IF: puzzle vs narration

4 views
Skip to first unread message

richard develyn

unread,
Jan 20, 2006, 7:08:21 AM1/20/06
to
Following on from discussions on "stuck" situations it seems to me
that interactive fiction is having to tread a difficult path between
"puzzle fairness", "narration" and "simulation". As far as
I can tell, "puzzle fairness" seems to have the strongest
supporters, or maybe the most sacred cows, which surprises me a little
because I thought that in changing the name of the genre from
"adventure games" to "interactive fiction" there was a move
away from puzzles towards simulation (i.e. "interactive") and
narration (i.e. "fiction").

For example, there is very little fiction that I can think of where the
main character knows everything that is going to happen in the future,
or even has a reasonable chance of knowing. Of course with simple
fiction the author is in total control of the narrative path; when you
go interactive, if you wish to provide "puzzle fairness", you either
have to find ways of providing foreknowledge (which can make your
narrative unrealistic) or you have to provide a satisfactory conclusion
to every choice.

It is always possible to do both these things, but I feel there's a
cost, which is either to bring a certain sterility to the narration, or
to change the interactivity into a conflict between author and player,
whereby the author only allows the player to get stuck if he was "too
stupid" to notice whatever clues were left in his path.

It may be better to allow a player occasionally to get stuck purely as
a consequence of the fact that this is a simulated environment and that
within it you can, such as you can with real life, screw things up. To
give someone freedom of choice (I read somewhere) is to give him or her
the "right to be wrong".

Of course IF is not a book, and you cannot simply turn the page and
carry on the story regardless. Neither is it life, and you cannot
continue living it despite having screwed up somewhere. This is why I
believe an IF player needs to be told in some fashion, preferably
within the story, that he can progress no further. Of course the author
has to be aware than in allowing this to happen he is challenging the
player to decide whether he has enjoyed the IF sufficiently to want to
go back and change something and then replay on from there. It's a
decision that shouldn't be taken lightly, and doubtlessly if you
challenge the player more then once or twice then they may well get fed
up and go and do something else. However I do think it's a compromise
that as an author you shouldn't be afraid to make, and I think the
whole IF genre will be the sadder if nobody does.

For example (a classic one way trip type), the Robinson Crusoe story:
in the 17th century you set off on a boat trip to somewhere civilised
but you hit a storm and end up ship-wrecked on an uninhabited desert
island. Your success in escaping depends on whether you've brought
certain things along with you that, quite frankly, you probably
wouldn't think of taking with you normally (such as a knife).

The "narration" elements say the ship-wreck should be something
unexpected (of course). The interactive elements allow you to choose
whether to carry a knife or not, and you hardly want to tell the player
("you better carry that knife with you in case you get shipwrecked"
or "people like you always carry knives around in order to
whatever"). You could contrive to have the knife delivered in some
other fashion but just having things washed-up or lying around breaks
your whole narrative tension (you are on your own - no deus ex
machina).

The only option for the player who didn't bring the knife therefore
is to restart the game and bring it (and then say to himself,
suspending disbelief for a moment, "lucky I brought that knife with
me", which I know might feel odd but it's what your story wants). Of
course you should try to indicate to the player, via the narration
preferably, that he has got himself stuck without the knife, pretty
quickly if you can. I believe this is a better option for this sort of
story than the other two (prevention or substitution).

I have no problem with the various guidelines that people have written
in various "guides to writing good IF" that exist on the net. I
think they're a good indication of the sorts of issues that players
have found frustrating in the past. However I feel that authors
shouldn't be frightened of breaking these guidelines occasionally
when they feel that their narrative or interactive needs require them
to. I believe that if the IF is good enough most players will forgive
you, as long as you're careful how you do this and considerate.

Richard

J. J. Guest

unread,
Jan 20, 2006, 7:42:52 AM1/20/06
to
In the early days authors of adventure games always seemed to assume
that the player would not complete them on the first play-through.
Knowledge gained from previous play-throughs was often crucial to
completing the game. I have even played old games where a death-message
actually contained crucial clues for completing the puzzle it was part
of!

Whilst I would not go as far as the death-message clue, this assumption
makes game construction a whole lot easier for the author. It removes
the need for meticulously eliminating unwinnable states, and if I
thought I could get away with it I would certainly make the same
assumption in my own games. I could probably turn out a game a week if
I did. But to the modern player, it seems, it is simply unacceptable.
There are exceptions to the rule; Varicella is a notable example of a
well-liked game which requires multiple play-throughs to complete
because of its very tight timing. But it is a knotty problem. I myself
do not mind playing a game over and over in order to get it right, like
Bill Murray in Groundhog Day, but I get the impression that most
players these days do not feel that way.

richard develyn

unread,
Jan 20, 2006, 9:11:56 AM1/20/06
to
Obviously you can go far too far with this, and maybe even two or three
small steps is one or two too far. But I think a purist attitude will
stifle creativity and prevent a lot of actually quite enjoyable IF ever
been written.

Richard

Jess Knoch

unread,
Jan 20, 2006, 9:31:22 AM1/20/06
to
richard develyn wrote:

> It may be better to allow a player occasionally to get stuck purely as
> a consequence of the fact that this is a simulated environment and
> that within it you can, such as you can with real life, screw things
> up. To give someone freedom of choice (I read somewhere) is to give
> him or her the "right to be wrong".

I don't think every game should be made so that it can't be put in an
unwinnable state. As long as the author is aware of when this happens, and
consciously makes decisions about gameplay, and (ideally) the player knows
before going in that this sort of thing might happen, then you're good to
go.

I hate to just point out things other people have written, but I believe
there was some discussion on this matter a few months back. For instance,
this post (you might need to copy/paste):
http://groups.google.com/group/rec.arts.int-fiction/msg/f1fe804cee7f3ae2?hl=en&


See also: Zarf's Cruelty Scale, at:
http://www.ifwiki.org/index.php/Cruelty_scale

--
Jess K., who considers a warning that the game is unfair to be perfectly
fair.


richard develyn

unread,
Jan 20, 2006, 10:18:21 AM1/20/06
to
Hi there.

Thanks for that. I agere with everything Daryl says in that post back
in December. I just felt that when I started the discussion about how
to flag up to the player that he is stuck people's first reaction was
to say you're a crap designer/author if you let this happen.

Richard

Jess Knoch

unread,
Jan 20, 2006, 10:37:23 AM1/20/06
to

I think that was because you said there was no way to make it so that there
was no stuck situation. You said:

> I've racked my brains to see if I can think of a logical game-true way
> out of this and I can't.
>
> Since I personally hate this sort of thing when I play games myself I
> want to put some sort of meta-game thing in place to tell the player
> that this has happened to him.

So of course no one suggested that it was perfectly fine to allow an
unwinnable situation -- since that clearly wasn't what you were looking for.

I think there is a time and a place for all varieties of "forgiveness" from
games. Personally, I favor Eric Eve's suggestion, which is to end the game
at that point (and I tend to prefer a humorous death message at this point,
but it depends on the game). I believe this is "polite" on Zarf's scale.

--
Jess K., getting stuck in unstuckable games since 2002.


David Thornley

unread,
Jan 20, 2006, 12:29:45 PM1/20/06
to
In article <1137758901.2...@g44g2000cwa.googlegroups.com>,

richard develyn <ric...@skaro.demon.co.uk> wrote:
>Following on from discussions on "stuck" situations it seems to me
>that interactive fiction is having to tread a difficult path between
>"puzzle fairness", "narration" and "simulation". As far as
>I can tell, "puzzle fairness" seems to have the strongest
>supporters, or maybe the most sacred cows, which surprises me a little
>because I thought that in changing the name of the genre from
>"adventure games" to "interactive fiction" there was a move
>away from puzzles towards simulation (i.e. "interactive") and
>narration (i.e. "fiction").
>
I think it a natural reaction.

In the old puzzle-fests, the emphasis was on solving the puzzles.
When you solved the puzzles, you were essentially done with the game.
The storyline was largely unimportant, being primarily something to
hang puzzles off of.

This means that the game was not meant to immerse the reader in
the same sense that a good novel does. The game was made to be
played with all sorts of metagame verbs, since it didn't need to
immerse, and so learning by dying was perfectly reasonable. It
meant that you had to go back to a saved game. People would
complain about games where you could lose the game and continue for
a long time, and so would have to go back several save files and
go through other puzzles again, but that's not the same thing.

If a game is meant to be immersive, the author wants to avoid
the use of verbs that would jar the player out of immersion.
"SAVE" and "RESTORE" are used only as bookmarks (I rarely finish
a book in one sitting, for example), and not as ways to try
different things out. The player is not expected to have to
play it through multiple times to get the intended experience,
but rather to go through once.

This means that puzzles are less important, since instead of being
the game's main attraction they become minor entertainment or pacing
devices, forcing the player to explore some, say, before moving on.

Since the puzzles are less important, they have to fit in more with
the rest of the game. When puzzles are important and the story is
background, the game needs to accomodate the puzzles and change the
story to conform. When story is important and the puzzles are pacing,
the game needs to accomodate the story and change the puzzles to
conform.


--
David H. Thornley | If you want my opinion, ask.
da...@thornley.net | If you don't, flee.
http://www.thornley.net/~thornley/david/ | O-

Dan Shiovitz

unread,
Jan 20, 2006, 5:48:08 PM1/20/06
to
>Following on from discussions on "stuck" situations it seems to me
>that interactive fiction is having to tread a difficult path between
>"puzzle fairness", "narration" and "simulation". As far as
>I can tell, "puzzle fairness" seems to have the strongest
>supporters, or maybe the most sacred cows, which surprises me a little
>because I thought that in changing the name of the genre from
>"adventure games" to "interactive fiction" there was a move
>away from puzzles towards simulation (i.e. "interactive") and
>narration (i.e. "fiction").

It's interesting to me that you talk about "puzzle fairness", "narration",
and "simulation" as three separate things here, but then the rest of
your post seems to conflate "narration" and "simulation". I think it's
useful to maintain a distinction between the latter two, and that
looking at that distinction addresses some of the points you raise
later. I also think that you're being slightly incorrect in talking
about the third axis being "puzzle *fairness*" -- I think it's more
just "puzzles", and "fairness" is one quality which determines how
good the puzzle is, along with "originality", "aha-ness", etc.

That said, lemme address stuff a little more specifically.
(Rearranging some of your points to make them easier to reply to)

[..]


>For example (a classic one way trip type), the Robinson Crusoe story:
>in the 17th century you set off on a boat trip to somewhere civilised
>but you hit a storm and end up ship-wrecked on an uninhabited desert
>island. Your success in escaping depends on whether you've brought
>certain things along with you that, quite frankly, you probably
>wouldn't think of taking with you normally (such as a knife).

This seems like a direct clash between narration and simulation, for
instance. It's true that if this were the "real world", the chance of
the person carrying useful supplies along when they get shipwrecked
are not that high. Similarly, even if the person gets shipwrecked
while carrying a knife, in most cases they'll drown, and in the cases
when they don't drown, they'll probably come down with pneumonia and
die of that, and if that doesn't happen they'll probably get eaten by
leeches or something. The idea of narration is not that we ignore
real-world plausibility, but that we choose between all the plausible
events to pick the ones that make interesting stories.

So, for example, in this case, the "correct" way to do the story isn't
to give the player a choice about whether to bring the knife or not,
because if they don't, the resulting stories are all uninteresting.
Instead, you start the story off by saying "You've been shipwrecked on
an island. Luckily, you brought your knife with you. What do you do
now?" and *now* we can let the player do whatever they want, because
whatever they do is cool and makes a good story. This is, incidentally,
a big deal in RPG theory too, where it's called "scene framing". If
you search for that term, you'll find a lot of interesting links about
it.


>whatever"). You could contrive to have the knife delivered in some
>other fashion but just having things washed-up or lying around breaks
>your whole narrative tension (you are on your own - no deus ex
>machina).

Note, by the way, that in the actual Robinson Crusoe story, he has
*all kinds* of loot salvaged from the ship -- provisions, tools, wood,
rope, guns. Talk about a deus ex machina! The only shipwreck that's
more implausible is the one in Swiss Family Robinson, where their ship
was fortunately carrying a small windmill.

[..]


>For example, there is very little fiction that I can think of where the
>main character knows everything that is going to happen in the future,
>or even has a reasonable chance of knowing. Of course with simple
>fiction the author is in total control of the narrative path; when you
>go interactive, if you wish to provide "puzzle fairness", you either
>have to find ways of providing foreknowledge (which can make your
>narrative unrealistic) or you have to provide a satisfactory conclusion
>to every choice.

That's an interesting way to put it, since I've always thought that
the principle of puzzle fairness is that the player *doesn't* have to
have foreknowledge to avoid getting stuck -- that is, a reasonable
person can know how to do things at the time they need to. On the
other hand, "unfair" games end up getting restarted a lot, so the
player does need to exercise foreknowledge. I guess it comes down to
whether the "character" (in the sense of a single playthrough by the
player) or the player (in the sense of the accumulated wisdom of all
playthroughs) needs to have the knowledge: I almost always find it
flows better to handle it at the character level, though there are
certainly exceptions (Lock and Key, Rematch, Varicella, All Things
Devours).

>It may be better to allow a player occasionally to get stuck purely as
>a consequence of the fact that this is a simulated environment and that
>within it you can, such as you can with real life, screw things up. To
>give someone freedom of choice (I read somewhere) is to give him or her
>the "right to be wrong".

Yeah, absolutely -- but "wrong" is not the same as "bored" by any means!

>Of course IF is not a book, and you cannot simply turn the page and
>carry on the story regardless. Neither is it life, and you cannot
>continue living it despite having screwed up somewhere. This is why I
>believe an IF player needs to be told in some fashion, preferably
>within the story, that he can progress no further. Of course the author

[..]


>up and go and do something else. However I do think it's a compromise
>that as an author you shouldn't be afraid to make, and I think the
>whole IF genre will be the sadder if nobody does.

Yeah, mm. I mean, what it comes down to is, you have to weigh the
change against the integrity of the work as a whole. It's no good
writing a game that you think is great but everybody will get stuck on,
but it's also no good writing a game that sucks but everyone can
breeze through. In general, if it's a central point of the game I
won't compromise, but in most other cases, even if I like the game
slightly better one way but it's more playable the other, I'll usually
change it. Often I end up changing my mind about which I like better
once I've got used to the other way, anyway.

Also, YMMV, but I find it extremely intrusive to be told I've made
the game unwinnable -- it feels like the author's stepped in and given
me a hint, which I suppose is exactly what it is. On the other hand, I
don't have any problem finding out a decision was final, even after I
made it -- that's just seeing how the game-world works.

[..]


>I have no problem with the various guidelines that people have written
>in various "guides to writing good IF" that exist on the net. I
>think they're a good indication of the sorts of issues that players
>have found frustrating in the past. However I feel that authors
>shouldn't be frightened of breaking these guidelines occasionally
>when they feel that their narrative or interactive needs require them
>to. I believe that if the IF is good enough most players will forgive
>you, as long as you're careful how you do this and considerate.

Oh, heck yeah. I certainly hope nobody holds back on writing something
they think is cool because they're worried about breaking some rule of
IF design. The people I'm worried about are more the people who don't
seem to be thinking about design at all -- they just assume that of
course the player will enjoy navigating this maze, or restarting
because the light source burned out, or the like. Those authors need
to be kicked. But anyone who's thought about the design and still
wants to do it their way, hey, I'm all for it. I may think it's a bad
choice and dislike the game afterwards, but I applaud the sentiment.

>Richard
--
Dan Shiovitz :: d...@cs.wisc.edu :: http://www.drizzle.com/~dans
"He settled down to dictate a letter to the Consolidated Nailfile and
Eyebrow Tweezer Corporation of Scranton, Pa., which would make them
realize that life is stern and earnest and Nailfile and Eyebrow Tweezer
Corporations are not put in this world for pleasure alone." -PGW

Fish

unread,
Jan 21, 2006, 12:13:34 AM1/21/06
to
richard develyn wrote:
> I just felt that when I started the discussion about how
> to flag up to the player that he is stuck people's first reaction was
> to say you're a crap designer/author if you let this happen.

To me, there's a continuum. On the one hand is Life; the other, Game.

In Life there's a set of reasonable expectations for any action. I
haven't fallen off a bridge today, but I have a good idea what would
happen if I tried. It would be highly unreasonable for me to fasten a
cord to my waist and climb over the edge of a bridge *just in case*
there's treasure down there. Also, I haven't seen a cop all day today,
but if I started monkeying around on the edge of a bridge, I know one
would show up.

A good game design *does* allow you to get stuck through your own
foolishness, if you persist in doing unreasonable things, but the player
depends entirely on the author's craftsmanship to know what "reasonable"
is. A fantasy world with magic and/or goblins (or whatever), with an
unknown location, requires much more work by the author. The player's
notions of what he might successfully do, and what the consequences of
his actions might be, diminish dramatically unless the author prepares
his material well.

Just my two cents.

FISH

Samwyse

unread,
Jan 21, 2006, 7:42:37 AM1/21/06
to
Dan Shiovitz wrote:

> Note, by the way, that in the actual Robinson Crusoe story, he has
> *all kinds* of loot salvaged from the ship -- provisions, tools, wood,
> rope, guns. Talk about a deus ex machina! The only shipwreck that's
> more implausible is the one in Swiss Family Robinson, where their ship
> was fortunately carrying a small windmill.

Personally, I've always wanted to write an IF based on "The Sailor Dog"
(http://images.amazon.com/images/P/0307001431.01._SCLZZZZZZZ_.jpg) which
is (IMHO) somewhat more plausible than either of those shipwrecks.

Well, more plausible once you get past the whole dog thing.

And I guess that finding a pallets-worth of bricks *was* a bit unlikely.

Mike Roberts

unread,
Jan 21, 2006, 4:28:13 PM1/21/06
to
"richard develyn" <ric...@skaro.demon.co.uk> wrote:
> For example, there is very little fiction that I can think of
> where the main character knows everything that is going to
> happen in the future, or even has a reasonable chance of
> knowing. Of course with simple fiction the author is in total
> control of the narrative path; when you go interactive, if you
> wish to provide "puzzle fairness", you either have to find ways
> of providing foreknowledge (which can make your narrative
> unrealistic) or you have to provide a satisfactory conclusion
> to every choice.

The contrast you're drawing about foreknowledge is the exact opposite of the
way I see it. In my view, the IF ideal is for the characters to be able to
operate with exactly the same amount of foreknowledge as characters would
have in static fiction - that is, none. (Unless it's a story about a
character who has premonitions, of course - but even then you'd have the
same foreknowledge in the static and interactive renditions.)

Here's the thing: in static fiction, the characters have to get it right
*the first time*. If they get killed, or they accidentally lose an artifact
that they absolutely need to reach their goals, they don't get to UNDO or
RESTORE "PAGE 267" and try again. (Unless it's a story about someone who
can go back in time or switch among alternate timelines or parallel
universes or whatever. And if it is, an interactive rendition could
implement the equivalent of undo or restore in the game world itself -
*without* meta-game commands like SAVE and RESTORE and UNDO - since the
timeline switches are part of the character's experience.)

To be realistic, static fiction characters have to make choices and take
actions based on what they plausibly know within the story. An action hero
can't just know that there's a bad guy hiding around the next corner; they
have to catch a glimpse of the bad guy ahead of time. A detective can't
just randomly know that the suspect will be coming out of a restaurant at
8pm; she has to have found information about the reservation somewhere.
Characters in static fiction have to have some plausible motivation for the
actions they take. If they do something important, and there's no good
reason for it, the reader is annoyed by the author's obvious manipulation of
events.

For what you're calling "puzzle fairness" in IF, the point is exactly the
same. The point is to allow IF characters to behave realistically, without
the *player's* obvious intrusion into the story. Suppose the player plays
through once and discovers that you get killed by a baddie hiding around a
certain corner. So the player restores the game from a point just before
the corner and tries again, but this time they know to throw the grenade
around the corner first, taking out the bad guy. If you saw this in static
fiction, you'd think it was ridiculous - how did the hero know to throw the
grenade around that exact corner? It's obvious author intrusion. In the IF
version, how did the PC know to throw the grenade just then? It's obvious
*player* intrusion.

> It is always possible to do both these things, but I feel
> there's a cost, which is either to bring a certain sterility to
> the narration, or to change the interactivity into a conflict
> between author and player, whereby the author only allows
> the player to get stuck if he was "too stupid" to notice
> whatever clues were left in his path.

Well, you have to differentiate between cases where the player *thinks*
she's stuck, and those where she's truly stuck. If the player simply isn't
picking up on clues, she just *thinks* she's stuck - that is, she can
proceed as soon as she figures out the right thing to do next. Being truly
stuck means that the player has irreversibly altered the game state to
foreclose success - she's lost or destroyed a crucial artifact, or locked
herself out of a key location, etc.

My big reason for wanting to avoid the "truly stuck" case isn't so much out
of "fairness," as you're calling it; it's more for the reasons above, that
truly-stuck situations make for more game-ness and less fiction-ness,
because a truly-stuck player will have to invoke meta-game mechanisms (UNDO
or RESTORE) to get unstuck.

> It may be better to allow a player occasionally to get
> stuck purely as a consequence of the fact that this is a
> simulated environment and that within it you can, such as
> you can with real life, screw things up. To give someone
> freedom of choice (I read somewhere) is to give him or her
> the "right to be wrong".

That seems like a reasonable argument on its face, but it has a huge flaw.
The flaw is that it's not more realistic to give the player the freedom to
screw things up when the screw-up itself isn't realistic.

When this "agency" issue comes up people always say they should have the
freedom to jump off bridges. That's fair enough, but it's also beside the
point. The kinds of screw-ups that matter aren't the simple, realistic
suicides. Rather, they're the ones that create unrealistic dead ends: the
ones where you forget to bring the knife on your doomed ocean crossing, or
you lock your keys in your car, or you eat the loaf of bread that you'll
need later to bribe the troll. These kinds of screw-ups mostly just can't
happen in the real world. They happen in IF only because IF is so
unrealistic: it's unrealistic in that certain objects are unique when they'd
be commonplace and numerous in the real world; they happen because objects
aren't as breakable or malleable or deformable as they are in the real
world, and don't exhibit the even a tiny fraction of the range of behaviors
they would in the real world; they happen because NPCs in IF are
simple-minded automata and not remotely like real people In the real world,
if you forget to bring your knife, you can fashion a cutting instrument out
of wood or stone or broken glass if you're determined enough. In the real
world, if you lock your keys in your car, you can call AAA, or jimmy the
lock, or just break a window. In the real world, if you eat the loaf of
bread, you can buy another when you encounter the troll, or maybe even talk
him into accepting one of the many shiny trinkets in your backpack.

When IF limits agency for the sake of avoiding dead ends, it's unrealistic.
But it's just to compensate for all the other bits of unrealism, so it
doesn't itself do anything to reduce the overall level of realism - it just
evens out the unrealism a bit, and in a way that improves the playing
experience for many players. A better solution would be to expand the world
model to make it more realistic; when you can do that, you can open up the
player's choices and no one will complain. But being ultra-realistic on one
axis while being ultra-fake on all the other axes is counterproductive.

--Mike
mjr underscore at hotmail dot com


richard develyn

unread,
Jan 24, 2006, 6:06:02 AM1/24/06
to
Excuse me answering myself - I just find it the easiest way to pick up
and collate on all the separate threads.

Scene framing only works at the beginning of a game. The general
problem I was trying to illustrate (and I picked a bad example) is a
one way trip occuring somewhere in the middle of a game (actually
travel is just the easiest example of a non-reversible state change -
which is where the problem happens).

As soon as we make the one way trip, and there is lots of play before
and after the trip, you have the problem I've been trying to
illustrate.

I'll try another example:

You're in a spaceship, which is boarded. Part 1 of the game is finding
your way out and, I don't know, maybe sabotaging the spaceship in the
process. You escape in a shuttle and land down on planet X for Part 2.
That's the one way trip. Now either:

1) part 2 is a completely self-contained adventure with no (critical)
reliance on anything that you did in part 1, or

2) part 2 does rely on part 1 but:

2a) all possibilities are catered for.

2b) you have some sort of "premonition" about part 2 which means you
are given reasonable clues about what to do in part 1 in order to be
able to complete part 2

2c) part 1 is impossible to complete without having done everything
"right" for part 2, even though you may not realise exactly why it is
right

2d) you can get stuck in part 2 and have to restore / restart in order
to do part 1 right.

In my opinion it has to be a judgement call, balancing the needs of the
story you are trying to tell with the needs of the player you are
trying to tell it to.

A strong simulationist might argue (2a) is better, because it
approaches realism, and because most of the time when you get stuck you
are unrealistically stuck.

A strong advocate of puzzle fairness would, I suppose, support
everything bar (2d).

The story-based designer would argue whichever one best fits the
narration. (2d) is sometimes necessary when you want to have "luck" in
the game. Most normal fiction includes strong elements of luck for the
main character. The argument is that, well, you could tell the story of
all the characters in a particular scenario who didn't get lucky (by
being in the right place at the right time when something interesting
was going to happen, if nothing else) but that would be boring. Your
main characters in fiction always are lucky - and you need to translate
that somehow into IF. Sometimes you'll be able to "guarantee" that luck
using (1) to (2c), sometimes it'll just be too awkward to do so so
you'll have to drop down to (2d).

Richard

solar penguin

unread,
Jan 24, 2006, 6:33:57 AM1/24/06
to
richard develyn <ric...@skaro.demon.co.uk> wrote:

>
> I'll try another example:
>
> You're in a spaceship, which is boarded. Part 1 of the game is finding
> your way out and, I don't know, maybe sabotaging the spaceship in the
> process. You escape in a shuttle and land down on planet X for Part 2.
> That's the one way trip. Now either:
>
> 1) part 2 is a completely self-contained adventure with no (critical)
> reliance on anything that you did in part 1, or
>
> 2) part 2 does rely on part 1 but:
>
> 2a) all possibilities are catered for.
>
> 2b) you have some sort of "premonition" about part 2 which means you
> are given reasonable clues about what to do in part 1 in order to be
> able to complete part 2
>
> 2c) part 1 is impossible to complete without having done everything
> "right" for part 2, even though you may not realise exactly why it is
> right
>
> 2d) you can get stuck in part 2 and have to restore / restart in order
> to do part 1 right.
>
> In my opinion it has to be a judgement call, balancing the needs of
> the story you are trying to tell with the needs of the player you are
> trying to tell it to.
>

At the risk of being contraversial... It _doesn't matter_ which one you
choose, just as long as you let the players know at the start of the
game. If they know from the start what sort sort of game they're
playing, they can choose whether it's the sort of thing they personally
want to play.


--
___ _ ___ _
/ __| ___ | | __ _ _ _ | _ \ ___ _ _ __ _ _ _ (_) _ _
\__ \/ _ \| |/ _` || '_| | _// -_)| ' \ / _` || || || || ' \
|___/\___/|_|\__,_||_| |_| \___||_||_|\__, | \_,_||_||_||_|
|___/
http://www.freewebs.com/solar_penguin/

** The Deuteronomist recast older materials to create clones of a
fearful, often violent, vision that reveals truths about past, present
and future times in highly symbolic and poetical terms. The poet may
represent himself as transported into a meteor crater!


Neil Cerutti

unread,
Jan 24, 2006, 8:11:51 AM1/24/06
to

Even strong advocates of puzzle fairness sometimes opt for 2d,
e.g., _Curses_, one of the most evil games of all time, developed
by the author of _The Craft of Adventure_. But then, he seems to
have been a fan of the Topoligica games, so evil is in his
nature.

I believe players are disposed to forgive authors of faults like
this, if care seems to have been taken in other ways. Perhaps
they aren't as generous when playing a 2-hour comp game, though.

> The story-based designer would argue whichever one best fits
> the narration. (2d) is sometimes necessary when you want to
> have "luck" in the game. Most normal fiction includes strong
> elements of luck for the main character. The argument is that,
> well, you could tell the story of all the characters in a
> particular scenario who didn't get lucky (by being in the right
> place at the right time when something interesting was going to
> happen, if nothing else) but that would be boring. Your main
> characters in fiction always are lucky - and you need to
> translate that somehow into IF. Sometimes you'll be able to
> "guarantee" that luck using (1) to (2c), sometimes it'll just
> be too awkward to do so so you'll have to drop down to (2d).

In the space ship game example, the author can strewn missing
items from part 1 around the part 2 area, ad hoc. A player
playing for the first time will not even notice.

One of Jon Ingold's comp games has no mandatory puzzle, but right
now I've forgotten its name.

--
Neil Cerutti

Ken

unread,
Jan 24, 2006, 1:55:46 PM1/24/06
to
>The only option for the player who didn't bring the knife therefore
>is to restart the game and bring it (and then say to himself,
>suspending disbelief for a moment, "lucky I brought that knife with
>me", which I know might feel odd but it's what your story wants).

Who say's a knife is even necessary? I mean Paleolithic man survived
just fine with nothing but sticks and stones and animals bones. For
your survival story you can do some research to find out the techniques
used by even some modern day tribes who still do not use metal or
modern conveniences of any kind. They make their own tools, clothing
and fire using only what nature provides. I think that would be an
interesting survival adventure. To figure out how to start a fire by
picking out the right supplies laying around on the island and picking
the right spot for it and using them in the right way. To construct
hunting, digging, cutting, chopping, sawing and sewing impliments out
of sticks and stones. To construct shelter out of branches and mud and
rocks and such. To spear fish or trap small game in makeshift traps you
have to construct yourself. To kill larger game for food and animals
hides which you have to sew together to make clothing. YOu could even
construct a native style canoo either duggout from a log or hide
stretched over a tree-branch frame. The possibilities for a realistic
survival game are really endless.

--Ken

Jim Aikin

unread,
Jan 24, 2006, 10:28:54 PM1/24/06
to
> Most normal fiction includes strong elements of luck for the
> main character.

As a writer of conventional fiction, I'm not convinced of this. I'd like to
see a couple of examples of what you're talking about. And something more
modern than Dickens. Dickens is FULL of unlikely coincidences. (There was a
reason for that. He was writing serials, and he had to keep tossing fresh
plot twists at his readers in every episode.)

The general precept for writing plotted fiction is that the lead character
should get out of her predicament **through her own efforts.** The luckier
she is, the more the reader feels cheated.

I've heard it said that a realistic novel or short story should have EXACTLY
ONE significant coincidence (i.e., bit of luck, good or bad). If there are
no coincidences, the reader will perceive that the writer believes the
universe to be a mechanical and therefore soulless place, like a giant piece
of clockwork. But if there are two or more coincidences, the reader will
feel that the writer is cheating.

--JA

*******************************************
"Those instances of it which lack the quality referred to
as 'swing' are meaningless." --Duke Ellington
*******************************************


Samwyse

unread,
Jan 24, 2006, 10:53:52 PM1/24/06
to
richard develyn wrote:

> The general
> problem I was trying to illustrate (and I picked a bad example) is a
> one way trip occuring somewhere in the middle of a game (actually
> travel is just the easiest example of a non-reversible state change -
> which is where the problem happens).
>
> As soon as we make the one way trip, and there is lots of play before
> and after the trip, you have the problem I've been trying to
> illustrate.

Those are called "scenes". Within a scene, the player has complete
freedom of movement, but once a scene is done, the player can't go back.

> You're in a spaceship, which is boarded. Part 1 of the game is finding
> your way out and, I don't know, maybe sabotaging the spaceship in the
> process. You escape in a shuttle and land down on planet X for Part 2.
> That's the one way trip. Now either:
>
> 1) part 2 is a completely self-contained adventure with no (critical)
> reliance on anything that you did in part 1, or
>
> 2) part 2 does rely on part 1 but:
>
> 2a) all possibilities are catered for.
>
> 2b) you have some sort of "premonition" about part 2 which means you
> are given reasonable clues about what to do in part 1 in order to be
> able to complete part 2
>
> 2c) part 1 is impossible to complete without having done everything
> "right" for part 2, even though you may not realise exactly why it is
> right
>
> 2d) you can get stuck in part 2 and have to restore / restart in order
> to do part 1 right.

Emily Short has done a fair amount of work on scene managers, which IIRC
handle case 2c; there's a daemon that runs each turn and checks if all
of the prerequisites for the scene change have been completed.
Actually, from a game mechanics point of view, 2b and 2c seem to be the
same except for what the author tells the player. For example, at a
formal dinner party the scene manager may wait until the player picks up
a butter knife, then introduce an escaped ax murderer to the game. The
difference between 2b and 2c would hinge on whether another NPC says
"Did you hear there's a criminal on the loose?" versus "Have you tasted
the herbed butter on the French loaves?"

Fish

unread,
Jan 24, 2006, 11:08:01 PM1/24/06
to
I'm a strong advocate of puzzle fairness *and* story, myself, and I
don't like 2b or 2c. I did once write a game that involved the player
being on a vessel which foundered and sank, and he had to have a life
vest (or scuba gear, or some damn thing) before the ship would ever
crash -- but that solution doesn't seem very organic to me now.

2a is the best of both worlds, to me: if the player finishes every task
on the spaceship, you reward his diligence and foresight. If he lands
on the planet and can improvise alternate solutions, you reward his
ingenuity.

For instance, on the spaceship you may find a plastic ID card, a broken
ray gun (with a perfectly good battery pack) and a helmet. You can use
the ID card to jimmy a door, the battery to start a fire, and wear the
helmet so when you climb the tree the monkey doesn't conk you out with
coconuts.

However, if you don't have those items, there is an alternate (and
longer!) way to accomplish the same things -- like trying to dig under
the door with a large clam shell. (If the player remembered to bring
along the ID card, you don't even need to spawn the clam shell item,
since he wouldn't need it.)

I don't like 2d very much. It forces the player to repeat the same
content ad infinitum (which is seldom as exciting the tenth or eleventh
time). It also doesn't reward ingenuity or foresight, but failure: the
previous times the player failed.

FISH

ems...@mindspring.com

unread,
Jan 25, 2006, 2:06:22 AM1/25/06
to

Samwyse wrote:
> richard develyn wrote:
[A scheme of relating scenes]

> > 1) part 2 is a completely self-contained adventure with no (critical)
> > reliance on anything that you did in part 1, or
> >
> > 2) part 2 does rely on part 1 but:
> >
> > 2a) all possibilities are catered for.
> >
> > 2b) you have some sort of "premonition" about part 2 which means you
> > are given reasonable clues about what to do in part 1 in order to be
> > able to complete part 2
> >
> > 2c) part 1 is impossible to complete without having done everything
> > "right" for part 2, even though you may not realise exactly why it is
> > right
> >
> > 2d) you can get stuck in part 2 and have to restore / restart in order
> > to do part 1 right.
>
> Emily Short has done a fair amount of work on scene managers, which IIRC
> handle case 2c; there's a daemon that runs each turn and checks if all
> of the prerequisites for the scene change have been completed.

I have, but not with the intention of using them in this kind of
situation; I think it's bad manners to make the player thrash around
trying to fill arbitrary requirements that are never made clear to him.
So I don't use the scene manager to wait until the player happens to be
holding the hammer and wrench but not holding the left glove, and then
pounce on him.

What I tend to do with scene managers is

-- watch for the completion criteria of a complicated puzzle -- which
might require that the player be holding the wrench and hammer and not
the rubber glove, but the player will have been given some rationale
for why this is a desirable state of affairs; in this case, the scene
manager is just there so that I *don't* have to append separate checks
to TAKE HAMMER, TAKE WRENCH, *and* DROP GLOVE.

-- watch for some other kind of narrative conclusion: e.g., the NPC has
asked the player a question which he might answer one of several ways,
but we are waiting for him to make a decision before the plot goes on.
In this case there isn't a puzzle as such, but the scene does have a
goal (force the player to take one position or the other) and it is
pretty clear what is going on (because the NPC asked).

Which doesn't address the original problem at all. Off the top of my
head, ways I might deal with this (depending on context):

-- design the game so that there is a legitimate puzzle-logic reason
why the player cannot go forward unless he has exactly the right set of
tools. Con: sometimes a very difficult design problem.

-- make the player character know, even if the player doesn't, that
he's going to need the hammer and wrench. ("You start for the door,
then bring yourself up short. What would Hank say if you showed up on
the job site without your tools?") Con: the PC has to have a legitimate
reason for this foresight, and too much of it will get annoying.

-- don't make the player play out the acquisition of every useful
object. "Gourmet" does a great job of eliminating arbitrary and boring
actions from the player's experience, for instance: the player
character needs to make a soup, but the game narrates many
dull-yet-hard-to-simulate steps and gives control back to the player
only when the stage is properly set and there is again something
interesting to do. If you've got some stuff that the player will need
in Part 2, but there's no pressing reason why it has to come *from*
Part 1, you may be able to slip in some narration about how he comes by
it during the scene change, instead. Con: doesn't work if the player is
going directly from Part 1 to Part 2 with no time leaps or setting
changes.

-- intervene overtly. "On the way to the job site, you remember that
you left behind the hammer, so you rush back and get it. (Time wasted
retrieving the hammer: 30 seconds. Time wasted explaining to your
neighbor why you were in such a rush as to knock over his trash can and
kill his cat: 27 minutes.)" Con: heavy-handed; works best in IF where
the author is already intervening a lot. Otherwise, it will just look
like an obvious patch job.

-- intervene subtly. If the player leaves the hammer behind, have the
game supply a different hammer-like tool that he can find in the new
location. Con: requires lots of extra objects and clever logic; may
undermine puzzles earlier in the game if there is no punishment for not
acquiring needed materials.

My personal favorite, though, is

-- avoid the problem.

In general, when I am writing a game with a lot of distinct segments, I
try to stick with a minimal inventory for exactly this reason. If the
player does have possessions that he absolutely must keep from segment
to segment, I try to set it up so that he never drops them. "City of
Secrets" does this; the player character has a narrative reason to be
wary of losing his stuff, so not letting him drop valuable things felt
like a fair solution.

Daphne Brinkerhoff

unread,
Jan 25, 2006, 3:56:46 AM1/25/06
to

Fish wrote:
> richard develyn wrote:
> > I just felt that when I started the discussion about how
> > to flag up to the player that he is stuck people's first reaction was
> > to say you're a crap designer/author if you let this happen.
>
> To me, there's a continuum. On the one hand is Life; the other, Game.
>
> In Life there's a set of reasonable expectations for any action. I
> haven't fallen off a bridge today, but I have a good idea what would
> happen if I tried. It would be highly unreasonable for me to fasten a
> cord to my waist and climb over the edge of a bridge *just in case*
> there's treasure down there. Also, I haven't seen a cop all day today,
> but if I started monkeying around on the edge of a bridge, I know one
> would show up.

Just had to say that your post reminds me greatly of one of my favorite
IF posts ever.

http://groups.google.com/group/rec.games.int-fiction/msg/73305f5b9d3d2fcd

You may return to your regularly scheduled analytical argument now. ;)

--
Daphne

richard develyn

unread,
Jan 25, 2006, 5:18:59 AM1/25/06
to
I think there's two sorts of "luck" involved here:

First of all there's co-incidence, or the luck of being in the right
place at the right time in order to make an interesting adventure.

Terry Pratchett I think captured the concept quite nicely with his book
"Interesting Times". All Rincewind wants is to have a quiet life. From
his point of view it is terribly *bad* luck that thrusts him into one
perilous situation after another. From the point of view of the reader
it is *good* luck because otherwise the book would be very boring to
read.

Then there's the conventional sort of "luck" about getting yourself out
of trouble.

Now I know what you mean about that writer's formula (did you see it in
MZB's pages, BTW?) But I think there is an element of smoke and mirrors
involved. Sure the reader should feel that the main character has
sorted their almost insurmountable problem by themeselves, but I'm sure
if you applied some cold mathematical risk analysis to it you would
find that, just as with the people who succeed in real life, luck
played one hell of a part.

The Hobbit / Lord of the Rings contains plenty of examples of the
hobbits getting rescued in the nick of time, but maybe that's not a
good example.

The Day of the Triffids starts off with an incredible piece of luck.

Most fantasy sagas, I think. I mean I knew when I started reading
Raymond Feist's Daughter of the Empire that that little girl scrubbing
the monastery floor was destined to become the heroine of three books.
Sure, she did a lot for herself, but she had a lot of help too, and
there were plenty of occasions when luck played a part in helping her
out.

Richard

Jim Aikin

unread,
Jan 25, 2006, 1:30:27 PM1/25/06
to
> First of all there's co-incidence, or the luck of being in the right
> place at the right time in order to make an interesting adventure.
>
> Terry Pratchett I think captured the concept quite nicely with his book
> "Interesting Times".

Yeah, well, Terry Pratchett gets to make up his own rules. That's a big part
of the fun of his books!

> Sure the reader should feel that the main character has
> sorted their almost insurmountable problem by themeselves, but I'm sure
> if you applied some cold mathematical risk analysis to it you would
> find that, just as with the people who succeed in real life, luck
> played one hell of a part.

Luck can present useful objects (the nuclear reactor at the end of Lucifer's
Hammer comes to mind), but it's essential that the characters figure out
themselves how to take advantage of the luck, and that it not be too easy!
In this, conventional fiction is a lot like IF.

> The Hobbit / Lord of the Rings contains plenty of examples of the
> hobbits getting rescued in the nick of time, but maybe that's not a
> good example.

Lord of the Rings -- don't get me started. A great work of literature, it
will be read with pleasure 500 years from now, but it's not without flaws. I
made a list once, just for fun. Here are two quick examples:

1) With two trivial exceptions, there are no wild animals whatever in the
book. The landscape is empty. This is especially obvious in the scene where
Legolas and Aragorn are tracking the bunch of orcs that has kidnapped Merry
and Pippin. The good guys are entering an ancient forest where humans have
not trod for hundreds of years; among their party are an amazing tracker and
an amazing archer; yet they voice worry that they'll run out of food,
because they didn't bring enough supplies with them. One can only conclude
that there were no animals in the forest. To Tolkien, it seems, a "forest"
consisted exclusively of trees. It was not a living bio-region.

2) Orcs are portrayed as, without exception, unredeemably evil. After the
battle in which the Rohan guys defeat the enemy at the mountain citadel
(sorry -- don't remember the name of the place, and I'm too lazy to look it
up), the defeated humans are given a chance to switch to the winning side,
but the orcs are simply slaughtered. Nobody thinks to ask the orcs whether
they'd like to switch to the winning side. Now, in the real world one would
have to assume that orcs love lady orcs and little baby orcs, if nothing
else. They would be capable of a variety of emotions, including tender ones,
so they would have some moral complexity. But because Tolkien has set up an
extremely simplistic moral equation (humans = capable of good or evil; elves
= always good; orcs = always evil), the slaughter of the orcs leads to no
soul-searching.

Depicting one's enemies as worthy only of wholesale slaughter is regrettably
common -- well-nigh universal -- among the human race, but that doesn't make
it defensible. On the contrary: it's to be deplored.

Them's mah 2 cents, and I'm stickin' to 'em.

--JA


Samwyse

unread,
Jan 26, 2006, 12:50:45 AM1/26/06
to
richard develyn wrote:

> The Day of the Triffids starts off with an incredible piece of luck.

Actually, I interpret it as an application of the Anthropomorphic
Principle. Of all the people in the world, 99.999% had very
uninteresting stories, so for the book to tell an interesting story it
has to be about someone at the far end of the bell curve.

richard develyn

unread,
Jan 26, 2006, 5:23:26 AM1/26/06
to
w.r.t. LOTR:

Interesting what you say about the wildilfe. It never occurred to me,
but then I expect I was thinking (as maybe was J.R.R.T.) of the forests
we have in england these days which are more or less devoid of wildlife
bigger than a rabbit or a squirrel.

The point about orcs brigns up an interesting point that I seem to
remember was made by Kurt Vonnegut in his book Hocus Pocus about the
difficulty of going to war with people you haven't demonised. It helps
if they have a culture different than yours, and it helps if you don't
speak their language.

Richard

richard develyn

unread,
Jan 26, 2006, 6:27:17 AM1/26/06
to
Well, exactly. The problem then is to translate that into IF where you
still want to give your player some free will.

Richard

Neil Cerutti

unread,
Jan 26, 2006, 8:16:57 AM1/26/06
to

I think of it like a phrase of western music. At the beginning,
after stating on the tonic chord (the basis, or key the phrase is
in), I'm allowed, in the first chord change, to make a jump to
*anywhere I want to*. But after that, I need to more-or-less
follow the circle of fifths to get back to the tonic (the
resolution). Stories seem to get away with a similar trick.
Something very unexpected can happen in the beginning, but things
must follow more-or-less a logical progression from then on.

--
Neil Cerutti

Ken

unread,
Jan 31, 2006, 11:46:32 PM1/31/06
to
>Interesting what you say about the wildilfe. It never occurred to me,
>but then I expect I was thinking (as maybe was J.R.R.T.) of the forests
>we have in england these days which are more or less devoid of wildlife
>bigger than a rabbit or a squirrel.

Some of you who enjoy a good Western now and then as I do might also
have noticed a similar lack of large game in cinematic portrayals of
the frontier lands or old west of N. America. There is actually good
reason for this.

While most of the dry portion of the earth's surface was, for millions
of years, naturally dominated by either desert or forest the
proportionately smaller remainder of land area was occupied by
grasslands of some kind or another. The savannas of Africa, the pampas
of S. America, the steppes of Central Asia and the prairies of N.
America were once the only major grass dominated ecosystems (before
lawn care services began sprouting up all over the place).

Grass is unique among plants in that it grows from its base rather the
tip and so can survive grazing by grass eating mammals. As a matter of
fact, grass flourishes due to the constant source of fertilizer
provided by the grazers. By contrast most plants in forests have
developed some sort of nasty or unpleasant defense against grazers due
to the fact that their family jewels are held at the tip of their
stalks rather than the base.

So for this reason wherever grass grows you will find herds of big fat
lazy grazing animals who frequently defecate where they eat - and often
simultaneously. Which is why the Serengeti teems with wildebeest,
zebra, and gazelles. The prairies were once where the great herds of
Buffolo roamed. And quite naturally, where there is big game there will
be big carnivores to hunt them - wolves, wild dogs, lions, cheetahs,
hyenas, and such.

So it is not really unusual that in the rain forest or the spruce
forests of the north or the oak woods of the temperate latitudes, large
animals are few and far between; there is less for them to eat.

--Ken

richard develyn

unread,
Feb 1, 2006, 5:03:55 AM2/1/06
to
Thanks for that - that was really interesting.

Richard

Reply all
Reply to author
Forward
0 new messages