[MUD-Dev] (fwd) Re: Issues from the digests and Wout's list

clawrenc at cup.hp.com clawrenc at cup.hp.com
Thu May 8 09:46:50 New Zealand Standard Time 1997


In <33738988.3422800 at post.demon.co.uk>, on 05/06/97 
   at 08:43 PM, Raz <muddyraz at mushroom.demon.co.uk> said:

>On Tue, 29 Apr 1997 19:33:59 PST8PDT, Chris L wrote (quoting me):
>> [ChrislL:]
>> >Remember the old, single-player text adventures, particularly the
>> >later games from Magnetic Scrolls/Rainbird?
...
>> They are also all single user systems turn based systems, where the
>> definition of state is extremely distinct and nothing is going to
>> happen while the user is sitting there at a prompt picking his nose. 
>> Multi-user realtime systems don't have a lot of that advantage.  The
>> definition of state is a lot more gray, and that state is highly
>> unstable dependant on other's realtime actions.

>Very strong point, of course, though I've recently started thinking
>that, in practice,  perhaps this won't always be such a major
>problem...

You got me thinking again onm this, and with a slight change in the
way my filters work, I think I agree with you.  The problem that I see
now is handling the error conditions intelligently, not handling the
state.

>I guess I'm writing off the cuff here, and not thinking of every
>possible situation, but I reckon that the goal is to have a finished
>parser which, even if the 'state' the user is in at the point that
>the parser prompts them to resolve ambiguity changes significantly,
>won't leave the player in a 'dangerous' situation.

Bingo.

>If it handles, or rather, is informed of these changes and responds
>intelligently, I'd hope that there shouldn't be too much problem.

The solution I see here for my model are lotsa lotsa watchers.

>An example, one player on the left, one player on the right: (no
>proportional fonts, please ;) )

> > n

> You're in a room; it seems it was
> created on naught but a whim.
> A big bag and a small bag lie here.

> > get bag                             > n
> Which bag - the big one or the small
> one?                                  You're in a room; it seems it
>was
>                                       created on naught but a whim.
> >                                     A big bag and a small bag lie
>here.
> A tall, female human enters.          A lean, male human is here.

> >                                     > get big bag
> The female human picks up the big     You pick up the big bag.
> bag.

> > big
> It doesn't seem to be here anymore,
> it appears you took too long!

A possible diagrammed solution:

 > n
 You're in a room; it seems it was
 created on naught but a whim.
 A big bag and a small bag lie here.

 > get bag                             > n
 Which bag - the big one or the small
 one?                                  You're in a room; it seems it
was
                                       created on naught but a whim.
[Filter installed for get selection]   A big bag and a small bag lie
here. [Watchers installed on all possible    A lean, male human is
here. slections]

 >
 A tall, female human enters.

 >                                     > get big bag
 The female human picks up the big     You pick up the big bag.
 bag.

[State change on the big bag is 
triggers the watcher]
[Watcher informs filter that bag
is no longer present]

 > big
 It doesn't seem to be here anymore,
 it appears you took too long!

[Filter gives some decent handling 
having already changed its internal 
state]

>The above seems to be a generally smooth handling of that particular
>situation. Obviously, nicer interaction could be passed to the two
>participants - the parser could respond to the male human in the last
>instance "The big bag was just taken by the female human, take the
>small one instead?" or, better, offer this prompt the instant the
>female takes the bag.

Exactly.  The filter can get all that data from the watcher very
transparently without having to know of the event that caused the
change.  This then allows the filter to respond in an intelligent
manner.

>Perhaps there will be situations that can't be resolved smoothly...
>come up with some and we can try and plan a chain of events that get
>through them =)

<muses>

--
J C Lawrence                           Internet: claw at null.net
(Contractor)                           Internet: coder at ibm.net
---------------(*)               Internet: clawrenc at cup.hp.com
...Honorary Member Clan McFUD -- Teamer's Avenging Monolith...




More information about the MUD-Dev mailing list