[MUD-Dev] "short" Introductory Message (fwd)

Martin Keegan martin at cam.sri.com
Fri Jun 20 09:34:14 New Zealand Standard Time 1997


On Thu, 19 Jun 1997 clawrenc at cup.hp.com wrote:
 
> >> Will Island ever arise again?  It would be a great loss to have it
> >> dissappear.  It did too may things far too well to lose, (and enough
> >> things far too stupidly to tolerate its loss as an example).
> 
> >"Oh no, not this chestnut again". I've already been swayed into
> >bringing it back - just trying to find a stable site for it. You
> >probably underestimate the amount of flak I *still* get from people
> >who hated that poor old game.
> 
> I caught some of the flammage during its life.  I'm not surprised. 
> What are the hardware requirements?

Well, it runs on various Unixes (Solaris, NeXT, Linux), and takes up about
8Mb of memory. Annoyingly, the atrociously written OxMUD code requires a
select() call with an insanely low timeout, which can keep the main
processes above 'top' on top without raising the load above .05 . If it
ever comes back, the OxMUD stuff would be ripped out, as it's more trouble
than it's worth.

> Re: Quests
> 
> >> And were for the uninitiated, utterly superb.  <I bow in their general
> >> direction>
> 
> >They were not too much fun if you were doing them for the second
> >time.
> 
> This is a fault of the quest system in general as implemented in MUDs.
> not of Island.  It is *rare* to find a quest on any MUD which isn't

Well, the game didn't attempt to make any allowances for people who
completed all the quests - once you were finished, that was it. You
normally sat around as some flavour of wizard until the next quest
upgrade. These only happened every 18 months or so, and the game simply
wasn't designed to deal with people who played that long.

> trite once you've done it the first time.  Cf my recent description of
> the Fortress Fract, King Mandel, and Princess Julia for a "quest"

The various scenarios mentioned here ought to be written up as some sort
of reference and posted every month, so that newcomers will know what is
being discussed.

> which was non-repeatable, just incredibly bloody annoying.  it is
> possible to do reset-less quests, and even to do non-trite,
> non-repeatable, non-automatable  quests, but they have to be designed
> that way from the start.  cf the recent thread on this area, and my
> comments on automatic systems to create conditions, not states.

It depends what your priorities are. Island didn't really care if the
quests were repeatable - the newer ones just had to be interesting and
challenging. There was a cool one where you had to break the bank at the
Casino. The cruel trick was that when you were in quest mode on that
quest, the roulette wheel would magically become deterministic:

nextnumber = (previous number + second to last number + 2) % 37

Once you'd worked out that it was Fibonacci + 2 mod 37, you could
complete the quest, reaping billions. This had drastic
longterm consequences for game balance.

I think it was Mtf who first said that Island was a great example of
things NOT to do.

> This is one of the reasons I am not a fan of democracies.  Equally I
> don't accept the "Most people..." assertion, (this is straying into
> philosophical areas not not meant for this list much tho I delight in
> them).  I profess the view that discrimination and exacting selection

I'm afraid that this issue has design implications. Most people don't like
to be challenged (see Marian's similar thoughts). Once you have enough
muds competing for players, muds can gain a competitive advantage by not
doing things players don't like - by not challenging the players. Any
doubters should read Alberto Barsella's frothings in r.g.mad (Combat
system design thread).

More seriously, unfamiliarity is a challenge. Muds which are similar to
other muds (such as two derived from the same codebase) are likely, all
other things being equal, to be less challenging for more players, since
there is a greater likelihood that some of the new players will have
played similar muds and be familiar with how they work. It's a vicious
circle - loads of similar muds engender a demand for exact that sort of
mud, and a rejection of innovations.

> have now been falsely equated with intellectual snobbery and a generic
> denigration of all those not practicing it.  Its a cheap half-arsed
> equation which doesn't stand up, but the media like it, and it sells
> newspapers almost as well as the Sun's page 3, so it must be good.

Hear, hear!

> >This system seems like a thought experiment but it's actually a
> >reaction against what I call "higher order imbalance" in many muds.
> >In some muds, powerful players can promote lowly players to high
> >ranks, as much as they want - they have limitless power. In a mud
> >with levels, a level 100 player could promote a level 10 player to
> >level 80 - as many times as she wanted. Really, she ought to use up
> >sufficient power to *drop* 70 levels.

[mayor snipped]
 
>   Perhaps investment economics would make a decent model?

There's a branch of economics called 'experimental economics', related to
game theory, which is involved in actually doing simulations to verify
axioms and assumptions that economists had previously worked out on paper.
These people are vaguely interested in muds, and probably know just how to
calculate a sane "cost" for creating objects.

> >The rationale is of not getting "something for nothing". The more
> >powerful/valuable the object/mobile/room you're creating, desto it's
> >gonna set you back.
> 
> How about, "The more capable the object, the more the creator is
> personally at risk for it?"  Then a "good" creator can create as much
> as he wants, but a "bad" creator gets negative investment feedback and
> zeros out.

Well, CamMUD is going to have a system where the arches penalise bad
spelling, grammar, punctuation and general Badness.

> >> Descriptions, definitions, or just names?
> 
> >Such strong typing, Chris! :) 
> 
> I know.  I wear out keyboards at an alarming rate.  <hang head>

This is from some very bad joke about strongly typed languages with
textbook examples in bold typefaces.

I tend to get keyboards dirty, for some reason. There's a mouse at the
Computer Labs into whose grime coating have been scratched the words
"Clean Me".
 
> >The system isn't really aware of any
> >distinction between words and sentences. 
> 
> <ponder>
> 
> Ahh. I see.  You use the same/similar macro form with the same
> replaceable parameters yada yada and then merely post parse the

There's no post parsing. So long as your definition grammar is sane, the
output will be too.

> results for acceptability.  Should work too.  But, how do you key this
> off a scene with given artifacts to generate a suitable description?

Well, I've been working on this, but said work is on a computer which has
been dismantled recently - I have the hard drive here, but not much else!
:)

Mk




More information about the MUD-Dev mailing list