When do computer languages die? The case of VRML, that didn't

Replies to This Discussion

Permalink Reply by Chris Thorne on April 15, 2009 at 7:23am

Hi Jordi. yes it is an interesting phenomenon. VRML is a good language and still addresses a lot of application requirements. However, no one vendor's VRML plugin or application reached an acrobat, flash or quicktime level of cross platform availability and public acceptance.

Re VRML/X3D, my feeling is that if X3D brought something truly compelling that was not in VRML, like direct support for portable networking and multiuser then the story would have been quite different.

I think that the new Khronos 3D initiative has the best chance of becoming the flash/acrobat of 3D because it is based on two pretty universally supported languages: javascript and opengl. Even when a large industry player brings out a competing "standard", as so often happened in the past, an API like C3DL is likely to withstand such an attack, in my opinion, because javascript and openGL form such a strong foundation across all the popular platforms.

▶ Reply to This

Permalink Reply by Hiperia3D - Jordi R Cardona on April 29, 2009 at 11:53pm

The reasons for the success of Acrobat, Flash and Quicktime is having companies like Apple, Macromedia or Adobe behind them, that know real marketing and do what is required. Good things don't sell by themselves.

Also, they did what their times needed. VRML came too soon.

For X3D I agree partially. XML syntax may be good for some applications, for x3dv files are much more understandable and take the good part of vrml. Also, multiuser was a popular demand that was ignored.

See, http://www.cs.man.ac.uk/~toby/writing/PCW/vrml3.htm
here it says:

"VRML 1.0 and 2.0 are single-user systems. When you visit a world (and there's an excellent list at vrml.sgi.com/worlds), it's downloaded onto your machine, and everything happens locally. The challenge now is to make VRML truly multi-user so that it will support shared virtual environments. There is intense activity to define what VRML 3.0 will do, and how objects such as avatars and bots will be represented and programmed."

What happened with Living Worlds and all this?

I think it was a great mistake to call it X3D, and not VRML 3.0. We don't know even if call it VRML/X3D or X3D, because X3D is much less popular. COLLADA is being much more widespread, and many add support for it, while companies that added support for VRML are not adding them X3D.

If it had been named VRML 3.0, those same companies may have continued the support, and others may have joined.

There's still a way, make extensions for X3D for those things and call it VRML 3.0 :)

And for Khronos and O3D, they are much lower level standards (although not too low, but they are). They're for low-level programmers and not for the big public. There is no comparison to Flash. If the designers that make flash animations and use its scripts had to code in low-level libraries, they won't use it.

Acrobat and Quicktime are for non programmers. The ease of use if a big factor.

▶ Reply to This

Permalink Reply by Chris Thorne on April 30, 2009 at 1:18am

Reply by Hiperia3D - Jordi R Cardona 1 hour ago
"The reasons for the success of Acrobat, Flash and Quicktime is having companies like Apple, Macromedia or Adobe behind them, that know real marketing and do what is required. Good things don't sell by themselves."

That's not all. These companies a very good at interactive design - creating something that is very usable and even a pleasure to use. That said, I was not really comparing VRML to those products, just their broad acceptance.

"...
See, http://www.cs.man.ac.uk/~toby/writing/PCW/vrml3.htm
here it says:

"VRML 1.0 and 2.0 are single-user systems. When you visit a world (and there's an excellent list at vrml.sgi.com/worlds), it's downloaded onto your machine, and everything happens locally. The challenge now is to make VRML truly multi-user so that it will support shared virtual environments. There is intense activity to define what VRML 3.0 will do, and how objects such as avatars and bots will be represented and programmed."

What happened with Living Worlds and all this?
"

A good question. Networking and multiuser capability support in the language spec has not progressed since 1997, when Gavin Bell defined the createVrmlFromUrl and LoadUrl calls.


...
"
And for Khronos and O3D, they are much lower level standards (although not too low, but they are). They're for low-level programmers and not for the big public. There is no comparison to Flash. If the designers that make flash animations and use its scripts had to code in low-level libraries, they won't use it."

yes that's right. the thing is, much like Flash I believe, we will need tools to help the general public build content for things like O3D.

"Acrobat and Quicktime are for non programmers. The ease of use if a big factor."

I did not know it was easy to write quicktime content by hand - like in a text file as we often do with vrml. I have only ever used a tool to generate quicktime. The runtime, I agree, has a high ease of use. As I said above, good interaction design is a key for the mass market.

Perhaps this is something that is harder to do with true 3D and that is one possible reason why 3D systems in general have difficulty gaining mass markets.
Demand on resources / performance would be another.
A big one: one of the presenters at muvemoot08 said they lose some 30% just at the installation stage. Getting ppl to install plugins to browsers is hard. That's a big advantage for those vendors that provide 3D capability supported directly by the browser. And guess what, vendors like google and khronos will have their system supported directly in their browser of choice. The rest of the market may have to follow or be left behind...

▶ Reply to This

Permalink Reply by Hiperia3D - Jordi R Cardona on April 30, 2009 at 7:35am

"I did not know it was easy to write quicktime content by hand "

I did not mean that. I mean that there are graphical tools to build things without programming, and some are free or open source.

Anyway we have 2 good projects now for X3D. One is X3D-Edit. It only supports the XML syntax, although my Module Suite for X3DV and VRML is compatible, and Don Brutzman has added it to their page as a compatible plugin. We should work more on the compatibility, something I hope to have the time soon.

Also, Xj3D. Xj3D is a Java X3D browser for all the encodings, and also for VRML. It can also run as an applet.

But we may need maybe a browser plugin, at least for Firefox, and may be some possibilities in using Xj3D with JavaFX to build RIAs that use X3D.

Xj3D is a great project, and may suppose big things in the future. It may be the base for a solid and powerful plugin for X3D on browsers.

▶ Reply to This

Permalink Reply by Hiperia3D - Jordi R Cardona on April 30, 2009 at 7:41am

"one of the presenters at muvemoot08 said they lose some 30% just at the installation stage. Getting ppl to install plugins to browsers is hard"

Yes, some have find a solution. For example, look 3DXplorer, built in Java.

O3D has a plugin, in fact you have to install it. Khronos will have it too. But most of people have already Java installed.So in practice, that is plugin-less.

Xj3D applet may be improved to something similar. These days I also think in JavaFX mixed with it, many RIAs may be born using Xj3D, and they may be executable on internet, in the desktop, or in mobile devices.

▶ Reply to This

Permalink Reply by Larry Rosenthal on April 23, 2009 at 5:32pm

and then google steps into the fray...o3d....
and it already works on in ie7

c3

▶ Reply to This

Permalink Reply by Chris Thorne on April 28, 2009 at 11:57pm

Yes, and nice to see they are still interested after unlively.

I put o3d in the same category as the Khronos one, javascript/low level graphics API that runs in the web browser. Only they will support both directx as well as opengl off the bat. The o3d installer for Mac was "not recognized". oh well, early days.

▶ Reply to This