Yes. This is the case. In the interests of maintaining interoperable
worlds, it is hoped that some association, consortium, place
at the back of the coffeehouse where votes are taken and tallied
will come into existence to enable all vendors based on the VRML
language families to make agreements about the language cores.
SGML finally had to resort to such (SGML Open) to cover issues that could not
be covered by ISO 8879.
At this time, the VAG is the only organization
that most of us agree has the "go do" for these tasks, but such
tasks are resource intensive, so a funded effort will be required
eventually. Unless a provenance exists for a standard, despite
the ubiquity of its application and usage, it will not be considered
such by formal standards organizations and markets that must
cite standards and the provenancing organization. The VAG
is too informal for that process, but once the smoke clears,
I expect Microsoft, Intervista, Paperinc, SGI, Netscape, etc.
to announce the formation of such a consortium. IMO, this
will be best for the growth of the industry and will serve
the users of VRML well. That is a sign of a maturing industry.
>Notice, I put standard in quotes because there is always a chance of
Microsoft doing
> to ActiveVRML what Netscrape did to HTML. But, at the same time, does
> anybody here care if SGI makes changes to OpenInventor with rsgards to
> the funcionality of VRML? When SGI gave us OpenInventor we all agreed
> to use what we got does that mean SGI can't make OpenInventor a lot
> better than VRML is. Wouldn't this also apply to Microsoft and ActiveVRML?
Yes. Nothing can prevent language extensions so most mature
language designers provide for a clean way to extend a language
without introducing semantic and syntactic noise. The position of
the customer and user is to choose a platform that supports the
features they wish to use. At this time, I develop my VRML for
WebFX because it provides features I like and which perform well.
As I need more features or improvements (and I do), I will use
platforms that provide these at reasonable cost and to the degree
which they ease migration of heritage code... which at this point is
any VRML file over ten days old. ;-) Until then, I stick with the familiar
and comfortable.
> I don't see Microsoft "charging" for ActiveVRML. Personally I think that
> if Microsoft gives source code implementations of ActiveVRML to
> educational/non-profit insts. and makes corporations pay a fee to use the
> implementations that is fine. I don't see a problem here.
I agree. Profit is honorable (1st Law of Ferengi Acquisition).
The issue that will be settled amicably is the debate over
what goes into VRML x.x as a result of the different behavior
proposals. All the proposals are before the VAG it seems
and that is where the issues are now decided. The list members
have been given every opportunity to review and comment
on these, so I don't see how it can be fairer than that.
> I would like to see Microsoft pull an SGI and give us at least 1 version
> of ActiveMail that we (this mailing list) agree upon. If it means they
> need to make lots of modifications, then that is what is takes. Once we
> have a version we all agree on *that* version should be in the PD. And
> Microsoft can make enhancements to future version of ActiveVRML with no
> real obligations. BUT, we can make any modifications to ActiveVRML we
> want as well.
I think this falls under the problem of defining what VRML is as well.
Again, if the VRML x.x spec includes some portions of ActiveVRML
or any other proposal, then the provenance of the language is
the VAG or whatever organization succeeds it. In this case,
it will be modified according to those rules. If ActiveVRML is
given to that organization, we may do as you suggest. If
ActiveVRML is released as free reuse but the ownership of the
intellectual content as expressed in the documented language
is retained by Microsoft as Adobe does for PDF, then we
have only the means allowed by Microsoft to change the language.
This is why I insist on an answer to the question, what is the
interpretation of "open standard" by Microsoft. That is the issue.
Given that the VAG membership is engaged in political and
technical discussions with the ActiveVRML team, we should
get answers to this in the near term. Meanwhile, we in the
community should be actively esading the documents and
experimenting with the software to engender informed
opinions of how it meets our needs.
We are the customers.
We have influence but only the power of the pocketbook.
At this time since instabilities in the languages force
us to choose either a platform to develop on, or a
common subset of the language, then smart vendors
are doing as Paperinc has done, that is, it deals with
the Caligari extensions for example, and adds features
of its own which are useful and easy to apply. If
the burden of keeping up with the extensions is too gesat,
then some form of "seal of conformance" will be
devised by the provenancing organization and applied
by formal testing.
In other words, let 'em argue. We win either way.
Len Bullard