely marking some text as
> "title" or "description" or "creator" is going to solve the problem of finding things on the web.
>
> Jim Weinheimer
> Princeton University
> [log in to unmask]
>
> PJ Browning wrote:
>
> > Markus Klink wrote:
> >
> > > Unfortunately there are more cataloging needs in the internet than in a library - that's why we
> > > only have a CORE set of tags to describe resources. What I wanted to express is that CORE tags
> > > without CORE rules are almost useless.
> >
> > I agree, but we must be careful that we create rules that fit the entire
> > community. This community includes librarians, archivists, computer
> > programmers and ordinary people. While many of the homepages out there
> > are junk in terms of what I might need as a librarian, they still
> > deserve the same treatment in terms of metadata.
> >
> > AACR2 is great, if you are a library cataloguer. But Joe Q Public has
> > no idea what it is and probably does want to know. The core list of
> > rules should be another dozen volumes but a one to two page list. Such
> > as list personal names in the form LastName, FirstName (as defined by
> > the native language) or dates should be listed as YYYY/MM/DD etc.
> >
> > I would prefer that some kind of name authority/subject authority
> > control should be used, but it's not going to happen anytime soon. It's
> > simply too complex for the general public. Schemes allow those in the
> > know to use LCSH or DDC or whatever tickles our fancy while the general
> > public can stick with the uncontrolled keywords they are used to using.
> >
> > Paula Browning
> > University of Oklahoma
> > GA--Bizzell Library
>
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|