Re: TOOLS: Virtus Walkthrough Pro 2.0 and VRML 1.0

Andy Norris ([email protected])
Tue, 23 May 1995 10:52:59 -0400


At 02:19 PM 5/22/95 EDT, Mark Waks wrote:
>Tony mentions:
>>Amen to that. I've been talking to the IMF about this-- how are we going to
>>get VRML out to the masses when the spaces live in 10Mb files?
>
>I will point out again that this is why we need a mechanism for
>distributing and/or caching common items -- spaces are a *lot* faster
>to download if you've already got some of the building blocks locally.
>We still haven't addressed this point in any sort of coherent way, and
>really ought to by, at the latest, the time of rollout for 2.0...

Am I right in thinking that you can specify a "file" URL in a document that
is not itself a "file"? IOW could http://www.mystuff.com/home.wrl include
a WWWAnchor with the URL "file:///c:/vrml/objlib/chair.wrl"?

If so, you could distribute a common object library for downloading. I can
think of two ways to do this (without coming up with a standard that's
new and time-consuming to develop):

1) Have a standard must-be-here PC directory, a standard Unix directory
(etc), and different versions of files for each of these platforms

2) Use a HTML form to get the path before entry into the WRL, and
CGI-generate each WRL with this path.

Either of these sucks away server resources (space or cycles) to save
transfer time and add speed on the client, but it would make VRML usable
for certain things that might otherwise be completely impractical.

Also, I'm not saying there shouldn't be a new standard for this--perhaps
there could be a standard "file:///object/..." path in the URL, and a
preference file in the browser that substitutes the correct directory.
In the long run, a new standard would be a lot easier.

I'm just saying that I think either of the two above methods could be
implemented this afternoon with no particular difficulty, and without
writing new specs or code.

--Andy
http://www.texas.net/users/andyn