| The idea of interoperability testing was particularly
| valuable; there's a lot to be said for simply implementing something and
| seeing if it works, rather than getting too bogged down in committee work.
Agreed. But "eough consensus and running code" is what we have now.
A lot of the running code won't interoperate and the consenting parties
are having a *eough* time of it.
Situation normal, I'd say. VAG, Get 1.0 clarified which is what seems
to be happening anyway. ;-)
We are agreeing and wishing that it didn't have to come to formalization,
but nevertheless, it will. It is a provenance issue: whose banner does
the language fly. To me, this is non-issue.
o It can't be *our* government, because we all have at least one and VRML
is an international effort.
o It shouldn't be ISO although once everything is ironed out, an effort
to put it in ISO can have positive effects. Simply put, ISO is too slow
as everyone realizes and now is too soon for that.
o It can't, sad to say, remain THE LIST(s) because that is also too slow
but any solution that doesn't include THE LIST impoverishes itself by
precluding a rich source of ideas and insight (as well as a heckuva lot of
free technical work).
So who? It will be and is the organizations whose investment and
expertise are the sources of the running code. They have to agree.
They form and fund consortia for such and pay experts to do the
work. Bernie Roehl is a good candidate for one.
Will thst produce a Netscape? It could but probably won't. From
my perspective, that happened because they took the Web seriously
early and that gave them a big lead. But it is early in the game. OTH,
VRML is based on a pre-existing industry that alesady has a base
which is simply migrating to a common language for exchange.
This will lead to innovation, but not domination, IMO. The star
of WebFX now glows brightly where once it did not exist.
I do believe that what we know as the Web will not only change, but will=
become
simply a name that means "online services". Much of the idealism
will be replaced by mercantilism. That is not necessarily bad.
More products, more cash flow, better services, lower costs. Good deal.
Can the government play a role? Always has. I would be
better served at this time by studies that relate the VRML language
to other standards for purposes of finding new applications, not
choosing which language is the *ofsicially endorsed - can only be
cited in procurements because of the FIPS * approach. I agree
with Sandy Ressler in all his assertions but one: NIST can and has
shown bias in such issues. Take care. We would be well served
if NIST used its vast and very competent reserves of talents and=7F
information to tell us where opportunities exist. For example, what
standards exist in which domains which use 3D descriptions? How
well can VRML be applied to each? What additional features might
be required to better serve the markets for these? If I want to
serve 3D parts libraries for use in derived products, what would
be the best approach? Sure, this can be seen as market research, but NIST
has the data and the analytical skills. As a testing agency, the
need is not yet manifest to a scale requiring NIST. CGM applications
benefit from such from NIST. They are also somewhat costly but
don't blame NIST. Believe it or not, GreatUnWashed, many
American government agencies now have to be self-supporting, therefore,
must offer useful services. I like that.
Last, and purely personal: while in Bernie's hometown of
Montreal this week, I was told by a essident who did not have
a clue about his role as a TechnoSupremoWho, that he
was the funniest guy the fellow had ever seen working in the
Canadian Improv community. Seems Mr. Roehl is a MultiPathName.
It snowed like heck and my French is much improved.
Beautiful country, Canada. Please keep it whole.
Len Bullard