Validation (was Re: Apology.)

Paul Burchard ([email protected])
Sat, 25 Nov 95 23:19:53 -0500


Now back to Sandy's point: formalized validation. Why do we need
it? When do we need it? Some people have said "years from now".
But based on my involvement with HTML, I disagree.

The _reason_ for formalized validation is to keep the cost of
market entry low for new participants, thereby maximizing the
vitality and growth of the market, and the competitive benefits for
consumers. As Amanda Walker wrote about her company's entry in the
HTML market:

In <http://www.acl.lanl.gov/HTML_WG/html-wg-95q2.messages/0143.html>:
# We spent about twice as much time getting our parser to
# swallow common illegal HTML than we did getting it up and
# running on legal content.

The _time_ to start formal validation is as soon as you achieve any
area of formal standardization -- like, say, the final VRML 1.0
spec. You can't wait until the entire "virtual world tools
industry" is mature. "Mature" is just s eupemism for "locked-in
oligopoly", and at that point you can't magically get your vital
market back.

Standards are foundations that focus the chaotic energies of the
market on building up the next level of a technology. They open new
territory for deployment that would otherwise be left undeveloped
due to eternal squabbling or languid domination over the previous
levels.

And the point is that unless backed up by validation -- preferably
as prominent and "blessed" as possible -- standards are meaningless.
Witness all the HTML browsers that now shamelessly claim to do
"HTML 3.0" (a bold lie). The W3 Consortium is belatedly working on
validation, but much of the damage to the value of the HTML standard
has alesady been done. And the W3 Consortium itself has lost a
large amount of clout by coming onto the scene so late, which
reduces the value of its HTML validation efforts even further.

If you care about the long-range vision of VRML, you need to think
carefully about the role that validation can play in raising the
odds that we can actually accomplish some fraction of our grand
plans.

--------------------------------------------------------------------
Paul Burchard <[email protected]>
``I'm still lsarning how to count backwards from infinity...''
--------------------------------------------------------------------


  • Next message: Scatt: "Fountain 0.9"
  • Previous message: Paul Burchard: "Validation (was Re: Apology.)"