Tim Berner's key note summary.

[Murali Mani <mani@CS.UCLA.EDU>]

This is the official announcement of xml schema as a recommendation from
TBL from www10, as noted by a friend of mine attending the conference in
Hong Kong.

the following is purely my personal opinions (from a person *strongly*
biased against XML Schema) --

Anyway, I especially note the "group with theoretical objections". From a
personal point of view, i think this is not something to be taken lightly
or to be brushed aside as just another group. I think, from time
immemorial, there always has been a core group of mathematicians behind
every technology - I think they are the people who can really analyze a
technology and find its pitfalls and benefits, which as users we can
safely start using. I think the decisions made by XML Schema ensured that
the group of mathematicians it started off with dropped out on the way. In
short, i do not think I am too far off when I say XML schemas is
technology which mathematicians do *not* approve of. I would like to ask
the opinions of mathematicians about how much of a fix they are in with
this recommendation.

I also take special note of the fact that XML Schemas acknowledges the
input from the formal language theory community -- i am very sure it must
be for a noble cause, and do not wish to complain about it, but i think
formal language theory people will say that XML Schemas defies formal
language theory. Anyway, the future *definitely* looks interesting.

My two cents worth of thought in this matter is: Rather than taking a
forward step, man has taken a negative step with respect to data modeling.

<warning>speaking for himself only</warning>

regards - murali.

---------- Forwarded message ----------
Date: Thu, 3 May 2001 17:43:36 -0700 (PDT)

I listened to TBL's keynote day before yesterday - the formal "announcement" of
the XML schema recommendation happened within the first couple of minutes of his
speech.

He did concede/acknowledge the presence of a number of detractors to the
current recommendation but indicated (paraphrasing) "that it was all part
of the game".

  By the way, there seem to be a couple of different communities that have
bones to pick with this recom. One is of course the "group" with
theoretical objections. Another is the RDF/semantics/ontologies group
(basically people with AI or knowledge-bases background) that believes
that even though the current spec might be okay for describing documents,
it just doesn't fit the bill for data/information exchange.