Lauren Wood writes:
> The following URL has been discussed a little on www-html, and I thought
> people on this list should also know about it.
>
> http://www.identify.com/welcome/idml-faq.html
>
> Especially the part that says why people shouldn't use META elements.
Actually, in a way they're quite right. The META issue can be viewed
as a point on the "resource-description axis" between full-text
indexing on one side and rich structured cataloging on the other. The
HTML 2.0 META element provides simple attribute-value pairing; a
simple step (hopefully useful) from full-text indexing. This list has
recently been grappling with designing a simple framework that allows
for more potentially richer, descriptive information (SCHEME and TYPE
attributes) to be associated with an object; a simple step above
attribute value pairing. The META element really starts to break down
at this part... There is only so much one can do with a "content-less"
element. The SYNTAX group at Warwick identified this as well, and
proposed (instead of creating a new content-less element with a
bunch'o'attributes) to use the LINK element as a next possible step
towards coupling the object with even a richer descriptive component.
(And yeah, IDML's "open standards" bit gave me a bit of a chuckle as
well :))
Side note:
- come to a consensus on the SCHEME and TYPE issues in HTML 2.0.
and provide examples, and services to create, harvest and search.
- come to a consensus on the SCHEME and TYPE issues in HTML 3.+
and identify a patron to run with this in the w3 groups.
- continue to work on a DTD and LINK formalization (syntax)
- address the protocol transfer issues with these metadata beasts
eric j. miller <URL:http://purl.oclc.org/net/eric>
[log in to unmask] OCLC, Online Computer Library Center, Inc.
[log in to unmask] Dept. of Geography, The Ohio State University
|