Good day, I was just mucking around in the #python channel talking with some guys, and ended up expressing my opinion of Reality, which at this point is not very high. Why you ask? I suppose my biggest problem is documentation and the lack of a proper distribution. In order to get reality, I had to use CVS, not a big deal, but I would have been much more contented with a tgz archive. You know, it could include that 'demo' world I've read about after searching through the mailing list archives. Along those same lines, nowhere in the cvs listing is _ANY_ documentation, save the docstrings in the source. Not to mention that on the twistedmatrix.com site, there is only passing reference to reality and how to get to it (get it from cvs, but you need to use the command: cvs -d:pserver:anon@twistedmatrix.com:/cvs co Reality Which I don't believe is listed anywhere on the twistedmatrix.com site. Along the 'demo' world I referred to above, information about that is also not listed on the twistedmatrix.com site, anywhere in the source (that I could find), documentation is nowhere to be found, save in the email archives...but email lists are not documentation. To get to the point; if you want people to use reality, get a distribution of any stable version, a world file with help about what is currently available in the mini-world, and a little how-to for getting the thing to run. It also wouldn't hurt to get a link to this distribution on some download page at twisted. I'm telling you this because I have just embarked on a project that quite possibly could last until I start my phd in september - possibly longer. Building a mud server and a mud using the server. I'm doing this because I don't believe any mud servers currently available are workable solutions to the problem of 'how do I make my mud do X', where X is anything. I want to make a DND 3rd ed rule-based mud. CircleMUD, SMAUG, ROM, ENVY, etc., are all 2nd ed based. Pulling out all that code for thaco, percent-based skills/spells, etc., and replacing it with D20 rules would be a daunting task (there are 3 guys who did it this past spring with CircleMud for a class, but their source is not downloadable). I'm not looking to re-make something that is doing too-much already. At 37k lines of code for circlemud, and nearly 140k lines for SMAUG, either is too huge. Reality is a nice change, but it still feels bulky (in terms of source code size) and I wouldn't want to muck around in the source for anything. Thanks, - Josiah Carlson