Adventurers Club Ltd. Member's Dossier, April, 1987 Hat

Richard Bartle's Page

Now that people have seen it's possible to write Multi-User Adventure games, and that such games are quite fun, it's not too surprising that a few of them go ahead and write their own. I know of at least a dozen ex-MUD players who have done this, and every four or five months I hear from someone else I've never heard of who's doing a MUA for their O-level Computing (or A-level, or 3rd-year BSc., or Msc. dissertation). Of the ones I've seen, the approach is inevitably to design a representation language which can accommodate most of the basic problems (and all of them if you twist its arm enough!), and which looks vaguely MUD-like (because MUD looks vaguely ADVENTURE-like!) .

Although language design is an important issue, it's not what I'll be discussing in this article. Having a decent game is only part of the battle; if you want to run it publicly, then you need some idea of what we call "game management".

Game Management is hard! Let's take an example: you receive a frantic 'phone call fram a player who just lost carrier on his modem and when he logged back in his persona had been eaten by a dragon. Being a prudent person, you check your logs and, sure enough, the player was killed by a dragon. OK, so do you give him his points back?

Well, the immediate answer is yes. Why should the poor chap suffer for something out of his control? OK, well let's say that it happens again the next day, and a few days after that. Do you suddenly stop believing him? What if your system is good at detecting carrier loss? Then, it would be unlikely (but still possible) that the dragon could have eaten the player before you could stop it. And how can you tell genuine carrier loss from that caused by the player pulling the 'phone off the wall in an effort to avoid being deaded?

That was just a simple example! Although "the customer is always right" should be applied as often as possible, the problem is that you get two customers saying different things. Substitute another player for the dragon in the above. Then you have one player demanding his points back, and the other demanding that he doesn't (keeping points gained for killing someone is rarely good enough - having your opponent stay dead is often the most important factor, to avoid their revenge!) .

It's easy to make a problem go away by chucking points at the people concerned. "You lost a necromancer? OK, here it is back, and 500 points for the inconvenience"; "Your machine overheated in a fight? I'll refund your lost points and maybe give a few extra to Conanne to make up for your sudden resurrection". This is an ideal way to keep everyone happy and ingratiate yourself with the players. The problem, of course, is that you get taken for a ride by anyone remotely unethical.

Although MOST players can be trusted, a good many can't! We've had people claiming they were killed unfairly at a time when MUD was not even running! They've been set upon by a wiz-controlled dragon when no wizzes were playing, and they've mysteriously died avyer being hit once by a raven (this complaint stopped abruptly when we let it be known that we log everything to do with fights and can show exactly how many points damage each blow delivered1). Rumours spread, of course, and as soon as one player gets back points, for a plausible reason we've not come across before, within days it's happening to everyone!

This is just the tip of the iceberg. Since MUD has been around for many years, I've made all the mistakes already and have a reasonably good idea of game management. Budding MUA-writers, who can readily copy MUD's structure, will find themselves less able to devise a consistent way of dealing with players' wails, because they only see it in action piecemeal. I don't want to discourage such people - the more MUAs there are around, the better - but I would like them to be aware that writing the game is only part of the problem; running it can, at times, be worse! At least then they won't be in for the same sudden shock which greeted me the first time I realised that processing players' moans took up 90 minutes a day!

Copyright © Richard A. Bartle (
21st January 1999: aclapr87.htm