Re: TAG Comments on XHTML 2.0 and HLink

Steven Pemberton writes:

> That the TAG doesn't yet understand how XLink gets in the way of combining
> with other specifications will be the source for much future discussion,

It's exactly this aspect of the TAG's decision that causes me the most
consternation.  The decision seems to have been taken on the grounds
of making the W3C specs consistent with one another rather than on the
basis of a deep understanding of the technical issues involving HTML and
XLink.  That is very desireable *goal* of the W3C, and was a reason for
the TAG's formation.  Unfortunatately, (to pick up on Didier Martin's
point)  coherence doesn't come from a committee vote, but only 
after lots of thought, experiment, creativity, and flexibility by
all concerned.  

As I understand the history of mathematics, for example, it was a
very long and hard struggle for any particular discipline to find a
set of "axioms" that were both powerful and consistent.  It can't be done
by fiat;  I'm reminded of the (possibly apocryphal) story of
the Kentucky legislature voting to set the value of PI at exactly 3.14 because
the rest of the digits were too, uhhh, irrational.  Or the story that Gauss 
didn't dare to publish his exposition of non-Euclidian geometry because 
he feared the wrath of of the mathematical community that valued "coherence"
over more profound truth. 

The W3C simply can't pretend that the first Recommendation 
in an area is sacrosanct, or that local consensus is the
path to global truth.  Local consensus codified in a Recomendation is
a way for competing companies and researchers to move forward in a way
that minimizes the gratuitous differences among their approaches, maximizes
the re-use of components, and minimizes the pain to our "customers." 
That's not "global consistency" but it beats hell out of chaos!
 
The TAG has a role to play in guiding WGs to operate in a way that encourages
global consistency, certainly.  Still, this has to be a two-way street -- 
old Recommendations may need to be refactored, current WG's more strongly 
encouraged to work toward a global consistency -- and this will require
hard work by the subject area specialists in the wG as well as guidance
by the TAG.  

We need to work for conceptual integrity, not foolish consistency.

[NOT speaking on behalf of the Web Services Architecture WG, only expressing
my personal opinion]

Received on Thursday, 26 September 2002 10:11:59 UTC