[MUD-Dev] Re: Parser engines

T. Alexander Popiel popiel at wolfskeep.com
Wed Mar 10 13:39:34 CET 2004

Brian Hook <brianhook at pyrogon.com> wrote:

> I've been looking at the various input parsers in different MUDs
> (although I haven't had a chance to look at commercial text MUDs),
> and by and large the parser technology seems to be roughly
> early-Infocom, if that.

> This is in contrast to the parsers in interactive fiction engines

Most MUDs are at ADVENT and DUNGEON level verb-noun stuff, though
(barely) more complex forms are used for multi-object actions, etc.
MUSH (and PennMUSH in particular) is on the slightly more complex
side in that it has a fairly elaborate noun-matcher, yielding stuff
like "give my 3rd box=this here 2nd marble"... but that
functionality is almost never used.

> So several questions:

>   - what MUDs do have IF-quality parsers?

None that I know of.  There've been several experiments in that
direction, all of which have been abandoned, AFAIK.

>   - assuming I'm correct that most MUDs (Diku and LP derivatives)
>   don't have IF-quality parsers, is there a good reason for this
>   and is this an impediment to less combat oriented, more puzzle
>   oriented play?

Yes, there's a good reason for it, and no, it isn't an impediment to
puzzle-oriented play (which is not uncommon on MUSHes).  The reason
is that in a multi-player environment, getting the command to the
server quickly (so you can interact with the other people in
realtime) is more important than having beautiful syntax.  Fewer
words is faster to type, so people can issue more commands in the
same timeframe, keeping the game moving.

> My guess for the second question above is that many players don't
> care about the quality of a parser so long as the basic commands
> they want are there

This is certainly true.

> and that a lot of the features in IF-engines [...] are considered
> fluff or irrelevant to most combat MUD players.

This is also true, though certainly not limited to combat MUDs.  If
you can type "get cardboard's grey" and "give machine=marble" to
accomplish the equivalent of "put the grey marble from the cardboard
box into the machine", and it takes half the time to type and a
quarter of the thought to come up with the two commands as opposed
to the pretty sentence, then the people who are doing this or
similar over and over in a multiplayer setting will pick the two

Another thing to consider is the amount of time that people spend
typing commands in MUDs vs. in IF games.  In a MUD, you tend to be
either issuing commands every few seconds or idling.  In an IF game,
you tend to be either moving from location to location with
'n.n.n.e.s.e.e.w.' (with brevity equal to a MUD), or pondering a
puzzle, issuing a brief spurt of commands every few minutes.

Thirdly, consider how long a person spends playing a game.  For
MUDs, you can reasonably expect that people will be playing the game
for a few months to years.  With an IF game, people work through
them in days to weeks.  Thus, there's a lot more time in a MUD to
learn its particular command structure, instead of making it accept
any odd sentence you throw at it.

So no, the simplicity of the parsers in MUDs is not because the MUD
developers haven't thought about making them accept more complex
syntax, but rather because with a MUD the players will adopt (after
a short learning period) the simplest form of command that's
available, because it's faster to use.  There's no point in having a
complex parsing engine that is a maintenance hassle for the devs and
nobody uses.

- Alex
MUD-Dev mailing list
MUD-Dev at kanga.nu

More information about the mud-dev-archive mailing list