----- Original Message -----
From: "Cal Lee" <[log in to unmask]>
To: <[log in to unmask]>; <[log in to unmask]>
Sent: Wednesday, April 26, 2000 10:39 PM
Subject: support for external metadata?
> Please forgive all cross-postings.
>
> I've been asked to provide some advice on the implementation of Dublin
> Core for the description of state government Web pages.
>
> As I understand it, there are several good reasons to create and
manage
> DC metadata in a separate file that's associated with a page through a
> <LINK REL="metadata" HREF=""> reference within the page, rather than
> coding all the metadata directly into the HTML as META elements.
>
> This is the approach taken by D-Lib Magazine
> http://www.dlib.org
> and the DC site itself
> http://purl.org/dc/
> (though their implementation in XML is a bit different)
>
> My question is related to support for such an approach by major server
> search and indexing software. I would specifically like to know if
> anyone has experience trying to use either Ultraseek Server or MS Site
> Server to crawl, index and provide access to pages that link out to
> external metadata.
Hi Cal
We use the indexing component of MS SiteServer to provide a search
facility for our Exploit Interactive web magazine. The metadata is
stored in external files, and is pulled in using a server-side include,
and transformed to <META ..> elements using a simple ASP script. We can
easily update the script to provide alternative representations, e.g.
RDF or other <META> descriptions (it's not clear, for example, how one
should use DC to describe the issue number or the funding body of the
project covered by an article - if, indeed, one should use DC).
A description of our approach is available at
http://www.exploit-lib.org/issue5/metadata/
A poster based on this will be given at the WWW9 conference in
Amsterdam next month - see
http://www.ukoln.ac.uk/web-focus/papers/www9/resource-discovery/
Note that some minor hacking is needed to get SiteServer to
understand DC (it doesn't like full-stops in META NAMEs). A description
is available at:
http://www.ukoln.ac.uk/web-focus/papers/www9/resource-discovery/tech-spe
c/
This approach isn't quite as you describe, but I thought it may be of
interest to you. get back to me if you have any questions.
Brian
--------------------------------------------------------------------
Brian Kelly, UK Web Focus
UKOLN, University of Bath, BATH, England, BA2 7AY
Email: [log in to unmask] URL: http://www.ukoln.ac.uk/
Homepage: http://www.ukoln.ac.uk/ukoln/staff/b.kelly.html
Phone: 01225 323943 FAX: 01225 826838
> This is distinct from the (also important) question of whether the
> software can correctly parse and index XML and RDF from that file once
> it gets to it. If the only approach currently supported by the major
> products is to for externally linked metadata files that are saved as
a
> set of HTML META elements, that seems like a reasonable approach.
> Though I would think markup in XML would be higly preferable.
>
> Any practical advice on this matter would be greatly appreciated.
>
> =======================================================
> Cal Lee
> Electronic Records Project Archivist
> Kansas State Historical Society
> Phone: 785-272-8681, ext. 280 Fax: 785-272-8682
> http://da.state.ks.us/itab/erc/
> http://www.kshs.org/archives/recmgt.htm
> "Obsolete power corrupts obsoletely."
> - Ted Nelson
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|