Re: VRML - The next step?

brandon@avon.declab.usu.edu
Mon, 03 Apr 95 23:08:45 -0600


< but as the next step in the MUSH paradigm. I seriously believe that we
< should start looking at making a 3D interactive MUSH based on already
< available code for various text-based MUSH schemes that already exist.

I really _really_ dont want to bring this to a religious war about which
virtual environment system driver is best, so I'll just state that there
are a pleathora of drivers which would work just as well as MUSH (such
as MOO, LP, CoolMUD and ColdMUD, to name a few).

< ( An aside: Many people view MUSH as just a game; my experience has proved
< otherwise. I helped set-up, administrate, and create a MUSH devoted to
< interactiveeducation over the network; it was called ViBES. Students from
< across the Internet could log in, and attend virtual classes taught by
< teachers which were either programmed, or actually logged on to the MUSH
< through their computers. What we learned from the project: the idea is,
< basically, a very good one - but with a very serious flaw: text based
< learning is difficult, if not impossible; it is difficult for a student
< to respect a teacher he only knows as a MUSH object :-)

I agree, but I would apply the same reasoning to any MUD, not just MUSH.
(I know everybody and their dog's dog has made an acronym for this, but
I like none of them so far (smile), although my current term of preference
is "virtual environment system/server". It may not be very catchy--but
its the best the local group of us could find, I'm always open to better
referencing names...) (Like all good humans, we must name an item before
any advancement or discussion can take place 8)

< I know this posting is totally generic, but this is purposefully so. I've
< been working on a very specific specification on an extended
< MUSH+VRML system which I intend to make public soon; until then, I'd like
< to hear everyone's thoughts on this exciting topic.

Since the creation of VRML I've been beating around designs and recently
implementing them for a VRML virtual environment server. The driver I'm
using is ColdX (descendant of ColdMUD which descended from MOO and CoolMUD).
I believe starting with a clean slate is the best way to go, but others
may disagree (shrug).

The database saves "text" in a "compiled" form online, so you can
output in html, plain text or some other type of markup (i've been thinking
of just tweaking html 3.0 a touch to work interactively (rather than a
static document)). It also implements all of the necessary physicality
aspects of a virtual environment system (located objects, locations,
movement through locations, etc), as well as some other nice things.

My plan is to strap VRML scenes to each physical object, and have them
generated on the fly (for instance, locations would take not only their
scene, but each located object in their scene and generate a complete
scene including anchors to exits and other items and pass it to the client).

Since I have nearly come to a point where the core is comparable to
MOO's core, I'm now moving forward on my VRML plans in the creation of
the Matrix core. If anybody is interested (in helping or generic
information), feel free to contact me.

-Brandon Gillespie-

(For information's sake, the ColdX driver handles a database language
which is object oriented. It disk bases the database by pulling only the
needed objects into memory (this is a good thing). The language is "ColdC",
which is a simple yet elegant version of C and/or C++).