Le 6 janv. 2007 à 12:24, Charles McCathieNevile a écrit :
It's a typical metadata/semweb scenario. You have some kind of
useful data, but
different people have different kinds and relying on one particular
version
fails as many people as it helps. (I like RDF because it was
designed to provide
useful answers in the face of lots of partial information).
Indeed and agreed.
Having written a lot of stuff in the traditional academic world
that is slowly
crawling to the Web, I don't see a lot of ISBN references.
Hear hear. Same here, when I was doing astrophysics research.
Though if you do http://www.google.com/search?q=astrophysics%20isbn
Résultats 1 - 10 sur un total d'environ 933 000 pour astrophysics
isbn. (0,40 secondes)
Another example
http://www.citeulike.org/user/ihuston/article/869979
With the tool (back end web server in this case) giving the BibTex
reference. :) in a form.
Useful in theory, but not a good basis for any serious work directed
at the people who are actually going to read what you wrote.
Agreed. It depends on the context of the usage.
What I usually do not want on the Web or in a language like HTML, is
to prevent a minority to use features of the languages, specifically
when these do not require anykind of actions (implementations) from
certain classes of products. (example with role and about attributes
for desktop browsers).
--
Karl Dubost - http://www.w3.org/People/karl/
W3C Conformance Manager, QA Activity Lead
QA Weblog - http://www.w3.org/QA/
*** Be Strict To Be Cool ***
