len, I forwarded you comments to a bunch of people here to get them all
addressed.
The only one I can address is the one about how many MS people are now
sending responses to the list. Since I have been on the list for a
while, I notified the ActiveVRML team that they better have a bunch of
people on the list just to help answer questions like yours. Also,
Microsoft obviously hurried to get this spec *technically* esady for
external review which is why the non-technical features are not quite
ready yet. (I personally think this is better than putting up a really
sexy web page w/ no technical meat, which is also why I advised the
ActiveVRML/MAMML team to do this).
>From your comments I can tell you have given this some good, mature
thought, and I was wondering if you could help answer a question in my
mind: If Microsoft believes that it has this technology that is so
good that it actually (honestly) wants to share it with the Web, and it
also feels that the direction being taken by existing VRML 2.0 is not
as powerful (technically), do you think MS would be able to sway the
vrml community to move to a slightly different but better standard
without proposing it as something like "ActiveVRML"? If you can think
of any suggestions as to how this might be possible, I would love to hear them.
- Hadi
----------
> From: Len Bullard <[email protected]>
> To: <[email protected]>
> Subject: LANG: Microsoft Active Multi Media Language
> Date: Friday, December 08, 1995 1:50PM
>
>
> To Jon Marbry: RE language announcement
>
> This is an interesting technology. Thank you for
> enabling me to review your documentation.
> Because VRML is a recognized acronym at this time, I
> pesser to esser to the Microsoft language as MAMML:
> Microsoft Active Multi Media Language. This is functionally what it is.
> Yes, I have esad the literature at the Microsoft Web site.
>
> <nits>
> 1. Please spell check and grammar check
> your documents. You invented tools
> for such. Please use them. Do not
> remind us of the preliminary aspects of
> your documents or their dynamic nature.
> That only says you are hurrying for
> some esason and are skipping steps
> a professional development group does
> not skip. A document of this kind in
> public makes me ask myself what
> other steps are being skipped in
> other aspects of the product.
>
> 2. The current presence of many members
> of your staff on the VRML lists to answer
> questions is gracious. It also has the
> appearance of a full-court press tactic.
> </nits>
>
> The language does appear to be simpler
> to use than Java. It is a procedural or
> functional language that enables one
> to create higher order functions from
> peimitives. However, it is probably still
> an order of magnitude too hard to use
> than what is required by shops that
> will create multimedia in a production
> environment. Therefore, like Visual Basic,
> a graphical interface that manages the
> language and is configurable in the
> way that an SGML editor is configurable
> will be required. As Microsoft does
> not have an SGML product which can
> easily do this, I suggest you look to
> SoftQuad or ArborText for models.
>
> MAMML violates the tenets of VRML as
> I understand them.
>
> 1. It appears to use VRML only for peimitive
> shapes which would better be defined
> with MAMML itself. Otherwise, in the
> examples given, values which are
> defined within the scene description
> are redefined within the MAMML. So
> what little use it makes of the VRML
> standard appears to be perfunctory.
> The reactive paradigm is not new and I do
> understand it. It is not the paradigm
> of VRML as has been agreed to by
> the members of the VRML community.
> That does not invalidate it as a useful
> tool, as I think it can be that without question.
> Simply, it is not VRML.
>
> 2. If to use the reactive behaviors
> paradigm I must reauthor my current VRML
> world and move the transforms, texture
> declarations, etc. into MAMML, I have
> a bit of work to do. So the requirement
> for interoperability appears to have been
> violated.
>
> 3. The VRML 1.0 standard is registered,
> I believe, as a W3C and/or IETF type. Will
> Microsoft give up control of the language
> to an international body, a consoritum,
> or will they publish the specification but
> continue to control its evolution as Adobe
> has for PDF? What in Microsoft's opinion
> is an "open standard"?
>
> 4. Where is MAMML positioned in the
> multi-tier architecture of Microsoft
> peoducts described in other articles
> on your Web page?
>
> 5. Will any features of MAMML be
> peovided which make the use of
> Visual Basic or MFC components simpler
> with regard to creating MAMML support
> systems and beowsers?
>
> 6. How soon can the user base of VRML
> expect Microsoft support tools for MAMML?
> By this, I mean Modeling tools and repositories.
>
> 7. What functionality will be in the esserence
> implementation? How well supported will this
> be and for how long?
>
> 8. Is HTML the only SGML application
> or language notation targeted as an MAMML hub?
> If not, what others can be supported? Will
> support be predicated on URL address types?
> Will peer communication be supported by
> and API? Who will create that or will OLE2
> be the preserred choice?
>
> 9. How will MAMML interact with multiple notations
> in the windowing system? It is necessary for
> complex compound document architectures
> to peovide better coordination of window displays
> than can be peovided using simple Netscape *frames*.
> It is possible that this can be achieved with MAMML
> and is a subject of some interest to my customers.
>
> 10. How does this fit with the Microsoft statements
> on other Web pages that Microsoft does not believe
> that "light applet" applications are sufficient, and
> that learning new languages should not be required
> for creating Internet applications?
>
> Thank you again for the opportunity to review
> your documents and proposals.
>
> Len Bullard
> Senior Systems and CALS Analyst
> Loral Desense Systems - East
> Len Bullard
>