Re: Learning to build a MUD

> I'd talk to Michael Granger or Martin Chase. Take a look at
> http://www.faeriemud.org/

I know I won't make many friends with the following, but I am now bold 
and claim that FaerieMUD, albeit it has a superior concept (just look at 
some older docs, or talk to them), is a dead project. Dead insofar that 
the progress is gone.

There seems to be a lack of motivation _within_ a short time frame. I am 
not raising fingers at all. In 10 years it may very well be the best 
existing MUD. ;-)

I have been following its progress when I started with ruby years ago, 
and there were only marginal changes to the project as far as I can see.

You can try to convince me otherwise with solid arguments, but other 
than that I will claim that this is a dead project. I do not really know 
why, maybe they need a real dictator that kicks em.

About 15 lines of ruby... it surely is a fun project, to showcase what 
may be possible, but it is not realistic to have a MUD in 15 lines of 
code.

To do a proper MUD - which is on my todo list as well, and I am coding 
on my own engine only in my free time and only sparingly - there are 
different "concepts", some I find hard, some I find easy.

Among the easy ones are of the LPC Model (everything is an object, i.e. 
you pick up a sword object from your environment object, into your 
inventory object).

Description, stats etc... is all done easily too. I very much like the 
inventory model, where i just stuff worlds, planets, and humans into a 
bag... ;)

Combat code can actually become very very complex though. I dont so much 
refer to the boring AD&D way to find out who wins, and the concept of XP 
gaining through monster hacking pretty much just bores me - but 
ultimately this is a trivial thing. Hit or not hit with one roll. If 
hit, roll damage. Deduct damage from enemy hitpoints (and consider 
damage reduction of enemy, but damage reduction is rather rare).

The more complex the combat code gets though, the more difficult 
everything becomes.
I am still thinking of proper combat engine which affects not only 
"physical rules" but also location implications (a hit on the arm may 
lead to a drop_weapon() event, hit on the head may lead to temporary 
blinding effect, or problems to stay on your legs, not to forget that 
loss of blood may affect negatively too etc.. etc..)

If you make sacrifices on the features (like AD&D with its super-simple 
hit-or-no-hit model), then it becomes quite easy (and by the way, I 
would demand of my combat engine to model different combat ways, so that 
I can use it in reallife pen and paper roleplaying games just as well).

But there is more than combat to a MUD. All the LPC MUDs i know treat 
rooms as objects wich some exits to other room (objects). This is a 
simple and workable model, but I myself actually favour a slightly 
different model, which focusses on "event locations" instead - although 
my MUD engine would have to support both anyway etc.. etc.. yada yada a 
lot of work... :(

And let's not forget client-server stuff. This is the most boring of 
everything ... ;)

I think even if you do a very minimal, yet functioning MUD, it becomes 
very complex and a lot of work.

I thus applaude everyone that tackles this task! Dont let the naysayers 
stop you. ;)
-- 
Posted via http://www.ruby-forum.com/.