[MUD-Dev] CoolMUD lives!, sort of.

Dan Root dar at thekeep.org
Fri Aug 17 14:37:00 New Zealand Standard Time 2001

In message <3B7C86F7.8010706 at cubik.org>, Bruce Mitchener writes:
> Robin Lee Powell wrote:

>> So, from an offhand comment here, someone mailed me and it turns
>> out that at least on full-scale CoolMUD server was running
>> somewhere, which means there may have been, or may still be, more.
>> So, let me introduce myself.  Hi, I'm the maintainer for CoolMUD.
>> The official page for it is at

>>   http://www.digitalkingdom.org/~rlpowell/coolmud/index.html

> I looked at your site and I looked at some of a diff of your 2.3
> vs. the 2.1.4 that I found elsewhere.

> I found it somewhat hard to identify the changes without a
> changelog. :) Any clues for us?  What plans, apart from scrapping
> it one day in favor of the Mozart/Oz codebase, do you have for it?
> What else might you have to say about it or the changes that
> you've made?

> And, for you or anyone familiar with it, what would you say its
> strengths are?  The weaknesses?

In many ways, CoolMUD is an awful lot like LambdaMOO.  It's fairly
obvious that they started from the same base (Stephen White's
original MOO 1.0) and diverged from there.  The internal languages
are quite similar, to the point where porting simple code between
them is quite straightforward.  However, LambdaMOO 'won', and has
the advantage of about 10 years worth of incremental improvements,
including exception handling, better internal process control,
enhanced list operations, a whole truckload of new builtin functions
and so on.  You could probably add some or all of the features from
LambdaMOO into CoolMUD, if you were devoted or bored.  The other
major lacking of CoolMUD is a significant and useful core library
along the lines of LambdaCore.  The included 'core' makes a
half-hearted attempt to be a TinyMUD, but that's about it.

On the flip side of that coin, CoolMUD has a bunch of very
interesting features, many of which still aren't commonly available
in other servers.  Perhaps the most significant is support for
'distributed' muds, using a UDP protocol called YO for (almost
completely transparent) message passing between objects on seperate
servers.  Objects can have multiple inheritence and the security
system is also a bit more generalized, as all objects are allowed to
define their own protocol and criteria for responding to messages
passed to them.  The server is also disk-based, using a variant of
the UnterMUD database routines by mjr and Andrew Molitor, fixing a
fairly common gripe with LambdaMOO.

I actually ran one for a year or so (actually two, so I could play
around with the YO code) until my entire userbase was subsumed by
another MUSH. :) I found it quite easy to work with and program for.
I also did a bunch of patches to the server myself, which I've
passed off to Robin already, which is presumably what prompted his
above post.

MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the MUD-Dev mailing list