11 Aug, 2012, chacham wrote in the 41st comment:
Votes: 0
Nathan said:
There is a really useful set of questions to help one figure out what kind of mud server you're trying to write, and what you want. It was written up by the guy who coded MUSHclient (Nick Gammon): http://www.gammon.com.au/forum/bbshowpos...


Thank you! That is a great thread.
14 Aug, 2012, quixadhal wrote in the 42nd comment:
Votes: 0
I don't know that writing Yet Another MUD Server from scratch is the best use of your time and energy. Especially if you're going to make one that works pretty much like all the other ones out there (IE: a loop that reads/writes to sockets, runs bits of code, mangles states, and then sleeps a bit to sync to a clock tick).

If you're coding from scratch because you have a totally different design idea, such as NOT having a single server process handle all the socket I/O, or NOT having everything in the game world be run inside that same server, great!

If you're coding from scratch because you intend to try to make money, and don't like any of the existing codebase license restrictions, great!

If you're coding from scratch, just because you think you can make a better tire…. there are plenty of tires out there that do the job just fine. People who aren't you would probably be happier if you just took an existing tire and improved it. You can, of course, not care and just do it anyways. :)
15 Aug, 2012, Alathon wrote in the 43rd comment:
Votes: 0
quixadhal said:
I don't know that writing Yet Another MUD Server from scratch is the best use of your time and energy. Especially if you're going to make one that works pretty much like all the other ones out there (IE: a loop that reads/writes to sockets, runs bits of code, mangles states, and then sleeps a bit to sync to a clock tick).


I thought a lot about it, actually. There are a lot of stock MUD bases out there, of varying level of detail. Some stripped all the way down, like SocketMUD; others more feature-complete.
Mostly, I imagine owners choice falls on knowing the language, and/or being happy with the feature-set. Or just random choice, maybe the name sounds cool :)

What I'd like to offer is something that is aimed at the same sort of crowd ('I want to create a MUD, but I'd like to jump straight into it'), but without having them chained down to C, or
stuck within a rigid framework that imposes a certain type of MUD down over their head. I'd like to promote a framework where forming the idea is core, and the programming happens
relatively easily.

I believe and hope that there are programmers out there who want a MUD, who may not like having to work off C. Who may feel intimidated by the number of things needed to get something
basic and workable, but would find it fun to take it the rest of the way. You can check out the list of features and see if that coincides with what you consider yet-another. Personally, I'm tired
of basic unscripted MUD schools; standard autoattack combat; MUDs where everyone has hp, mana and stamina; MUDs that don't encourage grouping together to do fun things that are hard; MUDs
that are completely and utterly static. And a lot of other things. I'm thoroughly convinced that most MUDs are stuck in the 1980s, and that there is SO much more that can be done with the medium.

I don't mean any insult to anyone in saying so - I've loved MUDs since I turned 8 and yahoo'd 'online game' and found RealmsMUD, which became my home for 5 years while I moved IRL every 6-12 months.
And people have done some very truly awesome things with MUDs, over the years. But I still think theres more :) And while a lot of this has to do with something entirely different than programming (Game design),
I believe that encouraging a framework that makes implementing features much easier, will also stimulate creativity and willingness to experiment.

Phew, sorry. That ended up being longer than I had expected.
15 Aug, 2012, Nathan wrote in the 44th comment:
Votes: 0
I think quixadhal is simply trying to point out that doing more with the medium is helped by working with an existing system since to make a new one you have to create a brand new pair of wheels, which may be inferior to existing ones. I.e. your new system needs to be as good as the old one for this to work out. No offense intended, but it's unlikely that your coding or mine is intrinsically better than someone who has been doing this kind of stuff for years.

My best argument for creating "Yet Another MUD Server from scratch" (nice phrasing there, quixadhal), is that it helps you to learn and you have intimate knowledge of the inner workings to some extent. Understanding how most of the stuff works leaves you free to implement new stuff more easily, still you could do that with an existing codebase as long as it is stable enough. It might even be better to work with an existing codebase so that you understand all of the intricacies of making it all work. The more codebases you have an understanding of, the better you can implement your own.

I think C++ is better than C for understanding readily what is being done and I think Java is better still in that regard.

You'll get a bit more notice if you use an existing codebase that is in use though.

In any case, more power to you, but please do inform us when you've made something you think is worth taking a look at, or else you'll have worked a ton and noone will ever know.
15 Aug, 2012, quixadhal wrote in the 45th comment:
Votes: 0
Alathon said:
What I'd like to offer is something that is aimed at the same sort of crowd ('I want to create a MUD, but I'd like to jump straight into it'), but without having them chained down to C, or
stuck within a rigid framework that imposes a certain type of MUD down over their head. I'd like to promote a framework where forming the idea is core, and the programming happens
relatively easily.


It sounds like you want an LpMUD.

An LpMUD is not easy to program, from the driver level… but an LpMUD driver bundled with an established mudlib (such as Dead Souls, Gurbalib, Discworld, or likely several others) provides a pretty solid system which can be modified in every way without ever having to recompile the driver. LPC itself is pretty simple to pick up, but powerful enough to do almost whatever you want to do (most LpMUD's have web servers, for example, programmed in LPC itself).

In any case, I wish you luck!
15 Aug, 2012, Alathon wrote in the 46th comment:
Votes: 0
quixadhal said:
It sounds like you want an LpMUD.

(…snip)
In any case, I wish you luck!


While writing my post, I actually did think of LPC.

Thanks! Hopefully someone else will find it useful as well, if anything I will!
40.0/46