Hash: SHA1

On Thu, 27 Oct 2005, Reed Hedges wrote:

On Thu, Oct 27, 2005 at 10:02:16AM -0400, Peter Amstutz wrote:
So what the hell, we can support both.  They are structurally identical
(since they represent the same thing) so I may even be able to reuse some
of the code from the XML loader.

We can have any number of such formats! Having seperate parsers and
parse-controllers or whatever will be a great benefit actually, for example
for loading from things that aren't even files.

Not sure what you mean by "things that arn't files" but you did just get me thinking about the loader/converter framework I have wanted to do for a while for handling non-COD files. What this would be is a unified interface for importing/exporting any relevant supported file format, not just VOS-specific ones. For example, loading ASE or VRML 3D models would automatically convert them into A3DL, loading a XHTML document would convert it into HyperVOS, and so forth. This could be very powerful.

Alright, after I finish XOD, I'll add a converter interface. Then we'll be able to move the functionality of a3dlconverter into a library so that it can be accessed by any application.

On a side note, we've wandered a bit from my original task, which was to rewrite the vosapp configuration implementation to not require flex. I think the tasks we've come up with are valuable, but a little bit of discipline might be helpful in setting priorities. Here's how I see the current work fitting together:

Goal: Make Ter'Angreal the software that enables the 3D web
Current challenges:
   can't make 3d worlds by hand & 3d data import sucks -> no content
   interactive apps need to be written in c++ -> no interactivity
   building VOS is a pain in the ass -> scares away developers

  no content:
   -> text format(s) make it easier to write by hand
   -> xml format(s) make it easier to write exporters from other programs
   -> converter architechture makes it easy to plug in data import and
      have it work seamlessly (every VOS app immediately gains ability to
      read that document format)
   -> maybe finish the vrml geometry importer (reed? :-)

   -> python bindings are underway (they kind of work?)
   -> need to provide highlevel API to support highlevel events and
      queries, such as proximity, collision detection, line of sight

  building VOS:
   -> eliminate dependency on flex to generate scanners & replace with
      TinyXML and boost::spirit
   -> take another look at generating Visual Studio project files
      automatically from automake files (currently there is a tool that
      generates nmake files from automake files)

Finally, I'd like to start thinking about a proof-of-concept interactive 3D application to develop with VOS. I have a vague idea of maybe doing a very simple first person shooter. Just deathmatch, no bots (unless Kao or dmiles want to write one :-), only a few weapons, arenas are just arbitrary VOS worlds. The goal being to test a) the network architechture, b) the extensibility of ter'angreal and c) the feasability and ease of implementing the all rules on the server in Python.

[   Peter Amstutz   ][ [EMAIL PROTECTED] ][ [EMAIL PROTECTED]  ]
[Lead Programmer][Interreality Project][Virtual Reality for the Internet]
[ VOS: Next Generation Internet Communication][ http://interreality.org ]
[ http://interreality.org/~tetron ][ pgpkey:  pgpkeys.mit.edu  18C21DF7 ]
Version: GnuPG v1.4.1 (GNU/Linux)


vos-d mailing list

Reply via email to