Unwinnable situations

9 views
Skip to first unread message

Mike Sousa

unread,
Mar 18, 2000, 3:00:00 AM3/18/00
to

To what extent should an author stop the PC from entering an unwinnable
situation? Should the game just let the PC continue with the game, thus
letting the PC figure it out? Should the game display subtle messages?
more direct message? Don't let it become unwinnable? Kill the PC/end
the game after a few moves?

I'm leaning towards limiting the situations and/or ending the game with
an appropriate sequence. Feedback welcomed.

-- Mike


BrenBarn

unread,
Mar 18, 2000, 3:00:00 AM3/18/00
to
I firmly advocate the creation of games with no unwinnable situations.
(Hey I kept it brief, for once :-)

From,
Brendan B. B. (Bren...@aol.com)
(Name in header has spam-blocker, use the address above instead.)

"Do not follow where the path may lead;
go, instead, where there is no path, and leave a trail."
--Author Unknown

Robb Sherwin

unread,
Mar 18, 2000, 3:00:00 AM3/18/00
to
On Sat, 18 Mar 2000 01:21:10 GMT, Mike Sousa <mso...@efortress.com>
wrote:

>To what extent should an author stop the PC from entering an unwinnable
>situation? Should the game just let the PC continue with the game, thus
>letting the PC figure it out?

My personal opinion is that it is difficult enough for interactive
fiction to come off as remotely approachable... if your audience is
purely hardcore r*ifers, then maybe you have a different slant. But if
not, the only thing I've ever envisioned after playing a game where
that happened to me is a mental image of the author suddenly getting a
brainwave that I completely messed up and they do a sudden spit-take
because they are laughing at me so hard.

There are exceptions, certainly, but I think making it 90% of the way
through a game and finding out that you forgot to deal a trebled fromp
on turn three means you have to play it all over again makes you feel
like a horse's ass. As an author, this is usually a feeling I do not
want my player to have.

> Should the game display subtle messages? more direct message?

Here it because more difficult. I'm playing through "Doppelganger"
right now. If you start machine gunning people before you leave your
head quarters then you are instantly told that the other person draws
*their* weapon and shoots you in the face, ending the game.

It could go on for a bit, I suppose. I could take out the librarian
and then the rest of the guys in the building could track me down like
a filthy, foaming cur. But it's a question of scope. Do you, as an
author, want to program all that stuff?

> Don't let it become unwinnable?

This way is always fair. I like how Andrew usually tells us right from
the get-go if it's possible to get his game in an unwinnable state. If
I know that going in then I'll keep a lot of different save game
files.

If you explain the rules to your players beforehand then your game
will come off as fair. If, at that point, they don't read the manual
or type "about" the first time they play the game, then that is their
own fault.

-- Robb


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Robb Sherwin, Fort Collins CO
Reviews From Trotting Krips: http://ifiction.tsx.org
Knight Orc Home Page: www.geocities.com/~knightorc

Mike Sousa

unread,
Mar 18, 2000, 3:00:00 AM3/18/00
to

Robb Sherwin wrote:

>
> .....<snip>, but I think making it 90% of the way


> through a game and finding out that you forgot to deal a trebled fromp
> on turn three means you have to play it all over again makes you feel
> like a horse's ass. As an author, this is usually a feeling I do not
> want my player to have.

I agree. I have spent many extra hours going through the story line to make
sure this doesn't happen. But on the flip side, I don't want to make it
idiot proof. There's a fine line I guess.

>
> It could go on for a bit, I suppose. I could take out the librarian
> and then the rest of the guys in the building could track me down like
> a filthy, foaming cur. But it's a question of scope. Do you, as an
> author, want to program all that stuff?
>

As an author, I want the PC to enjoy the game. If that means I have to
program all that stuff, then I guess it's not a matter of want, but need?

-- Mike


The Solar Echo

unread,
Mar 20, 2000, 3:00:00 AM3/20/00
to
I have a particular hatred for games that you can render unwinnable without
proper notification. I struggled with the Infocom classic Starcross for nearly
a year after I had rendered the game unwinnable. There are 2 ways to solve a
particular puzzle, and because I solved it the wrong way, it was impossible for
me to win the game. Only after I obtained a copy of the hints did I realise my
error. I worked long and hard to finish the game without hints, and came very
close to doing so. Being defeated by an unfair unwinnable situation left a
bitter taste in my mouth.

> Should the game just let the PC continue with the game, thus
>letting the PC figure it out?

NO! Unless you want the player to be disgusted with you and never pick up one
of your games again. The only exception to this is when the player does
something extremely stupid such as throwing their inventory off a cliff.


((((((:. The Solar Echo .:))))))

Andreas Hoppler

unread,
Mar 20, 2000, 3:00:00 AM3/20/00
to
Actually, I liked that particular puzzle idea in Starcross very much. Can't say
much more without spoilers.

Andreas

J.D. Berry

unread,
Mar 20, 2000, 3:00:00 AM3/20/00
to
The Solar Echo wrote:
>
> I have a particular hatred for games that you can render unwinnable
> without proper notification.

Andreas Hoppler <ahop...@bluewin.ch> wrote:

> Actually, I liked that particular puzzle idea in Starcross very much.
> Can't say much more without spoilers.


There's a meta-adventuring mode that a player needs to have to be able
to enjoy a lot of I-F out there. "I can't figure out how to proceed,
but based on other games THIS seems to be what I need to do..."
I'm definitely on the side of not liking unwinnable situations, but the
more I play the more that instinct of "I can now no longer win" kicks
in.

What I don't like about unwinnable (the kind where you have to go
back farther than a few turns) states is the break in mimesis. I was
absorbed in a story that is now a puzzle in and of itself. I'm no
longer "into" it.

All said, I think walkthroughs are very nice to have for the beginning
player and to encourage them to play more games. Oh, "show" is a
command that's useful. Oh, sometimes you can try brute force. Oh,
the sarcastic reply I got in one case doesn't mean it won't work in
another.

I know I'm treading on another thread now about walkthroughs and whether
or not to have them. Just add another vote "for." ;-)

Jim


Sent via Deja.com http://www.deja.com/
Before you buy.

Adam 'Jw' Biltcliffe

unread,
Mar 20, 2000, 3:00:00 AM3/20/00
to

The Solar Echo <thesol...@aol.com> wrote in message
news:20000320135654...@ng-xe1.aol.com...

> I have a particular hatred for games that you can render unwinnable
without
> proper notification. I struggled with the Infocom classic Starcross
for nearly
> a year after I had rendered the game unwinnable. There are 2 ways to
solve a
> particular puzzle, and because I solved it the wrong way, it was
impossible for
> me to win the game. Only after I obtained a copy of the hints did I
realise my
> error. I worked long and hard to finish the game without hints, and
came very
> close to doing so. Being defeated by an unfair unwinnable situation
left a
> bitter taste in my mouth.

IMO, the acceptable-ness of making a game unwinnable depends on how
obvious it is that you have done so. Actually allowing you to 'solve' a
particular puzzle, but rendering the game unwinnable in the process, is
an extreme example of *not* making this obvious, and it seems to me that
this is a perfectly valid reason for disliking the game, unless you know
in advance that some puzzle solutions might make the game unwinnable.
However, if, for example, the PC elects to walk along a road rather than
drive, and then after a while sees a distant bridge collapse that he or
she could have made it to in time by taking the car, I think it would be
fair to allow the player to continue with the now-unwinnable game.


Jw

Stewmail

unread,
Mar 20, 2000, 3:00:00 AM3/20/00
to
Hello Everyone,

I must say I agree with Mr. Echo. It drives me crazy. Also sudden death
which just means starting again (unless saved).

--
From Stewart in England ( o )===#
stew...@btinternet.com


The Solar Echo <thesol...@aol.com> wrote in message
news:20000320135654...@ng-xe1.aol.com...
> I have a particular hatred for games that you can render unwinnable
without
> proper notification. I struggled with the Infocom classic Starcross for
nearly
> a year after I had rendered the game unwinnable. There are 2 ways to
solve a
> particular puzzle, and because I solved it the wrong way, it was
impossible for
> me to win the game. Only after I obtained a copy of the hints did I
realise my
> error. I worked long and hard to finish the game without hints, and came
very
> close to doing so. Being defeated by an unfair unwinnable situation left
a
> bitter taste in my mouth.
>

The Solar Echo

unread,
Mar 20, 2000, 3:00:00 AM3/20/00
to
>Actually allowing you to 'solve' a
>particular puzzle, but rendering the game unwinnable in the process, is
>an extreme example of *not* making this obvious, and it seems to me that
>this is a perfectly valid reason for disliking the game, unless you know
>in advance that some puzzle solutions might make the game unwinnable.
>However, if, for example, the PC elects to walk along a road rather than
>drive, and then after a while sees a distant bridge collapse that he or
>she could have made it to in time by taking the car, I think it would be
>fair to allow the player to continue with the now-unwinnable game.

Yes, I agree with this very much. What made it so difficult is that I was
allowed to solve the puzzle in a way that made the game unwinnable, and was
given no warning. Then I was allowed to solve many more puzzles so that by the
time I got to the place where I was stuck, it was unclear where I had screwed
up, or if maybe there was just something I was missing.

My suggestion is to not let the game progress TOO much farther before making it
obvious that the game cannot be won. If I find out only at the end that I
can't win the game, I will probably not bother to go back and solve the puzzles
all over again.

Daniel Barkalow

unread,
Mar 21, 2000, 3:00:00 AM3/21/00
to
On Sat, 18 Mar 2000, Mike Sousa wrote:

> To what extent should an author stop the PC from entering an unwinnable

> situation? Should the game just let the PC continue with the game, thus
> letting the PC figure it out? Should the game display subtle messages?
> more direct message? Don't let it become unwinnable? Kill the PC/end
> the game after a few moves?
>
> I'm leaning towards limiting the situations and/or ending the game with
> an appropriate sequence. Feedback welcomed.

I've been playing So Far, and I like how that handles this: the area
you've messed up becomes pretty clearly broken in some way. You can keep
trying to do things with it, but they won't help. You can go elsewhere and
try to make progress in that way, but you know you've gotten stuck, and
are just playing to work on other puzzles.

Also, with the present bias towards relatively short games and can be
replayed quickly and easily, if the player becomes aware that the game has
become unwinnable for some reason, they can start over without too much
annoyance.

-Iabervon
*This .sig unintentionally changed*


Nicholas Shore

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
"The Solar Echo" <thesol...@aol.com> schrieb im Newsbeitrag
news:20000320173609...@ng-xe1.aol.com...

> >Actually allowing you to 'solve' a
> >particular puzzle, but rendering the game unwinnable in the process, is
> >an extreme example of *not* making this obvious, and it seems to me that
> >this is a perfectly valid reason for disliking the game, unless you know
> >in advance that some puzzle solutions might make the game unwinnable.
> >However, if, for example, the PC elects to walk along a road rather than
> >drive, and then after a while sees a distant bridge collapse that he or
> >she could have made it to in time by taking the car, I think it would be
> >fair to allow the player to continue with the now-unwinnable game.
>
> Yes, I agree with this very much. What made it so difficult is that I was
> allowed to solve the puzzle in a way that made the game unwinnable, and
was
> given no warning. Then I was allowed to solve many more puzzles so that
by the
> time I got to the place where I was stuck, it was unclear where I had
screwed
> up, or if maybe there was just something I was missing.
>

What I would really like is alternate solutions when you don't solve a
particular puzzle.
The most realistic turn for a game to take is to just let it run on, and
then give it a different ending. This isn't always easy, but theoretically
you could even display a winning message if the PC dies in the end. After
all, it happens often enough in real-life.
The art about that is giving the PC some satisfaction in any ending, and
that means: no score, no stupid two line ending messages of the style "you
are dead. game over." A truly good author could make death sound like a
triumph.

The frustrating thing is that usually in an unwinnable situation nothing
happens. To me the most obvious hint is usually if I wander around for ages
and I can't do anything (which is rather boring on the long run). Or I get
killed with the typical death message, which is rather frustrating.

Personally I should try to avoid this totally as an author. But I know its
difficult. Especially if you want the game to seem a bit realistic. One
thing that should definitely be avoinded are critical-item-unavailable
situations. All Items necessary for a puzzle should be accessable at the
time when the puzzle comes up. Anything else is inexcusable (to me).

> My suggestion is to not let the game progress TOO much farther before
making it
> obvious that the game cannot be won. If I find out only at the end that I
> can't win the game, I will probably not bother to go back and solve the
puzzles
> all over again.
>

Now TOO far is such an stretchable definition again. But of course you are
right.

Nicholas

----------------------------------------------------------------------------
-------------------------------
Visit my homepage:
http://members.xoom.com/ndsweb/index.htm
for literature, sci-fi-reviews (both planned), chemistry page (only german
so far), earning money on the web
----------------------------------------------------------------------------
--------------------------------

L. Ross Raszewski

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
On Wed, 22 Mar 2000 01:04:29 +0100, Nicholas Shore <nds.bl...@gmx.de> wrote:
>What I would really like is alternate solutions when you don't solve a
>particular puzzle.
>The most realistic turn for a game to take is to just let it run on, and
>then give it a different ending. This isn't always easy, but theoretically
>you could even display a winning message if the PC dies in the end. After
>all, it happens often enough in real-life.

Um. On average, I think it happens to most people exactly once ever.

--
"You should be women, and yet your beards forbid me to interpret that you are
so." -- Shakespeare, Macbeth I.iii

Nick Montfort

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
In article <8b9lq0$onj$1...@love.loyola.edu>,

lrasz...@loyola.edu wrote:
> On Wed, 22 Mar 2000 01:04:29 +0100, Nicholas Shore
<nds.bl...@gmx.de> wrote:
> >you could even display a winning message if the PC dies in the end.
> >After all, it happens often enough in real-life.
>
> Um. On average, I think it happens to most people exactly once ever.

The suggestion here is that when people live a good life and die, the
last thing they see is **** YOU HAVE WON **** ?

-Nick M.

The Solar Echo

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
>What I would really like is alternate solutions when you don't solve a
>particular puzzle.
>The most realistic turn for a game to take is to just let it run on, and
>then give it a different ending.

YES! Nicholas is talking about exactly what I've had on my mind. If there's
one thing I want to accomplish in I-F, it's to create I-F where the object is
not to "win", but to be involved in a story and affect the outcome with your
actions. THIS is the potential power I-F has over strictly linear books and
movies (IMHO)...the ability to change how it turns out. Isn't that why we call
it INTERACTIVE fiction?

Personally, I don't want to create games. I want to tell stories. I don't
want to fill them with puzzles. I want to fill them with situations that the
player/reader can make choices about, just like in real life.

Sorry (grin). I'm just really excited to see the thoughts that have been
spinning around my brain for months coming out of someone else! Thanks.

Daryl McCullough

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
thesol...@aol.com says...

>
>>What I would really like is alternate solutions when
>>you don't solve a particular puzzle. The most realistic
>>turn for a game to take is to just let it run on, and
>>then give it a different ending.
>
>YES! Nicholas is talking about exactly what I've had on my mind.
>If there's one thing I want to accomplish in I-F, it's to create
>I-F where the object is not to "win", but to be involved in a
>story and affect the outcome with your actions. THIS is the
>potential power I-F has over strictly linear books and movies
>(IMHO)...the ability to change how it turns out. Isn't that
>why we call it INTERACTIVE fiction?

>Personally, I don't want to create games. I want to tell stories.

The problem is that as the author of an interactive fiction,
you can't possibly anticipate all everything that a player will
do, and so you can't guarantee that all choices made by the
player will lead to *interesting* stories (unless you reduce
the interactivity by giving the player fewer choices). So,
if certain paths lead to interesting stories, and other
paths lead to boring stories, players finding a path of the
first type will *feel* like winners, and those who don't
find such a path will feel like losers---whether or not
you *call* them winners or losers.

Daryl McCullough
CoGenTex, Inc.
Ithaca, NY


The Solar Echo

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
>The problem is that as the author of an interactive fiction,
>you can't possibly anticipate all everything that a player will
>do, and so you can't guarantee that all choices made by the
>player will lead to *interesting* stories (unless you reduce
>the interactivity by giving the player fewer choices). So,
>if certain paths lead to interesting stories, and other
>paths lead to boring stories, players finding a path of the
>first type will *feel* like winners, and those who don't
>find such a path will feel like losers---whether or not
>you *call* them winners or losers.
>
>Daryl McCullough
>CoGenTex, Inc.
>Ithaca, NY
>

Yes, it is a problem. I realise that it would be almost impossible to create a
truely non-linear game. I also know that I can't anticipate every action that
a player will try, but hopefully (especially with a few good beta testers) I
can anticipate MOST of them, and while there may be a few possible "boring"
stories, they will be in the vast minority. I'm also of the opinion that good
writing can make even the "boring" stories somewhat interresting. (see Aisle,
by Sam Barlow)

I'm also in favor of giving the player fewer choices when necessary, especially
if it mirrors real life (on an airplane, our choices of movement are limited to
a few areas). While not perfect, I thought Adam Cadre's I-O did a good job of
this. The main character has fewer movement choices because she is stuck in
the middle of the desert. Knowing that to wander off into the desert would be
suicide, she's limited to areas of civilization, but still has quite a few
options, and at least the illusion of different outcomes.

okbl...@my-deja.com

unread,
Mar 22, 2000, 3:00:00 AM3/22/00
to
In article <8bb23d$1i...@edrn.newsguy.com>,
da...@cogentex.com (Daryl McCullough) wrote:

[snip, cut, paste, rearrange]


> if certain paths lead to interesting stories, and other
> paths lead to boring stories, players finding a path of the
> first type will *feel* like winners, and those who don't
> find such a path will feel like losers---whether or not
> you *call* them winners or losers.

This is undeniable. Of course, it's true even if you have only *one*
path that leads to a boring story.

> The problem is that as the author of an interactive fiction,
> you can't possibly anticipate all everything that a player will
> do, and so you can't guarantee that all choices made by the
> player will lead to *interesting* stories (unless you reduce
> the interactivity by giving the player fewer choices). So,

You don't have to anticipate *everything* a player will do. Right now,
plot-points/puzzles tend to be broken down into success or failure,
where success requires the exact right combination of inventory and
action. One could have two possible successful outcomes, where one of
the successful outcomes works out *despite* the player having made some
sort of mistake.

This could be done by providing the player with a missing item, as in
<<"I thought you might need this," she says, pulling the key to the door
from her purse.>> It might be tougher without NPCs, depending on the
circumstances, and it would be tougher if the player is supposed to
build some sort of Rube Goldberg contraption. But any puzzle or
plot-point trigger boils down to whether the player takes the correct
(finite number of) actions with any necessary (finite number of) items,
and programming a little leeway into these points couldn't be bad.

Or so it seems to me.
--
[ok]

BrenBarn

unread,
Mar 23, 2000, 3:00:00 AM3/23/00
to
>What I would really like is alternate solutions when you don't solve a
>particular puzzle.
>The most realistic turn for a game to take is to just let it run on, and
>then give it a different ending. This isn't always easy, but theoretically
>you could even display a winning message if the PC dies in the end. After
>all, it happens often enough in real-life.
>The art about that is giving the PC some satisfaction in any ending, and
>that means: no score, no stupid two line ending messages of the style "you
>are dead. game over."
Hear, hear! This is the holy grail of the whole medium, for me. But as
you say, it's not easy. Still, I think the ultimate in a game would be one
where you never noticed that you had done something "wrong" because there were
no "wrong" things; just "different" things, leading to different paths.

>A truly good author could make death sound like a
>triumph.

This is a very intriguing concept. I think about it a lot in terms of
plain old fiction, but applying to an IF work could be interesting.

>Personally I should try to avoid this totally as an author. But I know its
>difficult. Especially if you want the game to seem a bit realistic

I hear ya.

>One
>thing that should definitely be avoinded are critical-item-unavailable
>situations. All Items necessary for a puzzle should be accessable at the
>time when the puzzle comes up. Anything else is inexcusable (to me).

I agree that the "necessary" items shouldn't be locked away, inaccessible.
But I think an alternative viewpoint would be one in which the "necessary"
items are in fact defined as those accessible. In other words, if you don't
have the golden key, but you have the banana, the puzzle becomes a different
puzzle that can be solved in a different way by using the banana. If you have
both, either one may work, if applied correctly. Of course, for "traditional"
puzzles (open the door, acquire the object, etc.) this is somewhat difficuly to
implement (it's hard to find a plausible explanation for letting the player
unlock a door with a banana :-), but for more "wordly" puzzles -- puzzles that
are more subtly inserted into the game world -- it would be easier. Then
again, those worldly puzzles are hard to devise anyway.

BrenBarn

unread,
Mar 23, 2000, 3:00:00 AM3/23/00
to
>
>Personally, I don't want to create games. I want to tell stories. I don't
>want to fill them with puzzles. I want to fill them with situations that the
>player/reader can make choices about, just like in real life.
Man, me too. I like the idea of letting the player actually CREATE the
story by their actions (or letting them think they are), rather than letting
them move through a predefined one.

Sean T Barrett

unread,
Mar 23, 2000, 3:00:00 AM3/23/00
to
BrenBarn <bren...@aol.comRemove> wrote:
>[lost attribution]

>>Personally, I don't want to create games. I want to tell stories.
[snip]

> Man, me too. I like the idea of letting the player actually CREATE the
>story by their actions

These are two totally different things.

I snipped enough text that this is a fairly unfair criticism of
the posts as they were originally written, as they did have
some specifics in common. But I don't think you actually
can meet both goals with the same tools. Every step taken
towards player freedom reduces your ability to tell a story.

SeanB

BrenBarn

unread,
Mar 24, 2000, 3:00:00 AM3/24/00
to
>Every step taken
>towards player freedom reduces your ability to tell a story.
Not necessarily true, unless you stipulate that you're trying to tell THE
SAME story. Player freedom merely lets (or forces, depending on your point of
view) the author tell more stories with the same game.

The Solar Echo

unread,
Mar 24, 2000, 3:00:00 AM3/24/00
to

>>Every step taken
>>towards player freedom reduces your ability to tell a story.
> Not necessarily true, unless you stipulate that you're trying to tell
>THE
>SAME story. Player freedom merely lets (or forces, depending on your point
>of
>view) the author tell more stories with the same game.
>
>From,
>Brendan B. B. (Bren...@aol.com)


Nicely put.

Sean T Barrett

unread,
Mar 24, 2000, 3:00:00 AM3/24/00
to
BrenBarn <bren...@aol.comRemove> wrote:

>Sean Barret <buz...@world.std.com> wrote:
>>Every step taken
>>towards player freedom reduces your ability to tell a story.
> Not necessarily true, unless you stipulate that you're trying to tell THE
>SAME story. Player freedom merely lets (or forces, depending on your point of
>view) the author tell more stories with the same game.

Every step you take towards telling more stories requires you
to put less effort into telling each of those stories. If you actually
allow for exponential branching, it becomes impossible for
an author to meaningfully 'tell' a story beyond a certain
depth of branching. If you allow branches to merge, you still
are telling an exponential number of stories that happen to
have similar bits of plot later on, but unless your writing
accounts for the exponential branches, you will tell the story
poorly.

"Hunter, in Darkness" was a nice example of a branch-and-merge.
"9:05" is a good example of how you can write text that coheres
with two different storylines. Both of those were written by
very talented, very experienced IF authors, and neither attempts
any significant *depth* of branching at all. I have yet to see
any evidence that it's possible for an author to tell a significant
number of stories, and an awful lot of evidence of narrow storying.
There's no evidence it can be done and it doesn't particularly
*sound* tractable, so why would you think it was?

Looking at commercial games, there is a trend towards telling
a totally linear story while providing a simulation with player
simulation on a totally orthogonal axis. The Final Fantasy
style games are probably the best example of this; a piss-poor
adventure game (you have essentially no freedom or control)
coupled with a deep simulation for the hack and slash combats--total
freedom to pick your weapons, spells, equipment, combat strategies,
etc. Thief makes an explicitly linear narrative interwoven with
its linear mission structure, but within a single mission the
player has great freedom. The player can tell stories about the
events that transpired during a mission, and they aren't necessarily
authored stories--they're just the things that the player did and
how the simulation interacted with the player, but they're personal
to the player in a way that authored stories aren't. I cannot picture
how you could ever meaningfully author that many stories.

I'm not denying there's a *continuum* here. But the two extremes
are opposites and are incompatible.

SeanB

Sean T Barrett

unread,
Mar 24, 2000, 3:00:00 AM3/24/00
to
In article <Frxv0...@world.std.com>,

>BrenBarn <bren...@aol.comRemove> wrote:
>"Hunter, in Darkness" was a nice example of a branch-and-merge.
>"9:05" is a good example of how you can write text that coheres
>with two different storylines. Both of those were written by
>very talented, very experienced IF authors, and neither attempts
>any significant *depth* of branching at all. I have yet to see
>any evidence that it's possible for an author to tell a significant
>number of stories, and an awful lot of evidence of narrow storying.

I meant to mention this but spaced it. "Aisle" is definitely worth
mentioning here. Obviously the author skipped the exponential
branching problem and simply created a very large number of
one-way branches, but you could imagine those endings on a
deeper set of fewer branches. Except it's clearly far more
tractable to *not* have to keep track of the branches, to do
the "Aisle" thing.

One can certainly argue that "Aisle" proves you can tells lots
of stories which are explicitly authored, but as an example it
certainly plays into my claim of "less work per story"; not to
criticize the *quality* of the Aisle "stories", but rather to
point out their length. (Clearly Aisle is a bad example of
this in practice, since it's *meant* to be replayed, and multiple
plays are for the most part revealing facets of one larger
story.)

SeanB

Sean T Barrett

unread,
Mar 24, 2000, 3:00:00 AM3/24/00
to
Sean T Barrett <buz...@world.std.com> wrote:
>In article <Frxv0...@world.std.com>,
>>BrenBarn <bren...@aol.comRemove> wrote:
>>"Hunter, in Darkness" was a nice example of a branch-and-merge.

Oops, sorry, I was following myself up but the attribution got munged.
Now I'm following up my own followup of myself.

SeanB

BrenBarn

unread,
Mar 25, 2000, 3:00:00 AM3/25/00
to
>Every step you take towards telling more stories requires you
>to put less effort into telling each of those stories. If you actually
>allow for exponential branching, it becomes impossible for
>an author to meaningfully 'tell' a story beyond a certain
>depth of branching. If you allow branches to merge, you still
>are telling an exponential number of stories that happen to
>have similar bits of plot later on, but unless your writing
>accounts for the exponential branches, you will tell the story
>poorly.
I agree that this has thus far been the case. As I too often do, I was
speaking theoretically. Also, I feel I have misrepresented myself in two ways.
One, I am not saying that I advocate the inclusion of multiple storylines
in oppoisition to the in-depth development of a single plot; I think each could
be equally fulfilling for both player and author. I do think (again,
theoretically) that, because the multiple-story approach has, as you say, been
historically less common, it would be nice to explore the possibilities of that
concept.
Two, I feel that I am using story in perhaps too broad a sense. A
"multiple-story" game (i.e., one in which the player has lots of freedom to
alter the events of the game) may in fact be telling a single "story". This
"story" may not be a linear narrative, but instead be an overall picture of a
person, place, society, time period, etc. (For a literary example, check out
_Dictionary of the Khazars_, by Milorad Pavic, which has been discussed on this
newsgroup before.) I don't know exactly what term to use for the subject of
such a game; story seems inaccurate. Perhaps I could follow Mr. Pavic's
example and call it a "lexicon".

From,
Brendan B. B. (Bren...@aol.com)

YesuSlave

unread,
Mar 25, 2000, 3:00:00 AM3/25/00
to
>>Every step you take towards telling more stories requires you
>>to put less effort into telling each of those stories.

Perhaps, however, I believe if you actually work at telling the story instead
of "getting the game out" you may be able to focus on some great deal of depth,
but perhaps you could only work on one story at a time.

If you actually
>>allow for exponential branching, it becomes impossible for
>>an author to meaningfully 'tell' a story beyond a certain
>>depth of branching.

Why does it become impossible?

If you allow branches to merge, you still
>>are telling an exponential number of stories that happen to
>>have similar bits of plot later on, but unless your writing
>>accounts for the exponential branches, you will tell the story
>>poorly.

Again, why? Are you saying that a writer is incapable of doing this simply
because no one has before?

> I agree that this has thus far been the case. As I too often do, I was
>speaking theoretically. Also, I feel I have misrepresented myself in two
>ways.
> One, I am not saying that I advocate the inclusion of multiple
>storylines
>in oppoisition to the in-depth development of a single plot; I think each
>could
>be equally fulfilling for both player and author. I do think (again,
>theoretically) that, because the multiple-story approach has, as you say,
>been
>historically less common, it would be nice to explore the possibilities of
>that
>concept.

This is exactly what I am doing now...I am a bit paranoid, so I don't talk
about my games before I release them, but I expect to work long hours on this
particular game, but I don't see that being something that should stop me.


> Two, I feel that I am using story in perhaps too broad a sense. A
>"multiple-story" game (i.e., one in which the player has lots of freedom to
>alter the events of the game) may in fact be telling a single "story". This
>"story" may not be a linear narrative, but instead be an overall picture of a
>person, place, society, time period, etc. (For a literary example, check out
>_Dictionary of the Khazars_, by Milorad Pavic, which has been discussed on
>this
>newsgroup before.) I don't know exactly what term to use for the subject of
>such a game; story seems inaccurate. Perhaps I could follow Mr. Pavic's
>example and call it a "lexicon".
>
>From,
>Brendan B. B. (Bren...@aol.com)
>(Name in header has spam-blocker, use the address above instead.)
>
>"Do not follow where the path may lead;
>go, instead, where there is no path, and leave a trail."
> --Author Unknown
>
>
>
>
>

I guess I felt, after playing Photopia, that a game like this could be
designed, except that you could definately tell a different story with the
medium. I will say no more now, but I don't believe this to be impossible.

Josh

Play Deephome, an interactive exorcism and repair job.
Http://www.angelfire.com/nj2/Yesuslave
Come to the Alpha version of the new IF Files, where you can find games similar
to ones you like.
http://www.geocities.com/Yesuslave

Jon Ingold

unread,
Mar 25, 2000, 3:00:00 AM3/25/00
to
> Why does it become impossible?

Logistically - you get far too many branches far too quickly. Assuming you
any (a) people to react to events or (b) timed events to interact with.
Otherwise, sure you can tell as many as you like by letting all breakable
things break etc. But NPC reactions, for example - if they're going to be
the slightest bit believable in a any-outcome scenario you're going to have
to write three hundred and one responses to the same thing.

I'm trying something similar myself at present - however it is going to be
in no way not unwinnable.

Jon

YesuSlave

unread,
Mar 25, 2000, 3:00:00 AM3/25/00
to

There is, of course, the old "reign the player in" routine an IF author must go
through with any game, no matter how large. Though I do see what you are
saying.

BrenBarn

unread,
Mar 27, 2000, 3:00:00 AM3/27/00
to
>This is exactly what I am doing now...I am a bit paranoid, so I don't talk
>about my games before I release them, but I expect to work long hours on this
>particular game, but I don't see that being something that should stop me.
Can't wait to play it.

The Solar Echo

unread,
Mar 27, 2000, 3:00:00 AM3/27/00
to
This has certainly been an thought provoking thread to read, and I want to
thank you all for posting. It's encouraging, really, to find out that so many
of us have had similar thoughts about expanding the boundries of I-F. We may
disagree on the details, but it seems like what most of us want to create is
more realistic interactive fiction, and I-F with more...depth.

One of my favorite quotes from this thread is:

> I feel that I am using story in perhaps too broad a sense. A
>"multiple-story" game (i.e., one in which the player has lots of freedom to
>alter the events of the game) may in fact be telling a single "story". This
>"story" may not be a linear narrative, but instead be an overall picture of a
>person, place, society, time period, etc.

> ~ BrenBarn.

This is the sort of I-F I'd like to see more of.
For me personally, it's about creating I-F that doesn't FEEL like a game. One
of the really unrealistic things about traditional I-F is that it rewards us
for taking everything that's not bolted to the floor. In real life, would you
REALLY pick up that crushed soda can that's lying in the gutter, thinking that
you can use it to solve a problem later? Would you really steal things from
people's houses for no good reason? Probably not. So when I'm playing a game
that punishes me for not doing those things, it's wrecks the illusion for me.
There is no Suspension of Disbelief.

Any other thoughts of this or other related topics?

BrenBarn

unread,
Mar 28, 2000, 3:00:00 AM3/28/00
to
>In real life, would you
>REALLY pick up that crushed soda can that's lying in the gutter, thinking
>that
>you can use it to solve a problem later? Would you really steal things from
>people's houses for no good reason? Probably not. So when I'm playing a
>game
>that punishes me for not doing those things, it's wrecks the illusion for me.
I agree; but I must say that once you start trying to model actual
reactions that way, you can quickly run into the giant problem of using a
computer to imitate human behavior.
The thing about the game "punishing" you for not picking up random
objects, though, really gets me thinking. One thing I would like to see (which
means I'd like to be able to write :-) is IF where the notion of the game being
able to "punish" the player is minimal. It seems like most IF is centered
around the idea that the player is searching for a single "right" action among
all the possibilities. This seems incompatible another common IF trait:
namely, the attempt at increasing the realism of the game world.
If realism is the issue, then "right" and "wrong" actions shouldn't be,
because in the real world, concrete "right" and "wrong" don't exist (in the
sense of "furthering your goal"; I'm not talking morals here :-). So it would
be very cool to have a game where the possibilities of the PC's action are
limited more by the actual characters and objects in the game, and less by the
author's attempt at focusing the player's attention on the story (or whatever
we're calling it :-). Of course, it's impossible to totally model reality; all
I'm saying is that pushing the envelope a little would be very interesting.

Lucian Paul Smith

unread,
Mar 28, 2000, 3:00:00 AM3/28/00
to
BrenBarn (bren...@aol.comRemove) wrote:

: It seems like most IF is centered


: around the idea that the player is searching for a single "right" action among
: all the possibilities. This seems incompatible another common IF trait:
: namely, the attempt at increasing the realism of the game world.

If there is a 'right' action, it means that it's a puzzle. It doesn't
mean that it's less realistic. There's a great article at the games cafe
at:

http://www.thegamescafe.com/text/What%20Is%20a%20Puzzle.txt

One of the most intriguing quotes in it is, "This hierarchy leads me to a
useful rule of thumb for puzzle designers: to design a good puzzle, first
build a good toy. The player should have fun just manipulating the
puzzle, even before reaching a solution."

To me, this says that increasing the realism of an IF game makes it a
better toy, and therefore a better setting for a puzzle. Well,
almost. Increasing the *responsiveness* of an IF game makes it a better
toy, and therefore a better setting for a puzzle. So even if you have
only one situation that *solves* the puzzle, other potential situations
should be noticed and commented on by the game--even if it doesn't advance
the story, per se.

-Lucian

Jon Ingold

unread,
Mar 28, 2000, 3:00:00 AM3/28/00
to
..the other problem with really open, loose, responsive games, with multiple
paths, is they might not be fun to play. It's sound naive perhaps, but a lot
of players might get frustrated at having knowing there are different,
'hidden' alternatives which they've missed and may require some replaying to
reach. (I know I felt this way about Blade Runner, which allegedly had
mulitple story-lines, not that I ever found any. Well, actually I found one,
I think it was buggy because I got irretrievably stuck in a game with now
puzzles.. anyway)

A game like this would be really impressive if it's deeper, not broader
though. If you a wider spectrum for meaningful interaction (above "You jump
on the spot fruitlessly" etc) then the game becomes more interesting; but if
you have too many mutally exclusive paths it might detract from the
accessibility of the entire thing and discourage readers/players.

Then again, it may bait them. Who knows. Like I said, I'm trying to write
something kinda along these lines at the moment. From a programming point of
view it's harsh, because everything overlaps, it's very difficult to keep
track of all the branches which are unfinished (I'm sure I'll release it and
it'll still contain a few print outs of "..and this routine should now be
handling the policeman's reaction to the dead dog outside" and the like).
It's impossible to debug. And I have now I think about 23 different
scenario's (not *very* different, admittedly, but just by combinatronics)
and the game only lasts 15 moves or so. Is it worth it?

Jon

Michael Straight

unread,
Mar 28, 2000, 3:00:00 AM3/28/00
to
On 27 Mar 2000, The Solar Echo wrote:

> One of my favorite quotes from this thread is:
>
> > I feel that I am using story in perhaps too broad a sense. A
> >"multiple-story" game (i.e., one in which the player has lots of freedom to
> >alter the events of the game) may in fact be telling a single "story". This
> >"story" may not be a linear narrative, but instead be an overall picture of a
> >person, place, society, time period, etc.
> > ~ BrenBarn.

This describes one of the strengths of "Muse". It has multiple endings
but you don't fully appreciate the "optimal" ending unless you've also
seen the "suboptimal" endings. I think of them as all one story.

SMTIRCAHIAGEHLT


The Solar Echo

unread,
Mar 28, 2000, 3:00:00 AM3/28/00
to
> So even if you have
>only one situation that *solves* the puzzle, other potential situations
>should be noticed and commented on by the game--even if it doesn't advance
>the story, per se.
>
>-Lucian

I agree with this. When I'm playing a game that has a lot of single solution
puzzles, I'm often frustrated when my thoughtful but wrong attempts to solve it
aren't even noticed. Often that game won't even tell you WHY the puzzle can't
be solved that way, which only serves to remind me that I'm playing a GAME and
trying to find the RIGHT set of actions. My immersion in the game is broken
and there is no suspension of disbelief. A lot of times I'm tempted to cheat
just so I can forget about that puzzle and become immersed in the game again.

Personally, I'd like to see more puzzles with multiple solutions. In real
life, the majority of problems we run into can be solved more than one way.
Give the player some logical options, and it's not a game of "find the right
action" anymore. The flimsy door to the shed can be unlocked with the key
found in the tool box, but typing KICK DOOR several times will also do the job.
Bad example, perhaps, but you get the idea.

The Solar Echo

unread,
Mar 28, 2000, 3:00:00 AM3/28/00
to
>..the other problem with really open, loose, responsive games, with multiple
>paths, is they might not be fun to play. It's sound naive perhaps, but a lot
>of players might get frustrated at having knowing there are different,
>'hidden' alternatives which they've missed and may require some replaying to
>reach.

Hmmm.....I never really thought about it that way. I guess it would depend on
how easy it is to find a different storyline, and how engrossing the
reader/player finds the game in the first place. It also depends on who the
reader/player is. Some reader/players won't care that they've missed out on
some alternate storylines. For those that do care, I want to include a "plot
map" that shows where the plot diverges into different storylines. After the
reader/player has traveled all the paths they can find and runs out of ideas,
the plot map can help them find the ones they missed. Do you think something
like this might help? Also most of the plot branches will be fairly obvious in
my project, so reader/players don't get discouraged early on.

>A game like this would be really impressive if it's deeper, not broader
>though. If you a wider spectrum for meaningful interaction (above "You jump
>on the spot fruitlessly" etc) then the game becomes more interesting;

I couldn't agree more. A game that would really impress me would be a game
that is deeper because of the details.
The "McDonalds in France dialogue" at the beginning of 'Pulp Fiction' is not
essential to the plot, and it doesn't bring the characters any closer to
reaching their goals. But it's one of hundreds of details that set 'Pulp
Fiction' apart from dozens of other crime movies. The details give the
moviegoers a richer experience. The same could be applied to I-F with a little
effort on the part of the programmer.



> Who knows. Like I said, I'm trying to write
>something kinda along these lines at the moment. From a programming point of
>view it's harsh, because everything overlaps, it's very difficult to keep
>track of all the branches which are unfinished (I'm sure I'll release it and
>it'll still contain a few print outs of "..and this routine should now be
>handling the policeman's reaction to the dead dog outside" and the like).
>It's impossible to debug. And I have now I think about 23 different
>scenario's (not *very* different, admittedly, but just by combinatronics)
>and the game only lasts 15 moves or so. Is it worth it?
>
>Jon

Only you can decide if it's worth it to you, but I know that I'm going to be
eagerly waiting for your game to be released. It certainly sounds fascinating.
I encourage experimentation because, after all, without experimentation how
can their be growth?

BrenBarn

unread,
Mar 31, 2000, 3:00:00 AM3/31/00
to
>Increasing the *responsiveness* of an IF game makes it a better
>toy, and therefore a better setting for a puzzle. So even if you have

>only one situation that *solves* the puzzle, other potential situations
>should be noticed and commented on by the game--even if it doesn't advance
>the story, per se.
Right. The distinction you have made between "realism" and
"responsiveness" is very astute and broadens the possible approaches. For
example, in a game set in a fantasy world, there might no "realistic" puzzles,
but they could still be responsive.

BrenBarn

unread,
Mar 31, 2000, 3:00:00 AM3/31/00
to
>From a programming point of
>view it's harsh, because everything overlaps, it's very difficult to keep
>track of all the branches which are unfinished (I'm sure I'll release it and
>it'll still contain a few print outs of "..and this routine should now be
>handling the policeman's reaction to the dead dog outside" and the like).
>It's impossible to debug.
Of course, as you say, it's true that any such game is going to be very
hard to debug. But I think it could be made easier by not attempting to tell a
series of actual stories, but just allowing many apparently (or maybe actually)
unrelated events to form a collage of impressions. Then again, some people
(myself included) might not always be impressed with a game which is a bunch of
parts with no whole. Hmm. . .

Jon Ingold

unread,
Mar 31, 2000, 3:00:00 AM3/31/00
to
> Of course, as you say, it's true that any such game is going to be
very
> hard to debug. But I think it could be made easier by not attempting to
tell a
> series of actual stories, but just allowing many apparently (or maybe
actually)
> unrelated events to form a collage of impressions. Then again, some
people
> (myself included) might not always be impressed with a game which is a
bunch of
> parts with no whole. Hmm. . .

It depends entirely on what you're doing. If it's an art-house piece, there
to build up an impression in a disconnected way, then fair enough. The thing
I'm working on though - if it evers runs through - is very much less of a
story as a balancing act for the player; the idea being that things he does
will affect what happens and so he'll have to be compensating as well as
forwarding the story. As you may know from Mulldoon, I prefer puzzle games
to story games, and this is what this'll be; only losing the "fixed
solution" of the puzzle in place of a heuristic approach. How do you
convince your ex-wife you don't love her anymore? Make sure she's in the
room when you kiss Clarissa, or happens to be behind you when you
impassively drop her photo in the bin. That sort of thing, well, ish. It
would impossible to write a walkthrough really, more just guidelines and
strategies. I want something which responds to the player, that isn't a game
book but is somehow *alive*. It sounds impossible, AI-IF, I know, but - damn
it all, I've written the start few scenes and they *work*.

However, what I'm really interested in doing is having a game which drives
itself - rather than the impassive, well, what now approach of most games
the player will have to keep up with gallavanting NPC's to a strict
time-limit; not in the Varicella "whoops, you're dead style" but that the
stresses and problems will build up and build up.

This is the plan, at any rate. Writing about it is helping to clarify ideas
in my head as well. :0)

Jon

okbl...@my-deja.com

unread,
Apr 3, 2000, 3:00:00 AM4/3/00
to
In article <8c235m$fib$1...@newsg1.svr.pol.co.uk>,
"Jon Ingold" <j...@ingold.fsnet.co.uk> wrote:
> [stuff and then]

>
> solution" of the puzzle in place of a heuristic approach. How do you
> convince your ex-wife you don't love her anymore? Make sure she's in
the
> room when you kiss Clarissa, or happens to be behind you when you
> impassively drop her photo in the bin. That sort of thing, well, ish.
It
> would impossible to write a walkthrough really, more just guidelines
and
> strategies. I want something which responds to the player, that isn't
a game
> book but is somehow *alive*. It sounds impossible, AI-IF, I know, but
- damn
> it all, I've written the start few scenes and they *work*.

It's not impossible at all. I've been toying with something very
similar. The player character in my story is hip-deep in ethical and
criminal problems. The player has the option of resolving the problems
various ways at various ethical levels.

The puzzle is figuring out both what the "best" plan is *and* how to
implement the solution. Maybe you start out immediately with the
intention of resolving everything the most ethical way possible, but
find that's not as easy as it sounds, as what you say has no meaning to
others and your actions, no matter how noble, are interpreted as having
ulterior motives. And maybe one of the people you've screwed is going
to snap and kill you outright, or call the cops, etc.

My only real problem is deciding whether such an unpleasant scenario
will actually be *enjoyable*. I'm not sure if people enjoy black comedy
when they're the protagonist.

Reply all
Reply to author
Forward
0 new messages