Simulation vs convention vs completeness

2 views
Skip to first unread message

richard develyn

unread,
Jan 24, 2006, 6:31:47 AM1/24/06
to
Again I would like to canvas opinion on another aspect of game design
which is how to deal with interactions which are not related to the
game's narrative path.

I'm sure this has been discussed before - as this was alluded to in the
puzzle fairness thread (should I be able to jump off every bridge).

Should I?

Some more examples. In all cases where this has nothing to do with the
game:

Sould I be able to climb a tree anywhere where there is a forest?

Should I be able to set fire to anything flammable if I have a fire
somewhere?

Should everything be breakable?

Should everything have a description?

What about a smell or taste or sound?

Should everything which could be a surface / supporter be a surface /
supporter? Every table? Every mantlepiece? Every chest of drawers?

- I suppose there is a general point here, and my take on it would be
this:

1) Everything should have a description.
2) If the game narrative requires you to be able to perform an action
on an object, then you should answer sensibly any attempts to perform a
similar action on the object, or to perform the on similar objects. For
example, if in order to complete the game you need to cut a letter up
with a pair of scissors, then you would expect to be able to cut up any
paper with a pair of scissors. If you can't, then the designer needs to
come up with a very good reason why not.
3) If the game design doesn't require you to be able to perform a
particular action on an object, then you should provide a consitent,
and possibly conventional, answer to indicate that the action is
outside of the game's simulation-field (e.g. "you're not an arsonist").
4) I'm not sure about supporters / containers .... what do people
think?

IMO (2) is the real challenge of game design.

Richard

Uli Kusterer

unread,
Jan 24, 2006, 8:12:55 AM1/24/06
to
In article <1138102307.8...@z14g2000cwz.googlegroups.com>,
"richard develyn" <ric...@skaro.demon.co.uk> wrote:

> I'm sure this has been discussed before - as this was alluded to in the
> puzzle fairness thread (should I be able to jump off every bridge).
>
> Should I?
>
> Some more examples. In all cases where this has nothing to do with the
> game:
>
> Sould I be able to climb a tree anywhere where there is a forest?
>
> Should I be able to set fire to anything flammable if I have a fire
> somewhere?
>
> Should everything be breakable?
>
> Should everything have a description?
>
> What about a smell or taste or sound?
>
> Should everything which could be a surface / supporter be a surface /
> supporter? Every table? Every mantlepiece? Every chest of drawers?

Richard,

your answer is basically correct. But let me try another way of putting
it:

Many books on writing provide the answer to this question, and a quote
(supposedly from Einstein) can also help. Here are the two quotes I keep
in mind when I try to work out similar stuff for a story:

"When a gun is on the wall in Act I, Scene I, it should be fired by
Act III."

"Everything should be as simple as possible, but no simpler"

The first quote is usually read in reverse. I.e. if you want to fire a
gun in act III, make sure you've foreshadowed it in act I.

Essentially, these two quotes sum up to:

1) Everything used in the story should have been introduced before it is
actually used (otherwise it'll seem like a Deus ex Machina). In a game
that usually maps to providing clues instead of having the player die
and use afterlife knowledge to get around the death next time.

2) Everything that's not used in the story for a clear purpose should be
taken out. Mind you, I'm not saying a story (be it static or
interactive) should be terse and without backdrop. Instead, there needs
to be a purpose when backdrop is provided. Most authors do this
intuitively, but a good editor (or beta) usually has a talent for
spotting pieces of unnecessary exposition and interactions.

If a game reduces misleading distractions (like being able to set
everything afire even though the game only uses fire for light), people
playing it will never notice the restrictions placed on them.
Restrictions help guide the players along the most pleasing experience,
the main plot. If a story or a game don't provide this guidance, players
usually experience them as pointless, meandering, or plain boring.

Of course, some of the seeming distractions can be used for a purpose,
like providing more background on characters or the world, but if that's
used too much, players will get lost. Also, there are obviously red
herrings, whose whole purpose it is to be misleading. But then again,
those should be kept down as to not distract from the story.

Cheers,
-- Uli
http://www.zathras.de

Gregory Weir

unread,
Jan 24, 2006, 10:15:03 AM1/24/06
to
richard develyn wrote:
> Again I would like to canvas opinion on another aspect of game design
> which is how to deal with interactions which are not related to the
> game's narrative path.
>
> I'm sure this has been discussed before - as this was alluded to in the
> puzzle fairness thread (should I be able to jump off every bridge).
>
> Should I?

In my opinion, it depends on the game. Some games want a detailed
world with lots of incidental objects that you can put stuff on or
burn, while in some games anything not directly involved in the
puzzles/story feels like a red herring.

That's kind of a cop-out answer, isn't it? I guess it's just the old
simulationist/narrativist issue, really, which IMO should never be
solved.

Gregory Weir

Jeff Nyman

unread,
Jan 24, 2006, 12:51:07 PM1/24/06
to
Gregory Weir wrote:

> In my opinion, it depends on the game. Some games want a detailed
> world with lots of incidental objects that you can put stuff on or
> burn, while in some games anything not directly involved in the
> puzzles/story feels like a red herring.

I agree with this. Consider Infocom's "A Mind Forever Voyaging." There
you had boatloads of stuff you could look at (although, admittedly, the
interaction was sometimes quite limited). Not all of it was strictly
necessary for game completion at all, but it was there to provide
atmosphere or to provide the feeling of a distinctive world. On the
other hand, consider "Border Zone." There your interactions with
anything beyond the immediate narrative needs of the game was not
really allowed. You could still argue you had the distinctive feel of a
world but it was certainly a more limited one. That said: the narrative
structure of both games made it clear (I think) why these design
decisions may have been made.

I think it ultimately boils down to what your reader/player would like
and their expectations. (Those expectations are part of what you set up
with the game.) Think of a novel by Stephen King, for example. Here you
would tend to expect lots of characterization and description of
motivation. You might not expect that same level from a Clive Cussler
novel. In the latter, you expect more focus on the action elements. The
way a story is written at the start usually starts to clue a reader in,
regarding what to expect.

I am not sure if the same could be said for Interactive Fiction, but I
would think so. For example, you could subtly (or not so subtly)
indicate to your player that various elements of the "world" you have
created are not really there to be played around with or manipulated.
The idea is, rather, to go from point A to point B and then to point C
in a relatively directed fashion. (Somewhat like the "Border Zone"
concept.) Alternatively, I would think, you can make it clear that a
certain amount of realism is inherent in your world model and thus
encourage the player to try various things with various elements:
burning them, climbing them, etc. (Somewhat like the "AMFV" concept.)

Ultimately, though, I think what the player expects should be met to
some degree. If your game seems to indicate all these little details
about the world, those details should probably have some point or at
least the capability of being used to some purpose. Otherwise, the
player might get overloaded, as it were, and wonder why the heck all
these "annoying" things are in place when they serve no purpose.
Players, I am sure, tend to recognize they are in fact playing a game
and game-playing considerations/expectations surely have something to
say about this topic. Yet, the other side is also true: if you have set
expectations that many objects are useless (except as world-filling
detail), it could be bad for the player to then be expected to use just
one of those objects to solve a particular puzzle. Perhaps that is the
"foreshadowing" element mentioned by a previous poster; perhaps
"foreshadowing" in this case is "emphasis" on certain objects (or
certain interactions) over others, thus guiding player expectations to
a certain extent.

Just some thoughts. I am still thinking these things through myself, so
as you can tell what I am saying here is probably a little half-baked.

- Jeff

samwyse

unread,
Jan 24, 2006, 1:36:46 PM1/24/06
to
Uli Kusterer wrote:

> Many books on writing provide the answer to this question, and a quote
> (supposedly from Einstein) can also help.
>

> "When a gun is on the wall in Act I, Scene I, it should be fired by
> Act III."

Actually, it's Chekov, not Einstein. Makes more sense that way, since
he (Anton, not Pavel) was a playwright.

richard develyn

unread,
Jan 24, 2006, 1:55:01 PM1/24/06
to
I agree with the sentiment of what you are saying but I'm unsure to
what degree I can properly abstract the "player" as some sort of
objective entity. This is the equivalent of talking about convention,
which is why I'm canvasing opinion to see what the experience of other
designers here tell them that a typical "player" might expect.

Having everything have a description is a case in point. Following on
from your thoughts you might see this as something which depends on the
flavour of the game. However I get the impression that this is expected
of you whatever sort of game you write. Some of these conventions are
probably more an indication of competence / careful design - i.e. they
flag up to players whether your particular game is slap-dash or high
quality. It's important to know what these conventions are, though.
(Another one is not having messages along the lines of "You see the
policeman here." as opposed to "The policeman is leaning against the
lampost tying his shoe-laces." - or what have you.

Richard

Jeff Nyman

unread,
Jan 24, 2006, 2:52:38 PM1/24/06
to
richard develyn wrote:
> Having everything have a description is a case in point. Following on
> from your thoughts you might see this as something which depends on the
> flavour of the game.

Or, rather perhaps, the *usage* (as opposed to the *existence*) of
description itself is situational based on need. For example, a very
important gadget that the player has to manipulate quite a bit may be
given an extensive description. Another gadget that gives a little
"realism" to the game, but otherwise does nothing, may have a
description but not to that same level of detail.

Of course, whether or not a given player would "expect" that an object
with less detail is not so important is hard to guess, particularly
because it would depend a lot on the player, I would guess.

If anything is factored into your game, I would assume it should have a
description. Speaking as a player, I would dislike to be told that
something is in the room but then when I try to examine it, I get told
that I can see no such thing. Rather, I would expect some description.
However, again speaking as a player, if the description makes it clear
the object is not of much interest ("You see a bunch of these dusty
books; all of them tattered and unreadable") my expectations of using
the object are lessened.

In other words, just as an author of written fiction must set the
expectations of the reader by pointing out (through various means) what
is and what is not important in the story, I would think the same
applies to writers of interactive fiction. That, to me, seems to be at
least a large part of the convention. But, as you note, the design of
the game makes all the difference in terms of how this is done (i.e., a
"That's not important" message upon examining some widget).

Regarding description, I could see allowing for malleable descriptions
as well -- something I do not see many games do. For example, have the
description change based on what is happening in the game or what the
player has done. For example, the player examines some object and gets
this: "Upon closer examination, it looks like the dusty object is some
sort of control rod, but appears to be broken." Then if the player
tries things with the object and then examines it again, the
description could now be: "You've already determined that the control
rod is somewhat useless." This may not seem relevant to what you are
talking about but I think it might because in terms of simulation vs.
completeness, in the "real world" I would think we rarely examine some
object in the same level of detail each time we look at it. We build up
an internal description that we modify upon subsequent knowledge.

I think that speaks to your point 1 in your original post, but maybe
also your point 2: namely that if the narrative requires you to use an
object, you can constrain the expectations of the player in terms of
how to use the object by manipulating the description. (This is as
opposed to just saying something like: "You already used the scissors
to cut the only important piece of paper you are likely to find.") This
might also speak to your point 3: if the object eventually is shown not
to be useful other than as decoration, you can modify the initial
description to reflect that.

I think this can also apply to the usage of objects, not just their
description. That said, let's say you have a desk that is a supporter.
In that case, rather than saying should I allow the player to put
anything and everything on there, I would rather ask: why stop them?
Since the libraries tend to handle this logic automatically (assuming
there is not some limit to the items the supporter can hold), why not
just let the player put as much as they want on the desk, if they so
chose?

As far as breaking things/setting things on fire, let's say you have
fifty objects in your game that could be potentially flammable. Is it
better to allow the player to set fire to each of them to absolutely no
effect or is it better to simply set expectations by making it clear
that not everything can be set on fire, but giving subtle clues as to
when it might be logical to use fire. That said, I realize that is a
big balancing act and I agree with you: that makes your point 2 an
interesting issue in terms of design. It also speaks, I think, to
determining what may or may not be considered "logical to try" by the
player. I imagine that is what every designer is asking themselves:
what is the player likely to try to solve this problem? What
expectations have I set up that would lead them to think to combine
Object A and Object B?

Like you, I am curious to hear from people as well (both players and
designers).

- Jeff

Ashiq Alibhai

unread,
Jan 24, 2006, 10:55:52 PM1/24/06
to
I agree that simplicity is best. In general, don't spend extra effort
coding 50 flammable things when your player only needs to set one or
two on fire. This method also results in cleaner, shorter code -- less
bugs, and less potentially unexpected situations (they lit object X on
fire instead of object Y).

Fish

unread,
Jan 24, 2006, 11:42:58 PM1/24/06
to
I have yet to play any video game that made me forget I was sitting on
my butt playing a video game. Halo came close, but not quite.

Therefore, I tend to think in terms of game design rather than in
ultra-realism. I don't kid myself that I can ever fool any player into
thinking he's in the real world with infinite possibilities, because
even the dimmest player can imagine courses of action faster than I can
code answers to them. Pay no attention to the Fish behind the curtain, etc.

Thinking of IF as a game means you come at it from the angle of figuring
out how the player will act as he tries to gain some meaningful return
(score points, solve a puzzle defeat a level, etc).

Sometimes when a game is too difficult, the player decides the best
meaningful return is to turn the bloody game off and go watch TV. Or go
online and find a walkthrough or a hint book.

I believe everything in the room which is explicitly listed needs a
description and interactivity, even if it's "that isn't important."
Reasonable possibilities should be considered (if within the scope of
the game) or explicitly rejected (if not). Eliminating the impossible
is how players make logical choices about what to do next.

Some people really enjoy having multitudes of possibilities (as in
Morrowind). Some people I know *hated* Morrowind because they didn't
know what to do first. Paralysis through abundance of choice. But at
least in Morrowind, *anything* you did was productive, helped your
skills, earned you loot to sell, and so on.

The perfect simulation-IF, where anything is possible, means there are
1,000,000 options, 999,950 of which are crap. The player might feel as
if he's finding all kinds of really great things to do, but not one of
those things helps him a damn bit or advances to the end of the game.
At that point, he may turn off the bloody game to go watch TV.

I'm not like a bad Dungeon Master: I'm not trying to *defeat* the
player. I'm trying to entertain him. If someone likes to play
frustrating games where you have to start over endlessly, someone can
write 'em. :)

FISH

fel...@yahoo.com

unread,
Jan 25, 2006, 12:54:30 AM1/25/06
to
richard develyn wrote:

> I'm sure this has been discussed before - as this was alluded to in the
> puzzle fairness thread (should I be able to jump off every bridge).
>
> Should I?

Yes, I think you should, on one condition: the result should be instant
death, or otherwise game over. If you wake up at the hospital with no
chance of winning anymore, and you're not told, the game is broken.

That, in my opinion, makes all the difference.

Felix

ems...@mindspring.com

unread,
Jan 25, 2006, 4:19:06 AM1/25/06
to

richard develyn wrote:
> Having everything have a description is a case in point. Following on
> from your thoughts you might see this as something which depends on the
> flavour of the game. However I get the impression that this is expected
> of you whatever sort of game you write. Some of these conventions are
> probably more an indication of competence / careful design - i.e. they
> flag up to players whether your particular game is slap-dash or high
> quality.

Not having done anything beyond the default library behavior does
suggest a lack of polish, yes. But there are times when you might not
want a description-heavy game. I recall some good discussion on this
thread:

http://groups.google.com/group/rec.arts.int-fiction/browse_frm/thread/4a97fff85b8269dd/2148c2ac95636add#2148c2ac95636add

...and there are ways to get away with minimizing descriptions and
still signal to the player that you know what you're doing. Narcolepsy
and Book & Volume both take this course: B&V does so by replacing the
default "You don't see anything..." message with a different default of
its own.

The main thing, as noted elsewhere in this thread, is to let the player
know what kind of game this is and what kind of interaction is
necessary.

richard develyn

unread,
Jan 25, 2006, 8:50:20 AM1/25/06
to
I notice that TADS3 supports "examine all" - which presumably saves
time and frustration ...

Richard

ems...@mindspring.com

unread,
Jan 25, 2006, 4:43:14 PM1/25/06
to

Well, maybe. In a really dense room I'm not sure I'd be thrilled about
reading through two pages of descriptive text all at once, either.

Eric Eve

unread,
Jan 25, 2006, 5:12:58 PM1/25/06
to
<ems...@mindspring.com> wrote in message
news:1138225393....@g44g2000cwa.googlegroups.com...

Agreed. And as an author I'm not sure I'd be thrilled about this
short-circuiting of how I'd want a player to experience the game (in
the worst case, it could virtually act as a spoiler). But
fortunately, TADS 3 also lets authors disable the use of ALL in such
cases.

-- Eric


pesononline

unread,
Jan 26, 2006, 3:19:01 AM1/26/06
to
The problem with applying Chekov's Gun to IF is that it's not always
clear whether or not that gun actually *is* going to be fired later on.


So what was meant to be an elegant dramatic device of foreshadowing
becomes a... well, not a non sequitur, but the opposite: you hint at
things, and then through some plot decisions made down the line that
stuff never actually happens. So your subtle hint is lost. It's not
really a red herring, because it did point to something that could
happen. Maybe it's a blue herring: a hint that you could have followed
but didn't. Like blue hyperlinks, get it?

Of course you can claim that you're not hinting at things to come, but
merely at the potential of things to that might happen. And of course
there are plenty of things that you do know will happen later on in
most IF as well (in some, you know pretty much *everything* that will
happen, at least anything important enough to warrant a hint early on).

And you can claim that life is full of opportunities you don't pursue.
But that's largely irrelevant: Chekov's Gun is about drama, not about
life. Actually what Chekov said about the gun was pretty much said by
Aristotle some years before anyway. (_Poetics_, VII, about the unity of
plot - you can find it online.)

-- peson

Samwyse

unread,
Jan 26, 2006, 7:48:06 AM1/26/06
to
pesononline wrote:
> The problem with applying Chekov's Gun to IF is that it's not always
> clear whether or not that gun actually *is* going to be fired later on.

The vast majority of works of IF have only one intended ending; any
other endings are due to the player not completing the game. In that
case, the author's intent is that the "gun" will get used while the
other cases are similar to leaving the play during intermission. There
are a very few games where the author's intent is to provide multiple
endings. In those cases, yes, the hangman's noose may go unused, but
those stories risk having less dramatic impact, simply because the
author is refusing to make a stand.

> So what was meant to be an elegant dramatic device of foreshadowing
> becomes a... well, not a non sequitur, but the opposite: you hint at
> things, and then through some plot decisions made down the line that
> stuff never actually happens. So your subtle hint is lost. It's not
> really a red herring, because it did point to something that could
> happen. Maybe it's a blue herring: a hint that you could have followed
> but didn't. Like blue hyperlinks, get it?

I'm not sure what the games are that you're playing, but I can think of
few games where the herrings, red or blue, outnumber the usable items.
That's why players tend to act like kleptomaniacs, picking up and
carrying everything that they find in the hopes that it will be used
later. In fact, there are lots of games where things get used multiple
times, which train players to never discard anything. They instead
become pack-rats, carrying around the stumps of burned out candles out
of fear that the wax will be needed later. That's where "Simulation vs
convention vs completeness" helps an author decide what to do. If there
actually is a need for the wax later, using the candle should leave a
stump; if not, then the using the candle should leave no residue.

> Of course you can claim that you're not hinting at things to come, but
> merely at the potential of things to that might happen. And of course
> there are plenty of things that you do know will happen later on in
> most IF as well (in some, you know pretty much *everything* that will
> happen, at least anything important enough to warrant a hint early on).
>
> And you can claim that life is full of opportunities you don't pursue.
> But that's largely irrelevant: Chekov's Gun is about drama, not about
> life. Actually what Chekov said about the gun was pretty much said by
> Aristotle some years before anyway. (_Poetics_, VII, about the unity of
> plot - you can find it online.)

Again, I'm not sure what the heck you're talking about. This entire
discussion is about dramatic devices and how to use them in a story.
Simulationists generally write bad IF, because they are so busy modeling
life that they fail to introduce any drama. Having a kitchen in a story
does not mean you should model a refridgerator containg a dozen eggs and
a six-pack of beer, unless Act Three requires getting someone drunk and
feeding them an omelet.

Neil Cerutti

unread,
Jan 26, 2006, 8:22:35 AM1/26/06
to

In the interest of Inform programmers, who may be drawing a wrong
conclusion here, Inform allows authors to enable Examine All,
also. Consult guide about ChooseObjects(0, obj) and
ChooseObject(1, obj).

--
Neil Cerutti

pesononline

unread,
Jan 26, 2006, 12:45:58 PM1/26/06
to
Yeah, reading my own post I see that I really wasn't being very
coherent.

You're of course right that most IF works will generally just keep
moving down the track, so the problem of not being able to predict
what's going to happen is not really an issue. I had a bunch of CRPG
ideas in my head messing with my thinking.

And although it's hard to glean from my post, I do agree with you.
That's why I was talking about unity of plot - unless the eggs and the
frying pan are important to the plot, just leave 'em out. That's why it
makes sense to talk about (unintentional) red herrings at all - it's
not that they don't belong in the game world, it's that they don't
belong in the story.

Richard Bos

unread,
Jan 30, 2006, 7:00:00 PM1/30/06
to
"richard develyn" <ric...@skaro.demon.co.uk> wrote:

> I'm sure this has been discussed before - as this was alluded to in the
> puzzle fairness thread (should I be able to jump off every bridge).

No.

> Some more examples. In all cases where this has nothing to do with the
> game:
>
> Sould I be able to climb a tree anywhere where there is a forest?

No.

> Should I be able to set fire to anything flammable if I have a fire
> somewhere?

No.

> Should everything be breakable?

No.

But in all three cases, you should be able to _try_. Being told that you
can't manage to find a hand-hold/that would be dangerous/you might need
that vase/you don't want to wake everybody up is fine.

> Should everything have a description?

Yes. But not every sub-object of everything. If you're in a forest, the
trees should have a description; if that description mentions leaves,
it's OK by me if EXAMINE LEAVES results in the same description as for
the trees.

> What about a smell or taste or sound?

Only where appropriate. "You smell nothing special." is perfectly
adequate, indeed probably perfectly accurate, of a door key or a piece
of printer paper.

> Should everything which could be a surface / supporter be a surface /
> supporter? Every table? Every mantlepiece? Every chest of drawers?

Now you're getting into rapidly increasing complexity for rapidly
decreasing result. Tables, yes, you expect to be able to put things on
them. But every little ledge? No. Does the user take something out of a
drawer? Then he should either be able to put something in, or be given a
realistic reason why he can't (e.g., he's snooping around in someone
else's desk). At the same time you want to be careful that you don't let
your user put a folding ladder inside a desk drawer. It may fold and be
portable, but it's not going to fold _that_ small.

Richard

Reply all
Reply to author
Forward
0 new messages