Well, I've only just discovered the list after browsing the
Canberra page sent me by Renato Iannella ... too much time writing stuff
and not enough surfing ... ;-)
For the purposes of thread handling, I'll send several messages ...
I wrote an HTML/VRML DC generator a while ago now at
http://vancouver-webpages.com/VWbot/mk-dublin.html
While the provenance was perhaps vague requiring two clicks to
find an email address, I hadn't had any feedback until recently
from Renato though I now see it mentioned in the meta2 archive.
Is this considered at all useful, and should I update it?
Is the Internet draft the most recent consensus ?
I alos wrote the meta-generator at
http://vancouver-webpages.com/VWbot/mk-metas.html which gets
a lot more feedback. My own search engine (searchBC) uses
the keywords tags and stores description as does AltaVista, and
also notes some of the others such as language. Some of the
http-equiv ones (charset, language) actually do useful things
with Netscape and are defined in the HTTP spec. I was reluctant
to include the DC Subject instead of Keywords as it is currently
ignored (I believe) by major search engines. However, if it's
felt that DC is going somewhere I can move more tags into mk-metas
and drop the more obscure/informal rating and revist ones. There's
a danger of confusing people, though, with too much detail.
Description and Keywords I think they can understand.
I made some attempt to cross-pollinate the VRML working group
with metadata ideas a while ago. There was some interest, and there
is some support for information nodes in VRML 2.0, but feedback
has been limited.
Andrew Daviel mailto:[log in to unmask]
TRIUMF voice: 604-222-7376
4004 Wesbrook Mall fax: 604-222-7307
Vancouver BC http://andrew.triumf.ca/~andrew
Canada V6T 2A3 49D14.7N 123D13.6W
|