That's not important - leave it alone.

13 views
Skip to first unread message

Neil K. Guy

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to

A friend of mine was playtesting my game today, and made an interesting
comment that helped me articulate a problem in game design that I'm
having.

Basically it comes down to the way in which you're going to model the
world. How do you deal with irrelevant items?

The world is a pretty complex and rich place, and so you obviously
can't model every conceivable object that you might reasonably expect to
find. In early games, set in caves, it was pretty easy, since you could
just omit unimportant objects altogether. But games like mine, set in a
city, pose a problem. Particularly since my game is a bit more of a
simulation than some traditional pure puzzle games.


Some common approaches include:

1) Don't implement unimportant objects mentioned in descriptions. I
really don't like this. Reading "I don't know the word 'kumquat.'" or
"You don't need the word 'turnip' to solve this game." over and over
gets pretty tiresome.

2) Tell the player outright to ignore the object - implement it as a
basic decoration. "That's not important; leave it alone." I don't like
this either - it's rather rude and shatters any illusion of reality for
me.

3) Don't implement any objects at all. So you go into a grocery store
and it's totally empty - there isn't anything there. Likewise the
church, the fire station, etc. This I also don't like - it seems phoney
and empty, again wrecking any illusions.

3) Make things inaccessible. That's the approach my game has taken -
there are lots of buildings in the street, each with its own locked
door. Since they're just there for ambiance you can't go inside them.
Part of the puzzle, then, is figuring out what doors are important and
what aren't.


The problem is that my friend told me he got pretty fed up and
discouraged dealing with legions of locked doors until I produced a
spoiler-laden map for his benefit. Which is a pretty legitimate
complaint, of course.

So how to deal with this? How do you build up a convincing simulation
without bogging the player down in frustrating detail? Any ideas? I've
been toying with an optional spoiler mechanism for those players that
find my approach frustrating. eg:


>IS THE TIN OF BOILED YAMS A RED HERRING?

That would be telling.

>SPOILER

Spoiler mode on. From now on irrelevant decoration items will be
identified as such.

>IS THE TIN OF BOILED YAMS A RED HERRING?

Yes.

>OPEN THE PURPLE DOOR.

You pull at the locked door. Somehow you get the feeling it's not that
important, really, and give up.


Or is that too unspeakably clumsy for words?

- Neil K.

--
Neil K. Guy * ne...@sfu.ca * te...@tela.bc.ca
49N 16' 123W 7' * Vancouver, BC, Canada

David Baggett

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to
In article <4bspgt$4...@grid.direct.ca>, Neil K. Guy <n...@grid.direct.ca> wrote:

> 1) Don't implement unimportant objects mentioned in descriptions. I
>really don't like this. Reading "I don't know the word 'kumquat.'" or
>"You don't need the word 'turnip' to solve this game." over and over
>gets pretty tiresome.

This is so annoying that Brenda Laurel cites it as a favorite example of
"computer programs with brain-damage" in _Computers as Theatre_ (which
isn't even about interactive fiction). I think it's a hideous faux pas in
text adventures. If the talking part of the game uses vocabulary, the
listening part should handle that vocabulary. Otherwise the (already poor)
illusion that the reader's dealing with some kind of agent is completely
shattered.

>2) Tell the player outright to ignore the object - implement it as a basic
>decoration. "That's not important; leave it alone." I don't like this
>either - it's rather rude and shatters any illusion of reality for me.

This is lame, because the "not important" message can at the very least be
made interesting or amusing. When done well, this is fine (IMHO).

>3) Don't implement any objects at all. So you go into a grocery store
>and it's totally empty - there isn't anything there.

This overlaps a bit with #1. Also dreadful. I think IF requires something
of a naturalistic bent (describe everything in detail --- not just the
things that are directly relevant to the action), because it purports to
simulate a world.

>[4]) Make things inaccessible.

For buildings lining a street (your example), this may be the best
approach. But what's wrong with something between #3 and #4? Just
simplify the world so that you're simulating only an approximtion of a city
block. Perhaps you'll have three storefronts, whereas a real city block
might have eight. You only have to simulate enough for the reader to
suspend disbelief; besides, too much detail can be daunting.

>So how to deal with this? How do you build up a convincing simulation
>without bogging the player down in frustrating detail?

You want to have enough simulation for the reader to feel as though he's
really there. I don't think you have to go very deep to get this feeling.
But this is secondary to the requirement that the game understand the words
it uses to describe things --- that's an entirely different issue.

>Spoiler mode on. From now on irrelevant decoration items will be
>identified as such.

Not being a puzzle IF guy, I don't like this much. I see nothing wrong
with putting such "spoilers" in all over the place. And the technique
you've used here:

>You pull at the locked door. Somehow you get the feeling it's not that
>important, really, and give up.

is a basic and generally very effective one (IMHO) --- you inject thoughts
into the player/reader's mind via intuition.

Dave Baggett
__
d...@ai.mit.edu
"Mr. Price: Please don't try to make things nice! The wrong notes are *right*."
--- Charles Ives (note to copyist on the autograph score of The Fourth of July)

ErsatzPogo

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to
I agree that "That's not important; leave it alone" is a cop-out, and
annoying to play through. But I think you're ignoring a simple solution,
which is simply to describe in detail only those items that can be used
(and maybe a *few* red herrings):

MALL
Shoe stores, book stores, pet stores -- you name it, they can sell it to
you here, and all with the same annoying Muzak playing in the background.
You wander about for a while until you start to get vertigo, then finally
collapse in front of a drugstore.

> ENTER STORE

DRUGSTORE
Aisles and aisles of consumer goods stretch off to infinity. You're
starting to feel woozy again. You stagger onto a bench beside a candy
rack.

> X RACK

Hey! They've got Gummi Bears! Your favorite! [etc.]

OK, maybe a bit *too* contrived, but you get the idea. In your case, you
could describe all the doors, but lead the player to one in particular --
say, something catches their attention about the door you want them to
open. Then if they insist on trying all the other doors, they can't say
you haven't warned them.

This is hard to do well. But making a small amount of real information
seem like boundless vistas is the trick of good interactive fiction (*all*
fiction, one could argue).

Neil

Neil K. Guy

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to
David Baggett (d...@rice-chex.ai.mit.edu) wrote:

: >[4]) Make things inaccessible.

: For buildings lining a street (your example), this may be the best
: approach. But what's wrong with something between #3 and #4? Just
: simplify the world so that you're simulating only an approximtion of a city
: block. Perhaps you'll have three storefronts, whereas a real city block
: might have eight. You only have to simulate enough for the reader to
: suspend disbelief; besides, too much detail can be daunting.

True, but it seems that finding that cutoff point between enough
detail to be convincing and overburdening the player is very difficult.
I find the way most IF implements large areas like cities to be very
unconvincing. Infocom games, for example, often would have maybe half
a dozen locations as representing a town. Take the way "Bureaucracy"
implements a portion of a city, for example. I always found these
unsatisfying, because I didn't get any sense of size. My game has
around 4 dozen locations representing the main area of the city, but
one of my playtesters tells me he thinks it's too big. Daunting, as you
say. So I'm casting around for ways of building a reasonably convincing
city without having the player get tired of locked doors...

: >You pull at the locked door. Somehow you get the feeling it's not that


: >important, really, and give up.

: is a basic and generally very effective one (IMHO) --- you inject thoughts
: into the player/reader's mind via intuition.

The thing is, I'm not satisfied with that method either. I generally
find such placing of thoughts into the player's head to be rather
obtrusive and annoying. I've been forced to do it more and more in my
game, simply because I can't think of any other way of introducing
certain abstract concepts or emotions, but it still makes me uncomfortable.

Anyway, I don't mean to pick on your points, Dave. Just that I'm
frustrated with the conventional tools we seem to have at our disposal
for modelling text-based worlds. And I'm trying to think of ways around
that. So far with little success... Guess it's a bit of a windmill.

Julian Arnold

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to
Neil K. Guy (n...@grid.direct.ca) wrote:
> 1) Don't implement unimportant objects mentioned in descriptions. I
> really don't like this. Reading "I don't know the word 'kumquat.'" or
> "You don't need the word 'turnip' to solve this game." over and over
> gets pretty tiresome.
>
> 2) Tell the player outright to ignore the object - implement it as a
> basic decoration. "That's not important; leave it alone." I don't like

> this either - it's rather rude and shatters any illusion of reality for
> me.

1) is abysmal, don't be tempted to use it. 2) is common in Inform (as it's
so easy for the author to do without much effort -- the name property of a
location object). Graham Nelson's games seem to use this approach a lot. In
some circumstances I find it quite welcome (e.g., the bannister-thing in
Curses's attic), while in other others I find it frustrating (e.g., the baked
potato at the beginning of Jigsaw).

In the former instance, a bannister is a static, innocuous feature. It is
unlikely to be used for anything other than keeping one from falling down the
stairs. Of course, in text adventures it is often the seemingly most
innocent objects which are most useful, so the player is inclined to examine,
push, pull, etc. every object he comes across which doesn't have an immediate
obvious use. Now, a bannister, being a static object can't be moved. So,
the player must stay in the same place while fiddling with it. In this case
it is conducive to the flow of the game to just give the message "You don't
need to refer to that." -- the player is free to move on.

In the latter instance though, the baked potato is, in theory, a portable
object. So why not allow the player to pick it up and carry it away, even if
it is a red herring? Sure, he may try all those verbs on it to no avail, but
at least the game wouldn't be held up by this. The inventory is limitless,
so no problems there.

> 3) Don't implement any objects at all. So you go into a grocery store

> and it's totally empty - there isn't anything there. Likewise the
> church, the fire station, etc. This I also don't like - it seems phoney
> and empty, again wrecking any illusions.

This seems just a lazy way out. Maybe it's OK if you're writing purely
puzzle-oriented IF, where any extraneous detail would be nothing but...
extraneous, but I don't think anyone is writing such IF any more.

> 3) Make things inaccessible. That's the approach my game has taken -
> there are lots of buildings in the street, each with its own locked
> door. Since they're just there for ambiance you can't go inside them.
> Part of the puzzle, then, is figuring out what doors are important and
> what aren't.

This might work, but in the case you've outlined I would probably agree with
your friend (depending on just how many inaccesible doors there were) -- I'd
be discouraged from playing. I don't think I'd be too endeared to a puzzle
which involved working out something so mundane and arbitrary. It serves no
real purpose, and is no fun. If the player tries a wrong door it just means
he sits there for a bit longer -- there's no effect besides slowing the flow
of the game.

You might try treating the noun "door" on its own as a reference to one of
the inaccessible doors, while an accessible door would be "grocer's door" or
"florist's door". Thus,

>OPEN DOOR
You try a couple of doors, but they're locked.

>OPEN FLORIST'S DOOR
Opening the door your olfactory apparatus is assailed by the arresting
aroma of an array of attractive arrangements.

Would this work? I'm not sure.
--
Jools Arnold jo...@arnod.demon.co.uk


Neil K. Guy

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to
David Baggett (d...@rice-chex.ai.mit.edu) wrote:

: As with novels, this is part of the craft. I doubt there's a mechanical
: method that answers these kinds of questions. (Not a fun answer, of
: course.)

Oh, come on, Dave! You're a clever boffin! You work in a world-famous
AI lab. Surely you can whip up a little artificial intelligence that'll
do the trick? ;)

If things were that easy, eh?

: I guess it depends on how you feel about _Legend_, but I wouldn't
: underestimate the effectiveness of cut scenes. Instead of making your
: player go ">n, n, n, n" past five different buildings, you could have them
: type ">n" and noninteractively walk them past scenery that doesn't matter.

Hm. I've considered more overt ways of moving the player around, such
as cut scenes. But I guess I'm not entirely comfortable with this
approach. You know, if I'm sitting watching the puppets move on the stage,
and they glide along because someone is operating them with hidden
magnets, I don't much mind; it's part of the performance. But if a giant
hand reaches down from the Gods and picks up a player and puts it down
somewhere else, I feel the illusion is broken.

The difference, I suppose, is that one is cleverly disguised to look
internal to the action whereas the other is more clearly external. In an
adventure game I don't object to a game saying "As you pass through the
gate a cloud of noisome gas blows in. You collapse..." because the
action, no matter how contrived, seems part of the performance. But if
the game says "You hang around for a few minutes, then decide to walk
back to the more interesting part of the hallway." then I feel like my
character is being overtly manipulated by the author.

: I do find it encouraging that separate people have cited all three of the
: Barfee Outlet, the sequence on Foon, and the New Hell bit as portions of
: the game that evoked very vivid mental pictures. These are quite different
: from each other when it comes to level of interactivity. So I think many
: different styles can work.

That's a good point. Perhaps I should focus on trying to conjure up
evocative scenes rather than getting sidetracked by the issue of how
transparent the player's manipulation seems to be...

David Baggett

unread,
Dec 28, 1995, 3:00:00 AM12/28/95
to
In article <4btv49$a...@grid.direct.ca>, Neil K. Guy <n...@grid.direct.ca> wrote:

>True, but it seems that finding that cutoff point between enough
>detail to be convincing and overburdening the player is very difficult.

As with novels, this is part of the craft. I doubt there's a mechanical


method that answers these kinds of questions. (Not a fun answer, of
course.)

>Anyway, I don't mean to pick on your points, Dave. Just that I'm frustrated


>with the conventional tools we seem to have at our disposal for modelling
>text-based worlds.

I empathize with you. I felt this way many times while writing _Legend_.


I guess it depends on how you feel about _Legend_, but I wouldn't
underestimate the effectiveness of cut scenes. Instead of making your
player go ">n, n, n, n" past five different buildings, you could have them
type ">n" and noninteractively walk them past scenery that doesn't matter.

I do find it encouraging that separate people have cited all three of the


Barfee Outlet, the sequence on Foon, and the New Hell bit as portions of
the game that evoked very vivid mental pictures. These are quite different
from each other when it comes to level of interactivity. So I think many
different styles can work.

Dave Baggett

Dan Lyke

unread,
Dec 29, 1995, 3:00:00 AM12/29/95
to
In article <4bspgt$4...@grid.direct.ca>, n...@grid.direct.ca÷ says...

> Basically it comes down to the way in which you're going to model the
>world. How do you deal with irrelevant items?

> 1) Don't implement unimportant objects mentioned in descriptions.

Do this and you'll lose me _real_ fast. Games like this often deteriorate into
"guess which word the designer wanted":
You're in the produce department. In front of you are racks of
oranges, grapefruit, tangerines, peaches, apples, pears, kiwi
... potatoes, tomatoes, celery ...
> get oranges
I don't see any oranges here
> get grapefruit
I don't see any grapefruit here
> get tangerines
I don't see any tangerines here
[time passes]
> get tomatoes
You got the tomatoes. These are super crucial to the game play
and absolutely necessary to solving the puzzle.

> 2) Tell the player outright to ignore the object - implement it as a
>basic decoration. "That's not important; leave it alone." I don't like
>this either - it's rather rude and shatters any illusion of reality for
>me.

Unless I misunderstand, that's only a slightly less rude implementation of the
above, you're just putting everything in your descriptions in a list that gives
a "you don't need this" message. This should be part of the prose.

Now making the player think about what objects they'll need to solve a puzzle
is fine, and giving them red herrings is obviously sometimes critical (I
shouldn't blindly pick up _everything_ I see because I know I'll need it
later), so some extra objects are critical; implement them as thoroughly as
everything else.

> 3) Don't implement any objects at all. So you go into a grocery store
>and it's totally empty - there isn't anything there.

"You are standing on the corner of the business district. Numerous shops
selling various things dot the landscape. Of interest are the church, the
firehouse and the ice-cream parlor."

Doesn't sound too bad to me. Sure, you'll have to implement some _imaginative_
red herrings (otherwise the puzzle is just following a linear path, another
ho-hum).

> 3) Make things inaccessible. That's the approach my game has taken -

This to me is a repeat of #1, If this is going to deteriorate into a linear
search, count me out.
> North
You are standing in front of the firehouse
> Open door
The door is locked
> North
You are standing in front of the church
> Open door
The door is locked
> exit
I don't understand exit
> quit
I don't understand quit
> leave
Thank you for playing
C:\GAME>CD ..
C:\>DELTREE GAME
C:\>CD \DOOM
C:\DOOM>DOOM

When I played role-playing games I tried my hand at games-mastering and design.
I didn't understand abstraction. If you had to hack your way through 1000 orcs
to get to the goal, the castle included lodging for 1000 orcs. Not many people
played my games twice.

Abstraction is necessary to any fiction. Hint that there's a real world out
there, but please don't make your users interact with all of it.

Dan


Andrew C. Plotkin

unread,
Dec 29, 1995, 3:00:00 AM12/29/95
to
I think we've been round all the possible variations on this now,
right?

I dislike not implementing objects that are mentioned in the room
description. But I know that, unless you have very simple
environments, it's impossible to fully implement every object. You'd
go insane. So I prefer a lot of "that's not important." messages. They
aren't entirely in the mood (although it's really very common to tell
the player what to think), but at least they don't slow the game down,
and they don't give the impression that the author is amnesiac.

In Inform, actually, I've been doing a lot of objects which are one
step above "You don't need to refer to that object." I create a
scenery object whose before property is
[ "That's out of reach."; ]
or some other slightly individualized string. Maybe "Don't hassle
that." or "That belongs to the vendor, don't wreck it." It's there,
but you can't do anything to it.

If I want to expend a little more energy, make an exception for
"examine" and add a description. Now you can examine it, but nothing
else. And it still doesn't slow the game down, because when "take
thing" and "push thing" and "eat thing" all produce the same response,
the player catches on pretty quickly.

It's not as good as adding lots and lots of detail, but it's better
than taking three years to release your game.

--Z

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."

ErsatzPogo

unread,
Dec 29, 1995, 3:00:00 AM12/29/95
to
[Some Trinity spoilers ahead]

This thread had made me think of two items in Trinity that I hadn't
thought much about before. The first is an old favorite, the prohibition
on leaving Hyde Park in a certain direction because of the architectural
monstrosity (I forget the exact building) "you don't want to get any
closer to." It was an arbitrary obstacle that didn't make me think, "Hey,
why not?" Instead, it made me laugh.

The other is the boots and their role in the desert scene. I had always
thought of this as just a puzzle -- I'm only now realizing how important
it was in making the desert seem like a tangibly vast expanse.

I remain convinced that there's no problem like this that can't be solved
with the writing skills of Brian Moriarty.

Neil

ErsatzPogo

unread,
Dec 29, 1995, 3:00:00 AM12/29/95
to
One more thought: It's important not to overlook the importance not just
of keeping players away from certain objects, but of attracting them to
others. Like so:

ROCK-STREWN PLAIN
You see a rock, a rock, a rock, a rock, a rock, a rock and the Holy Grail
here. The grail sparkles majestically in the sunlight, making you feel as
if a higher power is personally blessing you, right now, at this very
moment, no kidding.

>TAKE ROCK
What are you, high?


Neil

Julian Arnold

unread,
Dec 29, 1995, 3:00:00 AM12/29/95
to
"Andrew C. Plotkin" (erky...@CMU.EDU) wrote:
> In Inform, actually, I've been doing a lot of objects which are one
> step above "You don't need to refer to that object." I create a
> scenery object whose before property is
> [ "That's out of reach."; ]
> or some other slightly individualized string. Maybe "Don't hassle
> that." or "That belongs to the vendor, don't wreck it." It's there,
> but you can't do anything to it.

This seems to be one of the better approaches, but I'd think you'd want to be
a bit careful.

For instance,
>GET SECURITY CAMERA


That's out of reach.

["Hm," thinks the player, "what can I do to get it in reach?"]

as opposed to
>GET SECURITY CAMERA
That belongs to the vendor, leave it alone.
[IMO this would make it pretty obvious you shouldn't touch.]
--
Jools Arnold jo...@arnod.demon.co.uk


Mark Green

unread,
Dec 29, 1995, 3:00:00 AM12/29/95
to
> MALL
> Shoe stores, book stores, pet stores -- you name it, they can sell it to
> you here, and all with the same annoying Muzak playing in the background.
> You wander about for a while until you start to get vertigo, then finally
> collapse in front of a drugstore.
> DRUGSTORE
> Aisles and aisles of consumer goods stretch off to infinity. You're
> starting to feel woozy again. You stagger onto a bench beside a candy
> rack.

> OK, maybe a bit *too* contrived, but you get the idea. In your case, you


> could describe all the doors, but lead the player to one in particular --
> say, something catches their attention about the door you want them to
> open. Then if they insist on trying all the other doors, they can't say
> you haven't warned them.

Yiiippe! Player actions described as part of room descriptions! It'll
seem a bit unnatural to have all these things happen *again* when you type
'look'. Probably better done is 222 Witherspoon in Princeton.
The *worst* "inaccessible object" I've seen is the boat mast in Curses.

Mg
--

Neil K. Guy

unread,
Dec 30, 1995, 3:00:00 AM12/30/95
to
Andrew C. Plotkin (erky...@CMU.EDU) wrote:

> It's not as good as adding lots and lots of detail, but it's better
> than taking three years to release your game.

Oh, cruel, *cruel* words!

But true, alas! :)

Andrew C. Plotkin

unread,
Dec 30, 1995, 3:00:00 AM12/30/95
to
jo...@arnod.demon.co.uk (Julian Arnold) writes:
> "Andrew C. Plotkin" (erky...@CMU.EDU) wrote:
> > In Inform, actually, I've been doing a lot of objects which are one
> > step above "You don't need to refer to that object." I create a
> > scenery object whose before property is
> > [ "That's out of reach."; ]
> > or some other slightly individualized string. Maybe "Don't hassle
> > that." or "That belongs to the vendor, don't wreck it." It's there,
> > but you can't do anything to it.
>
> This seems to be one of the better approaches, but I'd think you'd want to be
> a bit careful.
>
> For instance,
> >GET SECURITY CAMERA
> That's out of reach.
> ["Hm," thinks the player, "what can I do to get it in reach?"]

True. I was thinking in terms of a castle on the horizon, or something
at the top of a forty-foot pillar.

On the other hand, as an author, I *like* putting in things like a
security camera that's out of reach, which the player will simply
never be able to reach. Why should it be obvious which items are
soluble puzzles? :)

Julian Arnold

unread,
Dec 31, 1995, 3:00:00 AM12/31/95
to
"Andrew C. Plotkin" (erky...@CMU.EDU) wrote:
> > > In Inform, actually, I've been doing a lot of objects which are one
> > > step above "You don't need to refer to that object."
[...]

> > For instance,
> > >GET SECURITY CAMERA
> > That's out of reach.
> > ["Hm," thinks the player, "what can I do to get it in reach?"]
>
> True. I was thinking in terms of a castle on the horizon, or something
> at the top of a forty-foot pillar.
>
> On the other hand, as an author, I *like* putting in things like a
> security camera that's out of reach, which the player will simply
> never be able to reach. Why should it be obvious which items are
> soluble puzzles? :)

Well, of course there's no rule which says one should differentiate in this
way between soluble and insoluble puzzles. I was taking the "that's out of
reach" response as being directly analogous to "you don't need to refer to
that". In such circumstances that particular response wouldn't be much use,
as it raises doubts in the players mind (as above).

If, however, you're putting in such responses deliberately to play on their
ambiguity then that's a different thing altogether. (It's also very mean to
the poor old player, but I don't suppose the author of "Weather" minds that
8).
--
Jools Arnold jo...@arnod.demon.co.uk


ErsatzPogo

unread,
Jan 1, 1996, 3:00:00 AM1/1/96
to
> Yiiippe! Player actions described as part of room descriptions! It'll
>seem a bit unnatural to have all these things happen *again* when you
>type 'look'.

Oh, chill. You can put in a simple flag to change it the second time
around -- to, say, something like "Having explored all you want, you're
sitting in front of a drugstore." (Or just deleting the player action part
entirely.)

Anyway, I admitted that example was kind of contrived -- my point was that
the trick of good writing is to create the illusion of a rich, detailed
environment without using too many words. And this is doubly important in
*interactive* fiction, where the more words you use, the more words you
have to code.

Neil

Ken Leonard

unread,
Jan 1, 1996, 3:00:00 AM1/1/96
to
n...@grid.direct.ca (Neil K. Guy) wrote:

snip

> Basically it comes down to the way in which you're going to model the
>world. How do you deal with irrelevant items?

SNIP

---------------------------
You have entered Fumple's General Merchandise Emporium and Showplace
of Modern Wonders, which seems to be a rather ordinary hardware store.

There are the usual bins of bolts and buckets of chain. There are
several racks of hand tools and gardening implements. In the corner
to the left of the cashier's counter is a pail of nerf balls.
Scattered about at random are ladders and stools and a few cane
chairs. Among the hand tools are several bolt cutters. An epicene
gnome of Chicago watches you intently, as if you might be intending
to steal the entire store.
----------------------------

Now, what's so hard about that? I managed to mention quite a few
things contributory to the general ambiance and only two/three/four
things that might be of operational significance: the pail, a nerf
ball or the several nerf balls, (one of) the bolt cutters, and the
(shopkeeper.)

If you descriptions consistently follow the plan of identifying
non-significant items as compound subject/objects, and potentially
significant items as simple terms, your players are likely to
grasp the pattern quickly while you maintain the visualized richness
you desire.

regardz,

Ken Leonard
us01...@pop3.interramp.com


Julian Arnold

unread,
Jan 2, 1996, 3:00:00 AM1/2/96
to
Ken Leonard (us01...@pop3.interramp.com) wrote:
> ---------------------------
> You have entered Fumple's General Merchandise Emporium and Showplace
> of Modern Wonders, which seems to be a rather ordinary hardware store.
>
> There are the usual bins of bolts and buckets of chain. There are
> several racks of hand tools and gardening implements. In the corner
> to the left of the cashier's counter is a pail of nerf balls.
> Scattered about at random are ladders and stools and a few cane
> chairs. Among the hand tools are several bolt cutters. An epicene
> gnome of Chicago watches you intently, as if you might be intending
> to steal the entire store.
> ----------------------------

But now I want to SEARCH BINS, SEARCH BOLTS, SEARCH BUCKETS, SEARCH CHAIN,
EXAMINE RACKS, EXAMINE TOOLS, EXAMINE IMPLEMENTS, EXAMINE COUNTER, LOOK
BEHIND COUNTER, STAND ON COUNTER, EXAMINE PAIL, SEARCH BALLS, EXAMINE
LADDERS, EXAMINE STOOLS, EXAMINE CHAIRS, EXAMINE CUTTERS, and EXAMINE GNOME,
TALK TO GNOME, KISS GNOME, MOVE GNOME. I'll also probably try to GET
everything which isn't clearly immovable (i.e., everything but the counter).
I may try other verbs.

And I expect a sensible and/or witty comment for everything I try. 8)

> If you descriptions consistently follow the plan of identifying
> non-significant items as compound subject/objects, and potentially
> significant items as simple terms, your players are likely to
> grasp the pattern quickly while you maintain the visualized richness
> you desire.

Well, following a fairly consistent pattern in your descriptions is good to a
point, but it's surely more interesting (for both player and author) if you
occasionally stray from whatever rigid path you've set yourself. For
instance, why not have a medium-sized drill bit, fallen into a bin, and now
muddled with the bolts (an example of "non-significant compound objects" I
assume)?

>SEARCH BOLTS
You rummage around for a while. It's quite tiring, shovelling handfuls of
heavy bolts, but eventually your effort is rewarded as you uncover a
medium-sized drill bit.

You pick up the drill bit.

Basically, you need to define an object, or at least a set of messages, for
every potential object in the location, whether it is significant to the
gameplay or not. Otherwise, your "visualized richness" is nothing but
illusory.
--
Jools Arnold jo...@arnod.demon.co.uk


Carl D. Cravens

unread,
Jan 3, 1996, 3:00:00 AM1/3/96
to
On Mon, 01 Jan 1996 22:00:44 GMT, us01...@pop3.interramp.com (Ken Leonard) wrote:
>You have entered Fumple's General Merchandise Emporium and Showplace
>of Modern Wonders, which seems to be a rather ordinary hardware store.
>
>There are the usual bins of bolts and buckets of chain. There are
>several racks of hand tools and gardening implements. In the corner
>to the left of the cashier's counter is a pail of nerf balls.
>Scattered about at random are ladders and stools and a few cane
>chairs. Among the hand tools are several bolt cutters. An epicene
>gnome of Chicago watches you intently, as if you might be intending
>to steal the entire store.
>----------------------------
>
>Now, what's so hard about that? I managed to mention quite a few

>x bins
I don't know the word "bins".
>x bolts
I don't know the word "bolts".
>x buckets
I don't know the word "buckets".
>x chain
I don't know the word "chain".
>x tools
I don't know the word "tools".
>x racks
I don't know the word "racks".
>x implements
I don't know the word "implements".
>x counter
I don't know the word "counter".
>x ladder
I don't know the word "ladder".
>x stool
I don't know the word "stool".
>x chairs
I don't know the word "chairs".

That's what's so hard about it... these are all the things you mentioned
in the room description, but failed to see the need to code them up.
This kind of thing is very frustrating to the player... it is _not_
obvious by your style which things are important and which are not, so
the player has to try everything just to be sure. And the above series
of "try everything" gets really annoying. This was the fundamental
problem described at the beginning of this thread.

--
Carl (rave...@southwind.net)
* If it ain't broke, let me have a shot at it.

Norman Ramsey

unread,
Jan 3, 1996, 3:00:00 AM1/3/96
to
In article <2/m6wwIeE...@southwind.net>,
Carl D. Cravens <rave...@southwind.net> wrote:
%% On Mon, 01 Jan 1996 22:00:44 GMT, us01...@pop3.interramp.com (Ken Leonard) wrote:
%% >You have entered Fumple's General Merchandise Emporium and Showplace
%% >of Modern Wonders, which seems to be a rather ordinary hardware store.
%% >
%% >There are the usual bins of bolts and buckets of chain. There are
%% >several racks of hand tools and gardening implements. In the corner
%% >to the left of the cashier's counter is a pail of nerf balls.
%% >Scattered about at random are ladders and stools and a few cane
%% >chairs. Among the hand tools are several bolt cutters. An epicene
%% >gnome of Chicago watches you intently, as if you might be intending
%% >to steal the entire store.
%% >----------------------------
%% >
%% >Now, what's so hard about that? I managed to mention quite a few
%%
%% >x bins
%% I don't know the word "bins".
%% >x bolts
%% I don't know the word "bolts".
%% >x buckets
%% I don't know the word "buckets".
%% >x chain
%% I don't know the word "chain".
%% ... [and many more examples of same]

This is an example of something intelligent Dave Baggett said when
this discussion started: every output from the game must be a sensible
input.

Now, my business is programming languages and environments, not game
design. The natural question for me to ask is, what tools exist (or
can be written) to read TADS or Inform source and
- identify words used in output that aren't recognized as inputs
- help the author quickly fill in some reasonable behaviors for
those words
For my money, building such a tool might make a reasonable
undergraduate honors project, except I don't know how tractable the
TADS or Inform compilers are, or even if source code is available...

Norman Ramsey
--
``They who can give up essential liberty to obtain a little
temporary safety deserve neither liberty nor safety.'' ---Franklin
Norman Ramsey I support the Phil Zimmermann Legal Defense Fund.
http://www.cs.purdue.edu/homes/nr (http://www.netresponse.com/zldf)

Phil Goetz

unread,
Jan 4, 1996, 3:00:00 AM1/4/96
to
Just thought I'd beat one of my favorite dead horses, the Electronic Arts
game _Amnesia_. It implemented a complete street map of Manhattan, with
thousands of locations. (And you couldn't even wander through them -- you
died from hunger every 20 turns or so, so you had to restart hundreds of
times to even map out the game.)

This could be a good thing, *if* it is used to signal to the player that
exploring the map is not the proper approach in this game. But if you do
this, you had better make sure that the beginning puzzles are easy enough
that the player will get into the game, and not be tempted to
explore the map because he doesn't know what else to do.

Phil Go...@cs.buffalo.edu

Neil K. Guy

unread,
Jan 4, 1996, 3:00:00 AM1/4/96
to
Just a note - thanks to everyone who decided this was an important
enough thread to respond to and didn't leave it alone. :) Your comments
were very helpful!

Julian Arnold

unread,
Jan 5, 1996, 3:00:00 AM1/5/96
to
Mark Green (Ma...@antelope.demon.co.uk) wrote:
> This could be a problem. One of the main things that I think makes
> 'hidden-item' puzzles so unfair is that normally looking (in/behind/under/
> above) everything in sight gains no adventure value (usually there's nothing
> there) and very little entertainment value (you get stock library responses).
> It's very easy to assume from those responses that the game simply isn't
> going to involve searching and when it does you feel very cheated.

Well, I don't mind hidden items, but I know a lot of people do. I think as
long as there are clues in the text that further investigation may come up
trumps there isn't really cause to complain. I think making the player look
in/under/behind things can also help to give your textual world a certain
three dimensionality.

Now in a situation where there were 23 beds, and the author had decided
arbitrarily to put Apollo's winged flip-flops under number 17 I'd object.
Fortunately, I don't think anyone does this any more (if they ever did).

It'd be tedious if *everything* was in plain view always. A few subtly
signposted hidden items add to realism, and are very satisfying to find.

Also some things just beg to be searched or looked into (such as a bucket of
bolts). They're a signpost in themselves.
--
Jools Arnold jo...@arnod.demon.co.uk


Kvan

unread,
Jan 6, 1996, 3:00:00 AM1/6/96
to
n...@cs.purdue.edu (Norman Ramsey) writes:

>Now, my business is programming languages and environments, not game
>design. The natural question for me to ask is, what tools exist (or
>can be written) to read TADS or Inform source and
> - identify words used in output that aren't recognized as inputs
> - help the author quickly fill in some reasonable behaviors for
> those words
>For my money, building such a tool might make a reasonable
>undergraduate honors project, except I don't know how tractable the
>TADS or Inform compilers are, or even if source code is available...

The full source is available, so it should be manageable. Any takers? :)

In fact, such a tool would allow an author to start out coding the basivcs
and then flesh it out with the tool. That could change the whole
design-process a lot. You'd only need to code that which takes actual
coding, leaving the tedious work of inserting the right replies in the right
places for the Glorious Reply Tool (tm). It would also make it much easier
to implement answers requested by beta-testers and/or players. Just run your
code through the GRT and insert the proper reply.

Perhaps the GRT should check all words, giving you the option of making an
`ignore-file' for words that are unimportant. If shipped with a basic
`ignore-file', it shouldn't take long for that file to grow to a very usable
level.

Alternatively, the GRT could be a souped up interpreter (though it'd have to
compile and re-compile in run-time). This would make the whole process
extremely easy, since you'd be able to play your game while you built it.

Graham? A little something to keep you occupied? :)

I know for a fact that my programming skills aren't up to the task, so I'll
just sit back and shout suggestions at the people who do the thing. Butg
someone who can do it would certainly ease the lives of all IF writers.

What a messy post. And I'm even sober :)

Frunp,
Kvan.
--
kv...@diku.dk (Casper Kvan Clausen) | I think TWINKLE's a nice word. So's
| VIRIDIAN. I met a lady once who had
| an imaginary FISH.
http://www.diku.dk/students/kvan/ | Delirium.

root

unread,
Jan 8, 1996, 3:00:00 AM1/8/96
to
Neil K. Guy (n...@grid.direct.ca) wrote:
: Just a note - thanks to everyone who decided this was an important

: enough thread to respond to and didn't leave it alone. :) Your comments
: were very helpful!

Well, now that I have time to introduce my pair of pennies worth:

I figure that if you are going to mention something in a description,
then _have_ a description for it, for the inevitable moment when the
player decides to look at it, because they figure they've missed a clue,
or are thinking along radically different lines to you, the author.

If you mention the sun, you can bet your draft card that _at_least_
one player will try to look at it. Why? Telling the time? Trying to figure
which way is north, in the confusing maze of twisty canyons (all alike)?

Zarquon knows. Something. They'll try it. Don't skimp on these. Even if 90%
of your players are used to never looking at anything that doesn't appear
when hey try to 'get all', the other 10% _will_ see them, and _will_ judge
your writing as much on those bit-descriptions as on the prose of room-
descriptions, and on the more regular text with which they are confronted.

Verisimilitude. It's a pain to type, or to say when inebriated, but it's IT.
The big I-T of I-F (or indeed of _any_ -F).

If we can capture their imagination long enough for them to see...to really
_see_ the worlds we conjure, then we've won. We've brightened the world a
little, and truly created in the finest artistic medium known to man:
The Imagination.

D

Matt Ackeret

unread,
Jan 8, 1996, 3:00:00 AM1/8/96
to
In article <19960105....@arnod.arnod.demon.co.uk>,

Julian Arnold <jo...@arnod.demon.co.uk> wrote:
>Now in a situation where there were 23 beds, and the author had decided
>arbitrarily to put Apollo's winged flip-flops under number 17 I'd object.
>Fortunately, I don't think anyone does this any more (if they ever did).

Now, if it's actually under bed #11, and there's a clue like "If you
know the moon isn't made out of green cheese, you'll find the shoes"..
But then again, this may be construed as an "American" puzzle. (Just because
we were first! Yeah! U S A U S A! heh.)
--
unk...@apple.com Apple II Forever
These opinions are mine, not Apple's.

Zellyn James Hunter

unread,
Jan 8, 1996, 3:00:00 AM1/8/96
to
Matt Ackeret (mat...@eskimo.com) wrote:
: Now, if it's actually under bed #11, and there's a clue like "If you
: know the moon isn't made out of green cheese, you'll find the shoes"..
: But then again, this may be construed as an "American" puzzle. (Just because
: we were first! Yeah! U S A U S A! heh.)

Hahaha... but I get the shoes were made in Korea or some such place!


Really, why can't you guys just stop these petty arguments over whether the US
or the UK is better, and just admit that you wish you could be South African
so you could be cool too. ;-)

>DON FLAMEPROOF CLOAK
I only understood you so far as wanting to flame the don.

>AAARRGHH


-Zellyn

We really are mixing threads now, aren't we?

--
,d8b. d"b d"b "For You knit me together in my mother's womb.
`Yb d"b $ $ $ $ g g gd"b, I praise You because I am fearfully and
dP' "dP "dP "dP $ $ d `$ d wonderfully made."
"b,gd'"oP" "oP" "oP'YgP$'8 YP'
----dP-----------------,o"-P------------------------>Zellyn Hunter<-----------
YgdP' Ybgd' gt2...@prism.gatech.edu

Peter Wright

unread,
Jan 9, 1996, 3:00:00 AM1/9/96
to
n...@grid.direct.ca (Neil K. Guy) writes:

>root (dan...@brisnet.org.au) wrote:

>> Verisimilitude. It's a pain to type, or to say when inebriated, but it's IT.
>> The big I-T of I-F (or indeed of _any_ -F).

> Hm. Yeah, but it's also why I have a 1.2 Mb game binary 3 years later and
>the damned thing still isn't finished! :)

Ye gods. What the hell are you writing it in? A particularly obese
version of Visual Basic?

> - Neil K.

Pete.


Neil K. Guy

unread,
Jan 9, 1996, 3:00:00 AM1/9/96
to
Peter Wright (pjwr...@tartarus.uwa.edu.au) wrote:

> > Hm. Yeah, but it's also why I have a 1.2 Mb game binary 3 years later and
> >the damned thing still isn't finished! :)

> Ye gods. What the hell are you writing it in? A particularly obese
> version of Visual Basic?

No. TADS. And before you think my game is large because of TADS, don't.
It ain't TADS' fault. The game is just ludicrously overdetailed with an
enormous vocabulary. Not to mention a fair whack of text. It is, in fact,
the result of trying to create manipulable objects for everything described
in the game, as "Dancer" advocates in an earlier post.

Neil K. Guy

unread,
Jan 9, 1996, 3:00:00 AM1/9/96
to
root (dan...@brisnet.org.au) wrote:

> Verisimilitude. It's a pain to type, or to say when inebriated, but it's IT.
> The big I-T of I-F (or indeed of _any_ -F).

Hm. Yeah, but it's also why I have a 1.2 Mb game binary 3 years later and


the damned thing still isn't finished! :)

- Neil K.

Reply all
Reply to author
Forward
0 new messages