Frustration in text adventures
Mr A J Sithers 92
GOOD adventure game , I thought I'd introduce a new thread.
I've recently been playing some of the old magnetic scrolls games
(specifically Jinxter) and this brought to light on of the most frustrating
aspects of 'traditional' text adventures - having to repeat most of the game
because of a mistake I made early on. For example, in the afore mentioned
game, you find a box of matches early on, (with SOME matches - we're never
told how many). I accedently lit one of the little buggers outside, and it
was blown out by the wind (fair enough). However near the end game, I need
to use a match - and of course I've run out. My only option is
to repeat the whole game from when I first lit the match outside - and of
course I'm not encouraged to do this. Obviously the programmers thought -
"OK, he'll need three matches to finish the game - and that's what we'll
give him". What I feel would have been better (and more condusive to
finishing the game) would be :
1. More matches. (Obvious) - either in the box or better still ...
2. Matches near the puzzle - the puzzle should not utilise the matches
directly however
3. A warning "Looks like you've only got one match left" or something.
I find this problem occurrs throughout many commercial adventures I've played
and restarting the game really drives me to distraction. One of the nicest
features of the LucasArts games is that this simply does not happen, the
puzzle will stand in its own right and not be dependent on an (unrecognised)
accident that occurred early on the game.
Anyway, this is just some food for your brains, let's hear what others have
to say.
Cheers
Andy
A.J.Sithers Department of Pharmacology
email wph...@cardiff.ac.uk UWCM, Heath Park, Cardiff
sem...@thor.cf.ac.uk United Kingdon, CF4 3QD
It works! It works! I'm a Genius!
Brian Lane
: 1. More matches. (Obvious) - either in the box or better still ...
: 2. Matches near the puzzle - the puzzle should not utilise the matches
: directly however
: 3. A warning "Looks like you've only got one match left" or something.
: I find this problem occurrs throughout many commercial adventures I've played
: and restarting the game really drives me to distraction. One of the nicest
: features of the LucasArts games is that this simply does not happen, the
: puzzle will stand in its own right and not be dependent on an (unrecognised)
: accident that occurred early on the game.
You've hit upon one of Graham Nelson's rules of good IF! I know exactly
what you mean, you've spent many hours getting to where you are at and you'd
really rather just put the game aside than re-play it just for a match. I
think alot of the magic scroll games were like this - I never did finish the
Pawn although I did get pretty far.
I'd like to see IF move from the 'clean' worlds where you wander about
locations picking up things, maybe meeting a character or two, into a world
where you have more people to interact with. More like a interactive
text-movie simulation if you get my meaning. It sounds like Avalon is going
to be along these lines.
cypress
David Baggett
<sem...@thor.cf.ac.uk> wrote:
>For example, in the afore mentioned game, you find a box of matches early
>on, (with SOME matches - we're never told how many). I accedently lit one
>of the little buggers outside, and it was blown out by the wind (fair
>enough). However near the end game, I need to use a match - and of course
>I've run out.
Problems like this are dire. Adventure game authors should avoid this
problem even more than they should avoid the temptation to put in stupid
mazes. The Adventions policy (we've discussed this amongst ourselves at
great length) is to make you lose points whenever you do something that can
make the game unsolvable, or even to say something along the lines of "Now
you've gone and spoiled the game. Would you like to UNDO?" The players
will *thank* you for being so clear about it.
This is one reason UNDO is so important for playability. We didn't realize
this ourselves until Mike put UNDO support in TADS. Once you've played a
game with UNDO you won't be able to stand games without it. (I really
found this with Rylvania testing. I almost never saved and restored and
slogged through things I'd already done many times -- I just used UNDO all
over the place.)
Dave Baggett
__
d...@ai.mit.edu Boot up, log in, drop out. MIT AI Lab
ADVENTIONS: We make Kuul text adventures! Email for a catalog of releases.
PO Box 851 Columbia, MD 21044 USA / CIS: 76440,2671 / GEnie: ADVENTIONS
Phil Goetz
Brian Lane <cyp...@hebron.connected.com> wrote:
> I'd like to see IF move from the 'clean' worlds where you wander about
>locations picking up things, maybe meeting a character or two, into a world
>where you have more people to interact with. More like a interactive
>text-movie simulation if you get my meaning. It sounds like Avalon is going
>to be along these lines.
I'm wandering a deserted church after midnight in a small Rylvanian
town. The author has done a wonderful job of dredging up details gothic
and foreign yet familiar -- there is a graveyard behind the church full
of Rylvanian names which echoes every dark October night of my childhood --
when I come upon a coat hanging on a peg in the back room. I nearly
walk right past it.
Whoah. I'm in an Adventure. I almost forgot, and thought I was
exploring a church in a foreign country, in which I would never dream
of even looking in the pockets of someone else's coat, let alone stealing
a coat (probably the pastor's) from the church.
But I'm not. I'm just playing a game. Better take the coat.
'Nuff said.
Phil go...@cs.buffalo.edu
(who doesn't know if he actually needed the coat,
but the point remains the same)
Stephen R. Granade
In a previous article, go...@cs.buffalo.edu (Phil Goetz) says:
> I'm wandering a deserted church after midnight in a small Rylvanian
>town. The author has done a wonderful job of dredging up details gothic
>and foreign yet familiar -- there is a graveyard behind the church full
>of Rylvanian names which echoes every dark October night of my childhood --
>when I come upon a coat hanging on a peg in the back room. I nearly
>walk right past it.
>
> Whoah. I'm in an Adventure. I almost forgot, and thought I was
>exploring a church in a foreign country, in which I would never dream
>of even looking in the pockets of someone else's coat, let alone stealing
>a coat (probably the pastor's) from the church.
>But I'm not. I'm just playing a game. Better take the coat.
The problem is, in if, the player is the main character. In real life,
we don't have all of these wonderful adventures in strange places, or
even ordinary places. Imagine, if you will, a game in which you are
allowed only to work with what is explictly yours. Ugh, too much like
my life as it is: get up, get my clothes & wear them, go to work...
However, if we want to avoid these things, we should add other characters
who get understandably upset when you snarf their things.
I personally believe that there is a certain suspension of belief
necessary in games, just as in the theatre or in movies, or indeed in *any*
piece of art which attempts to mimic life to some degree. We should be
striving for a balance between enjoyment/playability and realism.
Stephen
--
_________________________________________________________________________
| Stephen Granade | "My research proposal involves reconstructing |
| | the Trinity test using tweezers and |
| sgra...@obu.arknet.edu | assistants with very good eyesight." |
Gerry Kevin Wilson
etc, etc. If the item is just laying around, or doesn't belong to the
player, there should be a reason that the character doesn't mind stealing
all of a sudden.
--
<~~~~~E~~~G~~~~~~~~~~~HEINLEIN~~~~~~~~~~~DOYLE~~~~~~~~~~~~~SPAM~~|~~~~~~~>
< V R I O Software. We bring words to life! | ~~\ >
< T | /~\ | >
<_WATCH for Avalon in early MAY!____wh...@uclink.berkeley.edu_|_\__/__>
A cohort's CoHort
Stephen R. Granade <bz...@cleveland.Freenet.Edu> wrote:
[...]
>Imagine, if you will, a game in which you are
>allowed only to work with what is explictly yours. Ugh, too much like
>my life as it is: get up, get my clothes & wear them, go to work...
>
if so, you ever try to open up Ford's satchel? I have never been able to do
more than move it about.
But, I can do anything with stuff that Ford had handed me... Hmm, explicit
ownership stuff happening here? I wonder...
--
CoH...@hebron.connected.com | Overheard at a bridge table:
@Stellar Genesis | "So, do you know the value of notrump?"
(jericho.connected.com 4040) | "Yeah, it should be just a bit larger than
(Stellar down due to TD :( ) | your shoulders."
Stephen R. Granade
In a previous article, coh...@hebron.connected.com (A cohort's CoHort) says:
>In article <2kc4ki$e...@usenet.ins.cwru.edu>,
>Stephen R. Granade <bz...@cleveland.Freenet.Edu> wrote:
>[...]
>>Imagine, if you will, a game in which you are
>>allowed only to work with what is explictly yours. Ugh, too much like
>>my life as it is: get up, get my clothes & wear them, go to work...
>>
>So, Stephen, you ever play Infocom's "Hitchhiker's Guide to the Galaxy"?
>if so, you ever try to open up Ford's satchel? I have never been able to do
>more than move it about.
> But, I can do anything with stuff that Ford had handed me... Hmm, explicit
>ownership stuff happening here? I wonder...
True, I had forgotten Ford's Satchel. However, later in the game you go
wandering around the Heart of Gold, scarfing up someone else's tools. And
I don't think it's polite to move other people's stuff around, but that's
just me. :)
At any rate, that still leaves us with the mindset that the player should
pick up anything that's not nailed down. Assuming that we want to alleviate
that mindset, I think there are a couple of options:
1) Make lotsa movable items. This will make the game more realistic (in
terms of the amount of things lying about real life), and force the player
to decide if they *really* need that wastebasket. However, this technique
can make code unwieldy. It can also cause player frustration: too many
red herrings and the average person will toss the game away in disgust.
It also doesn't address the mindset of "get everything," it just forces
more selective taking.
2) Tie every item to an NPC. Want that vacuum-ray gun? You have
to convince the evil professor to give it to you. Alternatively, NPCs
might notice that the player has a tendancy to take everything, and start
gathering up their belongings to protect them. Or fight with the
player if they won't return something. Good points: has the feel of
realism. Drawbacks: how many secretaries would be upset if someone took
one paper clip from their desk? Also, this too requires exponentially-
increasing complexity.
3) Divide things into "owned objects" and "unowned objects." For example,
a discarded screw would be unowned; a letter would be owned. Penalize the
player for taking owned objects. Good points: this is how many of us view
the world: mine and yours. Drawbacks: a bit simplistic and artificial.
I believe that a lot of this ties into NPCs. If the player is the only
game in town, why shouldn't he pick everything up? Noone will get upset
with him. The better we make the NPCs, the more realistic the whole
game becomes.
As I have said before, this quickly turns into a balancing act between
playability and realism, between effectiveness and complexity. If
we were writing simulations of the real world, an issue such as this
one would have to be explicitly handled. However, IF is a subset
of the real world, and we have to decide how much of the real world we will
attempt to model, and how much we will discard for coding/playability's
sake.
Any comments or thoughts?
Michael Kinyon
(Stephen R. Granade) writes:
>I believe that a lot of this ties into NPCs. If the player is the only
>game in town, why shouldn't he pick everything up? Noone will get upset
>with him. The better we make the NPCs, the more realistic the whole
>game becomes.
Indeed, do you recall the NPC adventurer from Enchanter who went around
picking up every item that the player hadn't taken? I thought that
was a beautiful parody of the typical Adventure/Zork player.
--
Michael Kinyon | mki...@peabody.iusb.indiana.edu
Dept of Mathematics & Comp. Sci. | "There is no quote in my .sig" -- M. Kinyon
Indiana University South Bend | Text-Adventure Betatesters Union (TABU)
South Bend, IN 46634 USA | *** Your score just went down ***
Fred M. Sloniker (L. Lazuli R'kamos)
>>>Imagine, if you will, a game in which you are
>>>allowed only to work with what is explictly yours. Ugh, too much like
>>>my life as it is: get up, get my clothes & wear them, go to work...
How about a game in which taking things that don't belong to you has
consequences appropriate to what you take? No one's likely to notice if you
take flowers from someone's grave, but the ship's captain is likely to object
with large-bore weapons if you try to take his security pass... unless, of
course, he doesn't *see* you do it.
He also wrote:
>At any rate, that still leaves us with the mindset that the player should
>pick up anything that's not nailed down. Assuming that we want to alleviate
>that mindset, I think there are a couple of options:
>1) Make lotsa movable items.
Ultima VI did this (in Lord British's castle, there were hundreds of movable
objects). Didn't work; we just restricted ourselves to 'snarf everything
that's either useful in combat or that we need to solve a puzzle, and remember
where the rest of it is for later.'
>2) Tie every item to an NPC. Want that vacuum-ray gun? You have
>to convince the evil professor to give it to you. Alternatively, NPCs
>might notice that the player has a tendancy to take everything, and start
>gathering up their belongings to protect them. Or fight with the
>player if they won't return something.
This approach I like, though it doesn't really address the question. It just
turns getting various items into a puzzle or a red herring, depending on
whether it's possible or not. ("Okay, what if I tell Jake Sterling the
Martians are invading, take his spare blaster when he dashes outside, then
blame it on his nemesis Dr. Unsure when he returns? Will that work?")
>Drawbacks: how many secretaries would be upset if someone took one paper clip
>from their desk?
So mark that item as belonging to no one, just like some objects exist nowhere.
>Also, this too requires exponentially-increasing complexity.
Exponentially? I don't see it. Polynomially at most (you're not going to
need to model past 'will NPC N react to the taking of his object X?' in most
situations; in some, you might want to go as far as 'will NPC N react to the
taking of any object Y?', but there are various ways you can cut down on how
many NPCs you have to check, and you're already limiting to a single object.)
Still, it is a source of complexity.
>3) Divide things into "owned objects" and "unowned objects." For example,
>a discarded screw would be unowned; a letter would be owned. Penalize the
>player for taking owned objects.
Ultima VI did this too. (You lost virtue points if you took something someone
hadn't said you could, not counting loot from killed enemies.) Didn't work;
we just cleaned out the areas we could.
>Any comments or thoughts?
Just one (besides the ones above): this doesn't really solve the packrat
mentality, does it? It just puts the player through more work.
How about this idea (it may be a little radical, but bear with me):
Write an adventure that isn't dependent on 10^6 iterations of "Get object.
Use object at location."
I've got a proto-adventure in mind (one of these days, I'll sit down and flesh
it out...) in which *all* the 'puzzles' have nothing to do with swiping other
peoples' stuff wholesale; instead, one collects *information* about what's
going on in the game, and/or properly *interacts* with NPCs. As an example,
how about the 'puzzle': "Find out who the hidden saboteur is"? That's not
something you're going to do by using the right object at the right location;
it's something you're going to do by going through surveillance tapes, talking
to various suspects, and possibly hiding out in some dingy service crawlway
to record the saboteur's chat with his compatriots (and, no, before you ask,
you don't have to get the tape recorder from some arcane location. You're
given it as part of your mission equipment.)
How about the 'puzzle': "Convince the beautiful lady/handsome gentleman/well-
brushed lemur to spend an evening with you"? It's not something you're going
to do by offering him/her/it the diamond brooch you found in the Pit of
Incredible Nasty Evil and Instant Death; it's something you're going to do by
the correct choice of conversation, taking cues from their reactions and
anything you happen to know about them (for instance, is the lemur a bit
ticked that people keep thinking of him as a harmless cute furry thing?
Maybe if you treat him as just 'one of the guys', he'll go with you to the
foreign film festival...)
Any comments on *my* comments?
---Fred M. Sloniker, stressed undergrad
L. Lazuli R'kamos, FurryMUCKer
laz...@u.washington.edu
DISCLAIMER: My evil twin wrote this message.
Gerry Kevin Wilson
one of their manuals had a section that said "Pick up everything that isn't
nailed down." So we have to either live with it or wean the players off of
it. Frankly, I don't see it as a terrible thing. There's no reason to me
to create endless items just to obfuscate what the player does and add a
little realism to the game. Just making sure everything has a discription
goes a long way towards that. And of course NPCs should react to theft. But
then, might not the character himself react to it (if the character has a set
personality). Ownership is one way to deal with it, but I don't see it as
neccessary unless you are going to allow the player to indulge in theft all
over the place. For me, it's not too hard to work around it though.
Stephen R. Granade
In a previous article, laz...@u.washington.edu (Fred M. Sloniker L. Lazuli R'kamos)) says:
[my original comments in '>>']
>>Also, this too requires exponentially-increasing complexity.
>
>Exponentially? I don't see it. Polynomially at most (you're not going to
>need to model past 'will NPC N react to the taking of his object X?' in most
>situations; in some, you might want to go as far as 'will NPC N react to the
>taking of any object Y?', but there are various ways you can cut down on how
>many NPCs you have to check, and you're already limiting to a single object.)
>Still, it is a source of complexity.
Ok, perhaps not exponentially. However, would you not want to model 'will
NPC N2 react to the taking of NPC N1's object X?' If you steal my
brother's car, I'm likely to want to rough you up a bit. If the NPCs
are interrelated at all, the complexity of their reactions would grow quickly
into a nasty network, unless we divide the NPCs into groups, or not have
them interrelated at all. I would prefer to have them react to each
other and to how you act towards all of them.
>>Any comments or thoughts?
>
>Just one (besides the ones above): this doesn't really solve the packrat
>mentality, does it? It just puts the player through more work.
You're right, it doesn't. And you're right, it won't until we make
puzzles that require no objects to solve, or at least no objects which
you have to hunt down and then figure out their use. I happen to like
the object-type puzzles, although a game should include other types of
puzzles. The only trouble with dispensing with object-based puzzles
is coming up with something else that gives as much freedom of movement.
I will never forget, in my misguided youth, spending money for Sierra's
Police Quest I and solving it in three hours. If you have to follow a
script exactly, it becomes rather easy to see what to do next.
It's possible to write a game with few object puzzles that would be harder
to solve, but I've personally never played one. If anyone could suggest
one (or more :) to me, I'd love to try them.
(Most likely this all leads back to having believable NPCs with whom you
can have a conversation of sorts...)
Fred M. Sloniker (L. Lazuli R'kamos)
the previous article for previous references):
>Ok, perhaps not exponentially. However, would you not want to model 'will
>NPC N2 react to the taking of NPC N1's object X?' If you steal my
>brother's car, I'm likely to want to rough you up a bit. If the NPCs
>are interrelated at all, the complexity of their reactions would grow quickly
>into a nasty network, unless we divide the NPCs into groups, or not have
>them interrelated at all. I would prefer to have them react to each
>other and to how you act towards all of them.
I think you could limit this sort of network explosion effectively with
knowledge limiting; that is, how do you know I took your brother's car?
Either you're in the room when I take it (one checks if any of the actors
in a room reacts to an action of another actor) or you see me driving it
(one of the things an actor does when deciding his own actions is to observe
his environment). We can also propagate the data of what the actor cares
about more directly, rather than relying on some massive inter-actor network;
that is, you're programmed to worry about the car, rather than programmed to
worry about anything belonging to your brother, of which the car is an example.
While this does increase the data size, it also means you don't have to
recursively check the interrelations of every actor in the game.
(Believe it or not, this leads to more realism than a 'brother's keeper'
approach, even if you have to explicitly state more: I wouldn't particularly
care about someone playing with my brother's LEGOs, for instance, while he
would certainly be ticked. On the other hand, we both agree that the computer
is not to be manhandled.)
[And, on all this not solving the 'packrat mentality'...]
>You're right, it doesn't. And you're right, it won't until we make
>puzzles that require no objects to solve, or at least no objects which
>you have to hunt down and then figure out their use.
Or, alternatively, one where some objects don't show up as objects until your
character has learned he needs them. As an example, I'm not likely to notice
the flashlight, power saw, screwdriver set, and pile of dirty laundry in the
basement work room if I'm there to get something out of the fridge. If the
lights suddenly go out, however, you can bet I'll search the room until I
find the flashlight.
Here's another idea to curb packrat tendencies: what if taking certain objects
makes certain puzzles harder to solve? (Sure, you can strip the power cell
out of the alien artifact. But you're going to be ticked when you find out
(a) it doesn't fit in any Fleet tech you can find, and (b) you've shut down
the protective force shield holding the space plague inert... now you have
hallucinations to deal with on top of everything else.)
>I happen to like the object-type puzzles, although a game should include
>other types of puzzles.
Good enough reason for having some. (:3 I rather like the way "The Secret
of Monkey Island" makes fun of the character's packrat tendencies from time
to time. (I also like quite a number of other things about the game, one of
them being you never get irrevocably stuck.) That "Oh good, more inventory!"
line is a classic. (:3
>The only trouble with dispensing with object-based puzzles is coming up with
>something else that gives as much freedom of movement.
>I will never forget, in my misguided youth, spending money for Sierra's
>Police Quest I and solving it in three hours. If you have to follow a
>script exactly, it becomes rather easy to see what to do next.
(grins) Don't confuse a lack of object-based puzzles with excessive linearity.
(Object-based puzzles can be *very* linear: find the copper key to open the
copper door, then the silver key to open the silver door, then the gold key,
you guessed it...) "Monkey Island" (yes, again!) let you wander all over the
place for most of the game, and while it had a lot of object-based puzzles,
there were also a lot of conversation-based puzzles (the first of which is
to get you some direction so you don't wander around aimlessly).
>It's possible to write a game with few object puzzles that would be harder
>to solve, but I've personally never played one. If anyone could suggest
>one (or more :) to me, I'd love to try them.
Ever tried "Nord & Bert Couldn't Make Head or Tail of it", one of Infocom's?
There were some object puzzles, but most of the puzzles involved various
plays on words. A word of warning to our international readers: you have to
have an excellent grasp of the English language to work out all the puns,
spoonerisms, and idioms that show up throughout the game (and a knowledge of
50's sitcoms helps).
>(Most likely this all leads back to having believable NPCs with whom you
>can have a conversation of sorts...)
An admirable goal, and one I'm shooting for. (:3
>"My research proposal involves reconstructing the Trinity test using tweezers
>and assistants with very good eyesight."
How dare you? Haven't you ever played "Trinity"? (;3
---Fred M. Sloniker, stressed undergrad
L. Lazuli R'kamos, FurryMUCKer
laz...@u.washington.edu
This message is shareware. If you enjoyed reading it, send $5 to...
a90...@zipi.fi.upm.es
> Mr A J Sithers 92 (sem...@thor.cf.ac.uk) wrote:
>
> : 1. More matches. (Obvious) - either in the box or better still ...
> what you mean, you've spent many hours getting to where you are at and you'd
> really rather just put the game aside than re-play it just for a match. I
> think alot of the magic scroll games were like this - I never did finish the
> Pawn although I did get pretty far.
>
It also happened to me in THE PAWN, the text displayed is so well chosen and
the screen format and char type so nice that I ended up thinking it was a
good game. I somehow got stuck and my memory kept filled with the idea that
the game was supperb. Some years latter I (yes, I did it I confess!) I got a
serialized solution and it showed how linnear and mechanical it was. I couldn't
even finish it with the solution!!!!
Anyway I still consider the beggining of THE PAWN as one of the best pieces
of IF to date.
By the way, did anybody here play THE HOBBIT? If so, I'd like to hear from you.
Javier B Díaz
a90...@zipi.fi.upm.es
"La mentira es algo que se esconde para no tener que existir..."
Santiago Auserón (De un País en Llamas, RADIO FUTURA)
Ville Lavonius
: It also happened to me in THE PAWN, the text displayed is so well chosen and
: the screen format and char type so nice that I ended up thinking it was a
: good game. I somehow got stuck and my memory kept filled with the idea that
: the game was supperb. Some years latter I (yes, I did it I confess!) I got a
: serialized solution and it showed how linnear and mechanical it was. I couldn't
: even finish it with the solution!!!!
How true, most of the reviews were based on the wonderful graphics (at that
time) I guess... And while the parser was good, it was nowhere near
Infocom's (eg. in Jinxter you can't build the command "lean ladder against
girder" piece by piece, as it is only accepted when typed in as a whole).
MS's latter games like Guild of Thieves or Jinxter were more traditional.
: By the way, did anybody here play THE HOBBIT? If so, I'd like to hear from you.
"Thorin sits down and sings about gold".
Yeah, played and solved ages ago.
: Javier B Díaz
: a90...@zipi.fi.upm.es
: "La mentira es algo que se esconde para no tener que existir..."
: Santiago Auserón (De un País en Llamas, RADIO FUTURA)
--
Ville Lavonius / Kangaroo lady with her bourbon in a pouch
lavo...@peruvian.cs.utah.edu / can't afford the rental of a bamboo couch
Erik Max Francis
> >Get key from chest on table
>
> That is getting the key from the chest standing on the table,
> referring to THREE objects in the same sentence.
>
> I don't think Infocom's parser could do that.
But is it really important for it to know how to do this? If you
_have_ to type this instead of just GET KEY or GET KEY FROM CHEST, I
would suspect that the game would be very frustrating -- now it's not
interactive fiction, it's merely finding the correct sequence of words
that the parser is looking for.
Erik Max Francis, &tSftDotIotE ...!uuwest!alcyone!max m...@alcyone.darkside.com
USMail: 1070 Oakmont Dr. #1 San Jose, CA 95117 ICBM: 37 20 N 121 53 W __
AGCTACTGTACGTACGTTTGCACGTATGCTGTGCAXTGCATACTGACATCGTGACTGATCTGCATGACTTGCA / \
"Omnia quia sunt, lumina sunt." (All things that are, are lights.) \__/
Bob Newell
>>>>allowed only to work with what is explictly yours. Ugh, too much like
>Use object at location."
This really IS a problem with the genre, isn't it? Even in a game as classy
as The Horror of Rylvania, I come up with something like:
You are carrying a fifty dollar travellers check, a cloak, a wine bottle,
a flyswatter, a horse hair, a wooden stake, a shovel, a book of matches,
a brass-bound book, a satchel, a flask, and an iron key.
A pretty eclectic collection, I must say. As the man said, get object, use
object.
A system I coded some time ago, called PLS, which had a very brief and
unglorious commercial life, was written in a manner that avoided all this
sort of stuff. (It avoided it in the simplest manner: the parser was so
lousy that it simply wasn't possible.) But those limitations did encourage
you to write games in which what you did and said mattered rather than
picking up objects because some puzzle will require them somewhere down the
line. I am currently, perpetually, trying to write event-driven IF as
opposed to possession-driven. The fact that I have yet to publish a game of
this type speaks to the fact that it isn't at all easy to break the mold.
Ville Lavonius
: >time) I guess... And while the parser was good, it was nowhere near
: >Infocom's (eg. in Jinxter you can't build the command "lean ladder against
: I'm very sorry, but I have to disagree.
: Magnetic Scroll's could handle sentences like this:
: >Get key from chest on table
> Plant plot plant in plant pot
was given out as an example of the parser's abilities. Quite nifty.
: That is getting the key from the chest standing on the table,
: referring to THREE objects in the same sentence.
: I don't think Infocom's parser could do that.
Yeah, but MS's parser was built to be fancy, not robust. The girder example
I gave above was just one of my frustrations.
Just my opinion of course.
: -Darius Katz
The Grim Reaper
Bob Newell <bne...@delphi.com> wrote:
>>>>>Imagine, if you will, a game in which you are
>>>>>allowed only to work with what is explictly yours. Ugh, too much like
>>Write an adventure that isn't dependent on 10^6 iterations of "Get object.
>>Use object at location."
>
>This really IS a problem with the genre, isn't it? Even in a game as classy
>as The Horror of Rylvania, I come up with something like:
> You are carrying a fifty dollar travellers check, a cloak, a wine bottle,
> a flyswatter, a horse hair, a wooden stake, a shovel, a book of matches,
> a brass-bound book, a satchel, a flask, and an iron key.
>A pretty eclectic collection, I must say. As the man said, get object, use
>object.
This is bad, but it isn't *that* bad. In real life, for instance, if
I was a traveller in a foreign country, my inventory might be:
You are carrying a wallet, a watch (worn), a warm jacket (worn), a
walking stick, a travel guide, a backpack (worn), and a baseball cap (worn).
The wallet seems to contain some money, a room key, a folded map,
some credit cards, and a note from your girlfriend. The jacket pocket
seems to contain some matches, a swiss army knife, [...]
We do carry an awful lot of things around in real life. And if I was stalking
a vampire, I'd pick up anything that might be useful in killing it. I think
the main problem is not actually getting items, but *disposing* of them.
It's not unrealistic for me to have the wine bottle in the restraunt, but it
is silly for me to continue carrying it when it's outside. How to handle this
problem? No idea.
>A system I coded some time ago, called PLS, which had a very brief and
>unglorious commercial life, was written in a manner that avoided all this
>sort of stuff. (It avoided it in the simplest manner: the parser was so
>lousy that it simply wasn't possible.) But those limitations did encourage
>you to write games in which what you did and said mattered rather than
>picking up objects because some puzzle will require them somewhere down the
>line. I am currently, perpetually, trying to write event-driven IF as
>opposed to possession-driven. The fact that I have yet to publish a game of
>this type speaks to the fact that it isn't at all easy to break the mold.
A worthy goal, to be sure. Good luck, though.
+----------------------------------------------------------+
| One .sig to rule them all, one .sig to find them... |
| One .sig to bring them all and in the darkness bind them |
+----------------------------------------------------------+
| The Grim Reaper (Reaper of Souls, Stealer of .sigs) |
| scy...@u.washington.edu |
+----------------------------------------------------------+
Ben Muckenhoupt
>We do carry an awful lot of things around in real life. And if I was stalking
>a vampire, I'd pick up anything that might be useful in killing it. I think
>the main problem is not actually getting items, but *disposing* of them.
>It's not unrealistic for me to have the wine bottle in the restraunt, but it
>is silly for me to continue carrying it when it's outside. How to handle this
>problem? No idea.
Hmm, an interesting thought. Perhaps if there were two levels of inventory,
"holding" and "carrying" or something, so that items you are simply holding
would be put back where they came from when you leave the room. To carry
something, you would have to first be holding it - think of it as putting
it in your backpack or whatever. This would get rid of unrealistic
situations like absent-mindedly picking up a bowling ball and then forgetting
about it, carrying it with you wherever you go. You would only carry things
all over the place if you intended to do so, as signalled by the "carry"
command (or "put in backpack", or whatever.)
Would such a two-level system be cumbersome? Not necessarily. It would have
to be reflected in your game design, of course. It would be unbearable
in something like Zork or Unnkuulia, where you would just put every item
you pick up in your backpack. But in a world with many insignificant
or redundant objects, where most things are located near where they are
needed, it could be an improvement. Of course, most adventures aren't like
that, but is that the nature of adventure games or the influence of the
inventory system? In a traditional adventure, you might be requires to
go through a long series of complicated and dangerous feats, the end result
of which is that you get a rope. Or a nail. Or any other common item, made
unique in the game by the fact that all objects in the game are unique.
BTW, there are some games that make you automatically drop certain objects
when leaving the room, producing a message like "You put the receiver back
on the phone before leaving." I don't think that the dropping of held
objects, under the two-level system, should produce a message. It should
be part of the ordinary actions that are taken for granted.
One big problem with this or any other inventory reform: The heavy players
(and that includes most people who still play text adventures) are used to
the old way. Well, what do you think?
Carl Muckenhoupt
Darius Katz
Erik Max Francis <m...@alcyone.darkside.com> wrote:
>ka...@marvin.df.lth.se (Darius Katz) writes:
>
>> >Get key from chest on table
>>
>> That is getting the key from the chest standing on the table,
>> referring to THREE objects in the same sentence.
>>
>> I don't think Infocom's parser could do that.
>
>But is it really important for it to know how to do this? If you
>_have_ to type this instead of just GET KEY or GET KEY FROM CHEST, I
>would suspect that the game would be very frustrating -- now it's not
>interactive fiction, it's merely finding the correct sequence of words
>that the parser is looking for.
>
I didn't say you have to type it. I said that you can type it, if for
example there also was a chest standing on the floor containing a key.
And that's superior to Infocom's parser, IMHO.
-Darius Katz
Darius Katz
Ville Lavonius <vpl...@u.cc.utah.edu> wrote:
>Darius Katz (ka...@marvin.df.lth.se) wrote:
>: >time) I guess... And while the parser was good, it was nowhere near
>: >Infocom's (eg. in Jinxter you can't build the command "lean ladder against
>
>: I'm very sorry, but I have to disagree.
>
>: Magnetic Scroll's could handle sentences like this:
>
>: >Get key from chest on table
>
>> Plant plot plant in plant pot
>was given out as an example of the parser's abilities. Quite nifty.
>
>: That is getting the key from the chest standing on the table,
>: referring to THREE objects in the same sentence.
>
>: I don't think Infocom's parser could do that.
>
>Yeah, but MS's parser was built to be fancy, not robust. The girder example
>I gave above was just one of my frustrations.
>
>Just my opinion of course.
>
Maybe a combination of Magnetic Scroll's and Infocom's parsers
would be the "perfect" parser?
-Darius Katz
Joanne Omang
interact in a realish way...bravo. Don't know if you read earlier postings
here on the subject of increased literary merit for interactive fiction, but
this sounds to me like a promising approach. It's clear that paper-based
forms of writing are not wholly suitable for cyberspace, so some new kind of
fictive form has to incubate. Consider yourself a broody hen...
Joanne Omang
PS That's meant as a compliment, please understand!
Duane D Morin
Jamieson Norrish <ja...@kauri.vuw.ac.nz> wrote:
>For some items at least a different method could be used, whereby
>using the item does not transfer it to the character's inventory. For
>example, you put the wine bottle on the table. Later, you pour
>yourself a glass of wine. Although you don't specify this, you put the
>wine bottle back on the table. Although this might not work for all
>objects and all places, it could still mean avoiding some of the
>problems of the method outlined above.
A more general variant of this was described by Schank years ago in
the AI field, known generally as "scripts." A common example:
Enter diner.
Sit.
Order meal.
Eat.
Leave.
Implies: There is an empty booth. You take it.
A waitress comes over to take your order. She brings you your order.
You pay for the meal when it's done.
>Those problems are that there are many situations in which you would
>want to have items carried in the hands while you moved around. For
>example, carrying a security pass while walking through a well-guarded
>complex. Or was the proposal on a more abstract level?
>wear security badge
or
>show badge to guard
So many different ways to "carry", "hold", etc...:(
[automatically dropping objects, etc, when leaving a room]
>I'm in two minds about this: sure, the actions are ordinary, and so
>taken for granted. However, wouldn't it be helpful to the player to
>have these little messages, and wouldn't they also add something to
>the atmosphere of the game? I'm keen on the idea of plenty of feedback
>from character actions, and this would seem to be a part of that.
It seems like the only happy medium here would be a system that knows
what objects you don't really want to carry, and which you want to keep.
If the system assumes that you don't really want it, but you really did,
then a message saying that you can't take it out of the room would be
nice. I'd hate to pick up the dustball from under the desk and then
spend 50 turns travelling to the bugblatter's lair in the hopes of using
it to make him sneeze, only to find out that the computer had dropped
the dustball for me before I left the office.
Or, like some other posters have mentioned, focus on puzzles where it's
not what you pick up that wins the game for you, but how you act in
certain situations. Must more unbroken ground there, it would seem.
>Jamie
Duane
Jamieson Norrish
muck...@math.rutgers.edu (Ben Muckenhoupt) writes:
Hmm, an interesting thought. Perhaps if there were two levels of
inventory, "holding" and "carrying" or something, so that items you
are simply holding would be put back where they came from when you
leave the room. To carry something, you would have to first be
holding it - think of it as putting it in your backpack or
whatever. This would get rid of unrealistic situations like
absent-mindedly picking up a bowling ball and then forgetting about
it, carrying it with you wherever you go. You would only carry
things all over the place if you intended to do so, as signalled by
the "carry" command (or "put in backpack", or whatever.)
For some items at least a different method could be used, whereby
using the item does not transfer it to the character's inventory. For
example, you put the wine bottle on the table. Later, you pour
yourself a glass of wine. Although you don't specify this, you put the
wine bottle back on the table. Although this might not work for all
objects and all places, it could still mean avoiding some of the
problems of the method outlined above.
Those problems are that there are many situations in which you would
want to have items carried in the hands while you moved around. For
example, carrying a security pass while walking through a well-guarded
complex. Or was the proposal on a more abstract level?
BTW, there are some games that make you automatically drop certain
objects when leaving the room, producing a message like "You put
the receiver back on the phone before leaving." I don't think that
the dropping of held objects, under the two-level system, should
produce a message. It should be part of the ordinary actions that
are taken for granted.
I'm in two minds about this: sure, the actions are ordinary, and so
taken for granted. However, wouldn't it be helpful to the player to
have these little messages, and wouldn't they also add something to
the atmosphere of the game? I'm keen on the idea of plenty of feedback
from character actions, and this would seem to be a part of that.
Jamie