Attitudes to playing (longish)

31 views
Skip to first unread message

Jamieson Norrish

unread,
Oct 4, 1994, 7:52:17 AM10/4/94
to
Warning: This gets into me saying that I like simulation, and that an
imposed plot need not curb wandering tendencies in the
player/character.

I've noticed that most people who post here have a great dislike of
finding out much later in the game that something went wrong and the
character can't go back. Now, this seems quite reasonable, because it
means it's a waste of time - you have to restart the entire game, or
at least go back to a saved game from ages ago.

But hold on. When I was playing the Zork trilogy, I rarely used save
(except, I admit, in Zork II, nearing the end of the game - all those
objects, sheesh) and would start a new game each time I played. I
didn't ever sit down determined to play until I finished it. I just
started, played, and when I left to do something else, I stopped
(which, I must say, was generally at a point when I was stuck).

Am I the only one who plays/played like this? Going back over all the
stuff at the beginning, redoing puzzles, rereading room descriptions,
getting vital objects stolen by the *&^%ing thief, etc. :) My point is
just that I can't imagine doing it in any other way, and it seems that
other people can't imagine doing it my way either. Why the difference?

Well, I could certainly suggest some - I can have a very high tedious
tolerance when I need to, and I really enjoyed getting immersed in
game. Trying to see how everything fitted together - not just in the
puzzles (and in many ways they were the worst bits of the game,
because of their slightly contrived nature), but in all the rooms and
little objects (the newspaper in Zork II was interesting because of
what it said, not because of its use in the door puzzle). What I
wanted was a world to explore, with puzzles deriving from the world
set-up.

Now, Dave mentioned, I think, that you need an imposed storyline on
the game. Perhaps, but ignoring the plotless simulation, I think that
there are different ways of doing this. Many games go for linear
plot-lines, where wandering around "just exploring" is ruled out. I
think that it should be possible to have the plot not so entirely
dependent on the character. Have stories (preferably more than one)
built into other characters, locations, and objects. Then let the
players do what they want, and discover their own plot.

I know that we've been here before, and people have said that players
can't or don't do that - if given the opportunity to wander, they
will, and not get any enjoyment from the game. Well, that doesn't cut
it. If the player wants to follow up on a plot thread that has been
come upon, then the player can. Or not. But that's the player's
decision, and why not? It just makes games more interesting to more
people - some can play it without deviation (provided they know what
they're getting into, so don't get overwhelmed by so-called "red
herrings"), while others can follow their own path.

Anyone who's read this far care to comment or object? :)

Jamie

Gerry Kevin Wilson

unread,
Oct 3, 1994, 10:24:23 PM10/3/94
to
Well, I guess I'll reiterate my old viewpoint. First and foremost, I'm a
writer, not a world engineer. I tell a story. I leave room to roam
around, play with things, and talk philosophy with NPCs. These are the
incidental parts to me, they may be the main part to you. I have to say
that puzzles are used as conflict in the story simply because I've got a
ways to go before I can figure out how to work in non-puzzle conflict.
In stories you read, the characters are faced with the same sort of
puzzles as the player in a text adventure. This is not a bad thing.
This is not a good thing. This is just a thing. I will continue to
write story and puzzle oriented games until someone finds a better way to
do things that everyone enjoys. While you can still get enjoyment from
exploring a primarily puzzlee-slanted game, others cannot get any
enjoyment, or only brief enjoyment, from exploring a virtual simulation
in text. Many people like to be faced with challenges and forge their
way through them. Not as many people take great delight in wandering
around a MUSH or similar setup where there are no puzzles, and no one
else is logged on. I take this into account when I write my games.

Now to focus more specifically on what you just brought up, room to explore.
I have really seldomly played a text adventure that gave me no room to
explore. Now, if you think about that, you will realize that a text
adventure with no room to explore is merely a hypertext novel. On the
other hand, what you suggest, the game that plots its own story, is:

1.) Out of my league. The exponential growth problem will get you every
time.

2.) Not satisfying to me, as an author. I have to have something
physical and concrete at the end of 300-400 hours work. (More, usually.)
If I have a plotted game, I can look at it and say, "I did that.
Everything that happens is something I wanted to happen. This story is
a result of my writing skills." I really don't know how I would react
to having created a 'system' of objects such as you describe. Actions
proscribed according to generic inter-relations rather than my own
conscious decision to add them. Frankly, I don't think I would feel
that I'd actually done anything. When you let a computer write the
story, are you still the author?

3.) Economic viability. A simulation is not something I can market. No way,
no how. Look at Arena. It was marketed as a huge world with all these
quests and such, but it flopped. It was a miserable failure because the
authors got too ambitious and spread themselves too thinly. What they
ended up with was an engine, not a game. Ultima 8 did this too. There
is plenty of room below the poverty line for IF authors. We need to
start looking at ways, not to commercially revive text adventures, but
to salvage what we can. There will be no text adventure revival. No
phoenix rising from the flames. Much less a surge of simulations
flooding the market. We each have to hold on to whatever is dearest to
us and carry it out of the burning building while we still can. No one
else is going to, that's apparent. Commercial games have moved in the
exact opposite direction of text adventures. Don't believe me? Load up
Doom and play it for 15 minutes. Story? What story? Explore, sure.
Manipulate objects, no. Interact with other characters? Only to kill
them. Hopefully this is more a sign of the field's infancy than the
final evolution of video games. Well, anyways, what started as an
explanation of why I prefer plotted to non-plotted games has degenerated
into a sermon. I'll shut up now.
--
<~V~E~SOF~~~~~~~~~~~~~~~CYBER~CHESS~~~~~~~~~~~~~NO~RELEASE~DATE~~~~~~|~~~~~~~>
< RTI T In the distant future, entire planets are won or lost | ~~\ >
< G O WAR E in a single battle. Vertigo's first strategy game. | /~\ | >
<_____DONT-HOLD-YOUR-BRE...@uclink.berkeley.edu__|_\__/__>

Andrew C. Plotkin

unread,
Oct 4, 1994, 12:15:05 PM10/4/94
to
Excerpts from netnews.rec.arts.int-fiction: 4-Oct-94 Attitudes to
playing (longish) Jamieson Norrish@akeake. (2961)

> Now, Dave mentioned, I think, that you need an imposed storyline on
> the game. Perhaps, but ignoring the plotless simulation, I think that
> there are different ways of doing this. Many games go for linear
> plot-lines, where wandering around "just exploring" is ruled out. I
> think that it should be possible to have the plot not so entirely
> dependent on the character. Have stories (preferably more than one)
> built into other characters, locations, and objects. Then let the
> players do what they want, and discover their own plot.

I would love a game like that, but I'm not capable of writing it. I have
enough trouble coming up with a single plot and making it playable.

Sigh.

--Z

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

David Baggett

unread,
Oct 6, 1994, 5:29:59 PM10/6/94
to
In article <JAMIE.94O...@akeake.its.vuw.ac.nz>,
Jamieson Norrish <ja...@akeake.its.vuw.ac.nz> wrote:

>But hold on. When I was playing the Zork trilogy, I rarely used save ...and


>would start a new game each time I played.
>

>Am I the only one who plays/played like this?

I know that if I made a game that didn't allow saving the game, players
would have my head on a platter. I don't know anyone who plays like this;
it seems to me that most people tend to save/restore or use undo liberally.

>Many games go for linear plot-lines, where wandering around "just
>exploring" is ruled out.

As an exercise, I tried coming up with a game concept that involved five
puzzles that could be solved in any order. My only requirement was that
there be a satisfying plot associated with each of the 32 possible solution
orders. I found this nearly impossible! So it is very difficult to
achieve this so-called "nonlinearity" in practice if you disallow "bad"
plots. (If you don't think it's hard, do it. I'd love to see such a
work.)

Furthermore, I think that most players *think* they want to be able to
solve the puzzles in any order, but only because they don't know what this
implies. It implies that you have no idea at any given time what puzzle
you should work on --- that you get no clues from the previous puzzle
you've solved or from your surroundings about which puzzle to solve next.
This is true when you first get underground in UU2, and is one of the
things many people have complained about --- that this circumstance makes
the game too hard. The basic fact is that giving the player more freedom
increases the size of the search space at any given time, which makes the
game correspondingly more different (moment to moment, at least).

>Have stories (preferably more than one) built into other characters,
>locations, and objects. Then let the players do what they want, and
>discover their own plot.

This sounds good at first, but the more I think about it, the more I'm
unsure what it really even means.

By "plot" I mean something you could draw a plot diagram for, like so:

CLIMAX
|
v
-----------
------------ \
----------- \
----------- \
-------- \

<-------------------- RISING ACTION-----------------><- DENOUEMENT ->


For the plot to be "good", it must (generally speaking) have these elements
-- it's got to build up to something; when that something happens, it has
to be compelling; and afterwards you need at least a brief "wrap-up". (Of
course, there are many effective plots that have instantaneous denouements,
but let's ignore that for the moment.)

Don't take this as criticism, but it is far from obvious given only a
comment like "build stories into the characters" how to apply the
traditional coneception of plot to interactive works. I think figuring out
how to do this is *the most important* currently tractable open problem in
interactive fiction.

I can think of a few basic ways, all of which have obvious problems:

1) Local interactivity. Imagine checkpoints on the plot diagram. The work
enforces the constraint that the reader never reaches a checkpoint until he
has reached all the previous checkpoints. Therefore the global integrity
of the plot is maintained while the reader is free to explore at will
between checkpoints. (Most text adventures do this by making puzzles be
the checkpoints.)

2) Plot tree. Instead of a single plot diagram, we have many. One way to
think of this is as a plot tree --- the reader begins at the root, and
encounters forks in the path. There are many different climaxes, and every
leaf (node with no children) in the tree is an "ending". For this to be
satisfying, the author must guarantee that every path from root to leaf is
a good plot (i.e., is of the form given above).

3) Plot graph. Same as a plot tree, except that plot paths may meet after
the beginning. You might imagine a work that has a single climax and
denouement, but has many different plots --- i.e., many different ways to
get to the single climax. E.g., no matter how you get there, you always
end up at the castle, where you have to kill the Bad Guy and save the
planet.

4) Absolute interactivity. Here, the player and program work together to
ensure that the player's path through the game generates a "good" plot
diagram. In other words, the work adapts to what the player's doing and
adjusts events so that "the plot so far" is good. This is AI-complete, and
therefore will not be feasible in the near future.

The problems:

1) Not satisfying to people who want interactive fiction to be *really*
interactive. Somehow seems to avoid interactivity more than it exploits
it.

2 & 3) Author must design an enormous number of plots.

4) AI-complete; probably not solvable within our lifetimes. (Even if you
believe that human-level intelligent machines can be constructed in the
near future, you have to realize that the human language and vision
faculties are suitably complicated and ad hoc as to prevent exact
duplication for a long time.)

Since I've found (and can argue at length!) that 2, 3, and 4 are not doable
in any reasonable amount of time, I advocate 1 while recognizing that it is
far from ideal.

Dave Baggett
__
d...@ai.mit.edu MIT AI Lab He who has the highest Kibo # when he dies wins.
ADVENTIONS: We make Kuul text adventures! Email for a catalog of releases.

David Baggett

unread,
Oct 6, 1994, 5:32:34 PM10/6/94
to
In article <371q8n...@life.ai.mit.edu>, David Baggett <d...@ai.mit.edu>
wrote:

>The basic fact is that giving the player more freedom increases the size of
>the search space at any given time, which makes the game correspondingly
>more different (moment to moment, at least).

^^^^^^^^^

Oops; I meant "difficult". Sorry.

Felix Lee

unread,
Oct 6, 1994, 11:40:42 PM10/6/94
to
David Baggett:

>Furthermore, I think that most players *think* they want to be able to
>solve the puzzles in any order, but only because they don't know what this
>implies. It implies that you have no idea at any given time what puzzle
>you should work on --- that you get no clues from the previous puzzle
>you've solved or from your surroundings about which puzzle to solve next.

umm, if you really can solve the puzzles in any order, then it doesn't
matter which puzzle you work on next.

The game becomes hard when:
- you can see puzzles that are currently unsolvable, and
- you can't tell if a puzzle is unsolvable or not, or
- you don't notice a solvable puzzle because it's camouflaged.

I've been wondering about plot-modelling in puzzle-adventures.
Built-in hints, like the game wizard in UU2 or the demon/angel in
Curses, these are pretty limited. If the game had an internal model
of its own plot, then it could guarantee a useful hint. It could
answer the "what do I do now?" question by pointing you at a
particular puzzle you can solve next.

Jamieson Norrish:


>>Have stories (preferably more than one) built into other characters,
>>locations, and objects. Then let the players do what they want, and
>>discover their own plot.

David Baggett:


>This sounds good at first, but the more I think about it, the more I'm
>unsure what it really even means.

I think you have to throw away the traditional conception of plot.
Simulationist IF will probably be quite different from literary IF.

There's no particular reason a player's role in a simulation has to
have a "good" plot in a literary sense. It just has to have the
potential of having a good plot. With luck and careful design, the
player will be able to find his own story in it. Sometimes baseball
games or nethack games or RPG sessions have dramatic stories to them;
other times they just kill time.

[Start random dreaming.] Traditional fantasy setting. Player plays a
character in a castle in a small kingdom over the course of years.
The player starts as a stable-boy or a page or whatever and advances
in status by accomplishing tasks, by accumulating wealth, by knocking
on opportunities, whatever. Give the player a whole character-sheet
of skills and statistics.

The kingdom is modelled as a simulation. There may be a fixed
timeline for some events: a dragon menaces the countryside, the
princess is kidnapped, etc. Other events may depend on the kingdom's
current state: diplomatic relations break down and a war begins, etc.

Whatever the cause, each event has a scenario or three attached to it.
The player can enter a scenario if he meets the requirements for a
particular role, otherwise he just hears about it.

Modelling these scenarios is plenty hard enough, but you don't have to
worry about branching plot complexity.

The goal for the player is to become a knight, or marry the princess,
or assassinate a wizard, or whatever else is provided for.

There may be a coherent overarching story involving the kingdom, like
the return of the king or the breaking of the world. But the player
does not have to be at the center of this, not all the time.

Well, this is something I'd like to see, though I don't think anything
like this will materialize soon. It's a few steps beyond the
RPG-style computer games I've seen. Not just random monster
encounters; random plot encounters as well.

It's certainly not something I'm going to produce; I'm still stuck at
trying to design puzzle-adventures. (If I find the right key, then
I'll be able to move on to writing story-adventures. :)
--

Matthew Amster

unread,
Oct 7, 1994, 11:05:17 AM10/7/94
to
On 6 Oct 1994 21:29:59 GMT,
David Baggett <d...@case.ai.mit.edu> wrote:

>As an exercise, I tried coming up with a game concept that involved five
>puzzles that could be solved in any order. My only requirement was that
>there be a satisfying plot associated with each of the 32 possible solution
>orders. I found this nearly impossible! So it is very difficult to
>achieve this so-called "nonlinearity" in practice if you disallow "bad"
>plots. (If you don't think it's hard, do it. I'd love to see such a
>work.)

How did you get 32 possible solutions from five puzzles? Last I checked,
5! = 120. I agree wholeheartedly with the sentiment, however.

Matthew

David Baggett

unread,
Oct 7, 1994, 11:06:45 PM10/7/94
to
In article <43177....@pomona.claremont.edu>,
Matthew Amster <mam...@pomona.claremont.edu> wrote:

>How did you get 32 possible solutions from five puzzles? Last I checked,
>5! = 120.

Good point; I deliberately left this unstated because I knew it would only
make an already potentially confusing post more difficult.

Since having a different plot for every possible solve order is clearly
silly (120 is a bit too many for even the most dedicated author) -- I
relaxed my requirements a bit, and decided to only guarantee that every
combination of solved and unsolved puzzles would give you a unique "current
game state". This worked well for my scenario, which was admittedly a bit
unusual. You were to be a time traveller who could instantly appear in any
of three different time periods. Your goal was to get The Artifact by
solving The Puzzles, which all involved doing something in the past that
altered The Artifact's history (or future, depending on which time were
in).

At the beginning of the game, you'd have a book describing The Artifact's
history, and this story would update as you solved puzzles. The key point
is that in all histories but the one where you've solved all the puzzles,
The Artifact is somehow lost or destroyed.

Each time you'd solve a puzzle, the history would adjust itself to explain
why The Artifact was still lost, until you'd solved the final puzzle, at
which point you could go and retrieve The Artifact. This is really quite
simple -- you'd just use the five solved/unsolved bits for the puzzles to
pick history 1 through 32. But this *would* allow you to solve the puzzles
in any order, and the would make sense for each such order. And if you
didn't know better, you'd think, once you'd completed the game, that you'd
solved the puzzles in the only proper order. (Of course, now I've ruined
that for you, haven't I?) :)

It's admittedly a bit different than having every solve order give you a
different plot, but it's certainly no *harder*, and I still found it
amazingly difficult to work out. (I eventually gave up, actually.)

Jamieson Norrish

unread,
Oct 8, 1994, 8:35:58 AM10/8/94
to
In article <FLEE.94O...@simula.cse.psu.edu> fl...@cse.psu.edu
(Felix Lee) writes:

Jamieson Norrish:
>>Have stories (preferably more than one) built into other
>>characters, locations, and objects. Then let the players do what
>>they want, and discover their own plot.

David Baggett:
>This sounds good at first, but the more I think about it, the more
>I'm unsure what it really even means.

["Random dreaming" deleted, but wonderful example.]

That, basically, sums up what I meant. That the plot isn't imposed on
the player to follow, but there is a plot or plots inherent in the
setting, in the NPCs, in actions, and the players is left to find them
and make of them what they will.

Jamie

Felix Lee

unread,
Oct 8, 1994, 2:51:27 PM10/8/94
to
David Baggett:
>[...] I relaxed my requirements a bit, and decided to only guarantee

>that every combination of solved and unsolved puzzles would give you a
>unique "current game state". This worked well for my scenario, which
>was admittedly a bit unusual. You were to be a time traveller who
>could instantly appear in any of three different time periods. Your
>goal was to get The Artifact by solving The Puzzles, which all
>involved doing something in the past that altered The Artifact's
>history (or future, depending on which time were in).

Interesting structure, but I can see why it's hard to work with.

Here's an attempt. The Artifact is the Wycott House, designed by the
architect John Dowlin long before he became famous, and now considered
to be one of his best works.

This is history "LCRAF", when everything goes right:

The Wycott House was constructed in 1952 by Delaney and Sons on a
pleasant hill overlooking Wycott College. The location was Louisa
Wycott's second choice; the house was originally intended to be
situated in a wooded area to the west, but Louisa changed her mind
for reasons unknown. Gossip at the time placed the blame on
various implausible plots involving Louisa's first husband,
Malcolm Russell.

In 1965 Louisa's sole heir, Andrew Wycott, abruptly left the
country, just a few steps ahead of angry creditors. The
belongings were auctioned off but the house itself was unsellable.
It remained abandoned until 1978, when the Birch Historical
Society, with a generous donation from Martha Dowlin, raised
enough funds to restore the house and turn the south wing into a
museum devoted to architect John Dowlin.

This is history "lcraf", when everything goes wrong:

The Wycott House was built outside of Belleville in 1953 by the
Colbert Construction Company. A continuous series of minor
problems culminated in winter of 1966, when the roof of the
southern wing collapsed. Before repairs could begin, a fire
finished the job of demolishing the wing. Arson was suspected,
but Andrew Wycott discouraged investigation into the matter. He
gamely remained in the half-ruined house until the Millstone River
flooded the entire valley in spring of 1967. An attempt was made
in 1978 to recover the derelict, but the project foundered from
lack of funds. In 1982, Malcolm Russell Jr razed the property to
build a shopping center.

There are five binary decision points:
Location - on a hill or in a valley
Contractor - good or bad construction
Rivalry - Malcolm Russell Jr will be involved or not
Arson - Andrew Wycott fled early or not
Funding - Martha Dowlin interested or not

I haven't written it all out, but I think there's a coherent history
involved with all 32 possibilities, and the house can end up destroyed
in 31 of them. Some of the differences are trivial, but it was hard
to make major changes to the house's history.

The problem isn't the house exactly; any major change tends to affect
future decision points, making it hard to keep a particular decision
point stable. Maybe it would be easier if there were greater
separation between time of intervention and time of consequence.
Like, five decision points in ancient Mesopotamia that affect today's
stock market. Silly example, but the larger the separation, the
longer the causal chain, the harder it is to make sensible puzzles.

I'm not really sure what sort of puzzles could be attached to the
decision points I chose for Wycott House. It was hard enough trying
to get it to work at all. Fun, though.
--

Paul Munn

unread,
Oct 9, 1994, 12:26:55 AM10/9/94
to
Felix Lee (fl...@cse.psu.edu) wrote:
[...]
: I've been wondering about plot-modelling in puzzle-adventures.

: Built-in hints, like the game wizard in UU2 or the demon/angel in
: Curses, these are pretty limited. If the game had an internal model
: of its own plot, then it could guarantee a useful hint. It could
: answer the "what do I do now?" question by pointing you at a
: particular puzzle you can solve next.

For my Senior Project as an undergraduate, I wrote an add-on module for
TADS that allows plot modeling much the same way as the adaptive hint
module on ftp.gmd.de does: by using plot diagrams in the form of
Directed Acyclic Graphs or DAGs.

I haven't uploaded the package yet because whenever I look at it, there
are always little improvements I can make. It's not the best code
around, and it could be written more compactly (but then it wouldn't be
as clear as it is now). If and when I do upload it to gmd, my paper will
accompany it so you can know exactly what it's all about.

The system is run by a daemon that keeps track of a "frontier" of nodes.
Each node has a number of child nodes, and the path to each child node
can only be traveled if the user-defined function returns a TRUE. A node
can only be traveled to if ALL of its predecessor nodes are satisfied.

The inspiration for this design, and plot modelling in general, came from
a paper from Carnegie Mellon's OZ Project researchers.

The plot diagramming method I used (I later found out) is being used by
others in TADS games already. This format makes adding hints that change
with the game state not too hard to do. I think Mr. Baggett is using a
plot-dag like system.

I can say that just in writing the simple nonsense adventure that I used
to test it, the DAG plot modelling method forces you to be more
systematic in your plot layout. You have to know what your game is going
to do and when before you code it. In this way, perhaps plot modelling
is a bigger boon to game development than it seems.

Yours,
Paul Munn
pm...@westnet.com
--
--------------------------
"Just because you're paranoid doesn't mean they're not out to get you."
pm...@westnet.com // Paul Munn
modem 914-967-7802 = low cost Lower Westchester, NY net access = Westnet.Com

David Baggett

unread,
Oct 9, 1994, 2:46:49 PM10/9/94
to
In article <377ref$u...@westnet.westnet.com>,
Paul Munn <pm...@westnet.westnet.com> wrote:

>I haven't uploaded the package yet because whenever I look at it, there
>are always little improvements I can make. It's not the best code
>around, and it could be written more compactly (but then it wouldn't be
>as clear as it is now).

Phooey! Upload it anyway! :)

No code is as good as it could be. Anything complicated is going to come
out messy the first time you write it...

>The inspiration for this design, and plot modelling in general, came from
>a paper from Carnegie Mellon's OZ Project researchers.

What does the plot modelling module you descibe actually *do* for you? I
(re-?)invented the plot-DAG model just for the task Felix descibed -- an
adaptive hint system. It sounds like what you've done is more complex.
Could you describe it in a bit more detail, or upload your paper to
ftp.gmd.de?

>The plot diagramming method I used (I later found out) is being used by
>others in TADS games already.

It's an obvious enough idea, but I think that _Legend_ is the first game to
implement it. Does anyone know of any other games that do this? (I know
about David Allen's code on ftp.gmd.de; that was derived from my initial
description of how the hint system in _Legend_ works, and I don't think
anyone's yet used it in a new game. I was sort of hoping to "debut" this
techique -- for hints, at least -- in _Legend_, actually...)

As an aside, the hint system caused most of _Legend_'s well-known delays.
The way I wrote it originally was simple and elegant, and used TADS objects
for the nodes in the DAG (like David's code does). Unfortunately, this
gave me a game with zillions of objects, and the DOS run-time was simply
unable to handle it. With the advent of TADS 2.2 this problem may go away,
because 2.2 under DOS makes better use of memory in 286+ machines.

In any case, the sort of interesting thing is that I lost the original code
to a drive crash long ago, and am left with only the hideous (and I do mean
HIDEOUS) work-around code that is based on having humongous, unreadable
lists of gunk.

Fortunately, David's code (adhint.zip) is much closer to my original code
than the code that I'm actually using now. Anyone interested in this idea
(using a DAG to represent puzzles or plots) should grab Allen's
if-archive/programming/tads/examples/adhint.zip file from ftp.gmd.de and
read adhint.doc. It clearly explains the theory and practice.

David Baggett

unread,
Oct 9, 1994, 2:57:55 PM10/9/94
to
In article <FLEE.94O...@simula.cse.psu.edu>,
Felix Lee <fl...@cse.psu.edu> wrote:

>I haven't written it all out, but I think there's a coherent history
>involved with all 32 possibilities, and the house can end up destroyed
>in 31 of them.

Nicely done! I was locked into thinking of said artifact as a jeweled
scabbard or some such thing.

>Some of the differences are trivial, but it was hard to make major changes
>to the house's history.

The problem you're describing seems similar (perhaps identical) to one I
had with the whole idea --- that it's hard to make a game where you can do
a few things that affect the future so dramatically, but that none of the
other things you can do have dramatic effects.

For example, you can go back in time and convince the builder to build the
house in another place by doing something-or-other. This changes the
future, as we'd expect. However, what happens if you go back in time and
shoot the guy? Shouldn't that affect the future? It's seems harder to
weasel out of these things in a game where you're *supposed* to do things
that affect the future than it is in a traditional game, where your actions
won't generally have such drastic effects.

More to the point, in the time travel game, a *trivial* action may have an
incredible impact on the future. One puzzle idea I thought of was that
you'd plant a tree in a particular place to divert the course of a river
away from one town and towards another. Hence some boat accident wouldn't
occur (ship crashing into dock), and the boat wouldn't sink, and the
artifact wouldn't be lost. (Or perhaps the accident crucially depended on
something about the "bad" town that was absent from the "good" town.)

But what happens if you plant the tree somewhere else? Gee, it seems like
a lot of stuff might happen --- that we've magnified the job of making "all
reasonable things work" a thousand-fold, and that we've rendered ourselves
unable to get out of it just by saying "Now why would you want to waste
your time doing *that*?" everywhere.

Jamieson Norrish

unread,
Oct 10, 1994, 8:29:27 AM10/10/94
to
In article <371q8n...@life.ai.mit.edu> d...@case.ai.mit.edu (David
Baggett) writes:

Furthermore, I think that most players *think* they want to be able
to solve the puzzles in any order, but only because they don't know
what this implies. It implies that you have no idea at any given
time what puzzle you should work on --- that you get no clues from
the previous puzzle you've solved or from your surroundings about
which puzzle to solve next.

No. Just because you are able to solve some puzzles in any order (it
is easy to add "checkpoints" by having external circumstances change
depending on what the character has done), does not mean that the
player will do them in any order, nor that there are no hints, if you
like, as to what can be done next. For example, a character might
wander around, find some "puzzles", ignore them, find another, do it,
be informed about another "puzzle" which continues from the one just
completed, but instead choose to go back and do one of the ones
ignored earlier. Plots are present, they may interact to some
extent (solving one of the earlier puzzles might have changed the
outcome of the later puzzle), and the player can choose to follow
one 'line of enquiry' or go off on another.

And the beginning of UU2 that you mentioned sounds slightly
interesting - I like the idea of wandering around, coming across a few
odds and ends, and then finding something which intrigues me enough to
follow it up.

I write:

>Have stories (preferably more than one) built into other
>characters, locations, and objects. Then let the players do what
>they want, and discover their own plot.

This sounds good at first, but the more I think about it, the more
I'm unsure what it really even means.

Okay, what I mean is that there are stories in the game. Like, this
NPC wants to buy a horse and travel to see her uncle in a far-off
town, while that old shack is haunted by the ghost of the dog who used
to live in it. These are, if you like, "puzzles" (although they can
involve numerous "puzzles" themselves), but are not necessary for
"solving" the game. They have an existence independent of the
character, and some stories may well 'sort themselves out' without the
character ever finding out about them. Or the character could ignore
them.

Does that make it clearer what I mean? The character can wander
around, doing something here, something there, and getting their own
story built out of what is done.

For the plot to be "good", it must (generally speaking) have these
elements -- it's got to build up to something; when that something
happens, it has to be compelling; and afterwards you need at least
a brief "wrap-up".

This is quite possible both for the little stories in a game, and also
the overall story that is created by what the character does and
experiences.

2) Plot tree. Instead of a single plot diagram, we have many. One
way to think of this is as a plot tree --- the reader begins at the
root, and encounters forks in the path. There are many different
climaxes, and every leaf (node with no children) in the tree is an
"ending". For this to be satisfying, the author must guarantee that
every path from root to leaf is a good plot (i.e., is of the form
given above).

I think that it should be possible to do something along these lines
without having to sit down and generate an exlicit plot for each path.
Nor do I think doing this would automatically mean that some outcomes
might be unsatisfying. Also, since there might be many plot trees in a
given game (rather like your option 3, perhaps), a leaf may be an
ending, but not *the* ending.

Say, for example, there was a situation rather like in Zork II, where
there was a object (the brick) which can be used basically anywhere,
but is particularly designed for one puzzle (getting the card and
crown). Now, if we change it so that the puzzle does not have only
this one solution, or that the completion of the puzzle is not vital
to the enjoyment of the game, the object can be used elsewhere without
ruining the game. This opens up the possibility of using it in other
contexts to useful purpose - I was convinced that I could solve the
dragon puzzle by exploding the brick behind the dragon so that it
couldn't retrace its steps, and that I could therefore sneak around
and rescue the princess that way. Alas, it can't be done, but it would
have been easy enough to add that in.

That's a very simply example of course, but I hope it serves to
illustrate both that stories might spring from events (I imagine
someone might be pissed off at a lunatic blowing up their gazebo, for
example), and also that plots don't need to be totally designed
beforehand.

4) Absolute interactivity.

Well, of course this is the goal. :)

Jamie

Jason Noble

unread,
Oct 9, 1994, 7:55:39 PM10/9/94
to

>As an exercise, I tried coming up with a game concept that involved five

^^^^


>puzzles that could be solved in any order. My only requirement was that

^^^^^^^


>there be a satisfying plot associated with each of the 32 possible solution

^^^^^^^^^^^^^^^^^^^^


>orders. I found this nearly impossible! So it is very difficult to

^^^^^^


>achieve this so-called "nonlinearity" in practice if you disallow "bad"

>plots.


I hate to be a pedantic bastard, Dave, but shouldn't that be 5! = 120
possible solution orders? (Which only makes the problem more intractable, I
know).

Having said that, I will confess that I *want* to disagree with Dave. I
want to write and play wonderful interactive fiction with boundless plot
possibilities. However, I fear he may be right. When you get down to
actually coding your ideas, it's pretty hard to make something that's 100%
interactive and 100% interesting at the same time.

Yours pessimistically,


---------------------------------------------------------------------------
Jason Noble | jno...@bunyip.bhs.mq.edu.au
National Centre for HIV Social Research | jno...@laurel.ocs.mq.edu.au
Macquarie University, Sydney, Australia | ph. (61 2) 850 8667
---------------------------------------------------------------------------

Matthew Amster

unread,
Oct 9, 1994, 9:02:12 PM10/9/94
to jno...@laurel.ocs.mq.edu.au
On 9 Oct 1994 23:55:39 GMT,
Jason Noble <jno...@laurel.ocs.mq.edu.au> wrote:

>I hate to be a pedantic bastard, Dave, but shouldn't that be 5! = 120
>possible solution orders? (Which only makes the problem more intractable, I
>know).

I brought this up as well, and David explained on this group what I should
have realized immediately: that the order of puzzles only mattered in
terms of *which* puzzles have already been done. That is, it's not P(5,5),
but a five-bit field, or 2^5, which is 32.

Matthew

Felix Lee

unread,
Oct 9, 1994, 10:34:20 PM10/9/94
to
David Baggett:

>Nicely done! I was locked into thinking of said artifact as a jeweled
>scabbard or some such thing.

yah, I spent a little while on that track too, but I couldn't see how
a future puzzle would remain the same if the object kept moving
around. Though now that I've looked at why the Wycott House scenario
works, I think I might be able to make those work too.

>The problem you're describing seems similar (perhaps identical) to one I
>had with the whole idea --- that it's hard to make a game where you can do
>a few things that affect the future so dramatically, but that none of the
>other things you can do have dramatic effects.

I'm not sure this is really a problem. If you're intervening at a
particular time-travel point, it's because you've traced a causal
chain and identified a particular critical change you can make. Why
does it matter if it's trivial or not?

If you make other trivial changes, then you might affect the history
of other objects and other people, but that's not relevant to this
story, and it can be waved away in the same way that you can wave away
questions about how all this cheez got there, and what happens to it
after you finish the game.

For major changes: if they're not in the scope of the planned puzzles,
then they don't have to result in one of the 32 histories. There can
be a few dead-end offshoots: the house was never built, it was torn
down soon after, etc. Not much different from other types of
dead-ends in games.

It only gets tricky if you're allowed to intersect your own timeline,
and that's easy enough to disallow.
--

Felix Lee

unread,
Oct 9, 1994, 10:42:29 PM10/9/94
to
Dave Baggett:

>One puzzle idea I thought of was that
>you'd plant a tree in a particular place to divert the course of a river
>away from one town and towards another. Hence some boat accident wouldn't
>occur (ship crashing into dock), and the boat wouldn't sink, and the
>artifact wouldn't be lost. (Or perhaps the accident crucially depended on
>something about the "bad" town that was absent from the "good" town.)

>But what happens if you plant the tree somewhere else?

Stuff like this can be handled if you design it so that there are just
two critical states: In the bad history, the tree was planted at X-
and caused event P-. In the good history, the tree gets planted at X+
and causes event P+. If you plant the tree anywhere else, the result
is a neutral event P0 which doesn't accomplish your goal.

It's a three-way decision, not just a two-way one. But it shouldn't
be much harder to create.
--

S.P.Harvey

unread,
Oct 9, 1994, 10:41:04 PM10/9/94
to
Jamieson Norrish (ja...@akeake.its.vuw.ac.nz) wrote:

: Does that make it clearer what I mean? The character can wander


: around, doing something here, something there, and getting their own
: story built out of what is done.

True, and an excellent point made and well-taken. However, we cannot (as
designers) allow so much openness in our games that the player is able to
form their own conclusions as to what is going on. The players must be
kept on-course, as unobtrusively as possible, in order to fully "play the
plot", if you would.

To me, part of the enjoyment of playing IF are the games that have the
following question and answer set as an integral part of the design:

Q: What am I supposed to do in this game?
A: You're supposed to discover what you're supposed to do.

This seems a good method of immersing the player in the world you're
creating. The player will find bits and pieces of the plot scattered
about in different places - solving puzzles, talking to NPC's, examining
objects - and slowly but surely, all the bits will fall into place and
the jigsaw puzzle finally shows a complete picture.

At least that's what I'm striving for in the game I'm working on. I'm
thrusting the player into a mildly interesting situation which will
rapidly unravel itself, leaving the player to weave it back together. I
already have the plot worked out, now I'm placing puzzles to slowly
release that information to the player. Player solves an easy puzzle,
they get a tiny tidbit of the plot. Solve a more challenging puzzle, get
a whole additional chapter.

Am I still making sense?

Scott

--
----------------------| S.P. Harvey |--------------------------
"True! - nervous - very, very dreadfully nervous I had been and am;
but why _will_ you say that I am mad?" - Poe, 'The Tell-Tale Heart'
----------------------| sha...@interaccess.com |--------------------------

Mark B Sachs

unread,
Oct 10, 1994, 2:14:16 AM10/10/94
to
In article <37a9k0$i...@nntp.interaccess.com> sha...@interaccess.com writes:
>To me, part of the enjoyment of playing IF are the games that have the
>following question and answer set as an integral part of the design:
>
>Q: What am I supposed to do in this game?
>A: You're supposed to discover what you're supposed to do.
>
>This seems a good method of immersing the player in the world you're
>creating. The player will find bits and pieces of the plot scattered
>about in different places - solving puzzles, talking to NPC's, examining
>objects - and slowly but surely, all the bits will fall into place and
>the jigsaw puzzle finally shows a complete picture.

In theory, I highly approve of this design. However...

Here's the reason why I haven't been able to get into Curses, despite its
excellence of design. I play the game and find myself wandering an
attic, looking for a tourist map. But as I wander along, all these
strange and bizarre things intrude -- interesting things, certainly,
but I have no idea what they have to do with any larger context, and
once I'm done with them I'm back in the attic no wiser than when I
started. I had no clue what I was really supposed to do, and so just
eventually gave up.

Hmm, how to put this more concretely?

Okay: _I have a problem with a game where I'm faced with puzzles to solve
and no particular reason in the game why one should bother solving
them._ For example, in Lurking Horror, you are a student trying to
complete a term paper. Sure, all kinds of weird and freaky things
are happening to you, C'thulhoid monsters are everywhere, but -- so what?
Why is it _my_ problem? Why should I risk my life over it? One puzzle involves
bashing open a brick wall using an elevator and a rope. I'm sorry, but if in
real life I'm wandering through a campus building and see a brick
wall, my first thought is not "Ooh, how am I going to smash that
wall open so I can see what's behind it?" I never finished Lurking
Horror either, needless to say.

The reason to go around and solve puzzles doesn't have to be an
elaborate or literary one; just a self-evident one, a reason that,
if you were _really_ in this situation, you would personally find
convincing. For example, in Planetfall, you're stranded in a complex
on an unknown planet. Naturally, in _this_ context, finding out how
the place works and then getting it running again would be foremost
on your mind. In Enchanter, you're supposed to break into a castle
and defeat a bad guy, so again, solving puzzles is natural. In
Unnkulia II, you're a professional adventurer whose easy life on the
talk-show circuit is going to dry up if you don't get off your duff
and have another adventure. Given the motivation, I perservered
at _these_ games until I completed them.

So IMHO: that's what a game needs -- _motive_. Not necessarily
a big or a fancy one, just enough of one to get the player off
his or her duff and get moving.

(Am I the only person who thinks this way -- who needs direction to
this degree?)

-Mark

russell wallace

unread,
Oct 10, 1994, 1:54:33 PM10/10/94
to

Agreement.

One of the problems that I have with a lot of adventure games is the
character they require you to play. In particular, very many adventure
games seem to start off with 'You are Mr. Nobody who knows nothing and
has no skills applicable to the weird environment in which you are going
to find yourself'. (The Unnkulia games as far as I can remember started
off the introduction with actual insults.) Apart from being both
irritating and disheartening, this tends to create the above problem: if
I'm Mr. Nobody, why am *I* expected to solve all these problems? (Not
to say this approach doesn't sometimes work well - 'The Hitch-Hiker's
Guide to the Galaxy' and 'Trinity' being examples - but a lot of the
time it doesn't.)

I preferred, as you mentioned, the 'Enchanter' approach, in which there
is a valid reason why your character should be attempting to solve the
problems (and prefer still more the RPG approach in which you are
allowed to start off by creating your own character).

--
"To summarize the summary of the summary: people are a problem"
Russell Wallace, Trinity College, Dublin
rwal...@vax1.tcd.ie

Phil Goetz

unread,
Oct 10, 1994, 3:23:23 PM10/10/94
to
In article <37a9k0$i...@nntp.interaccess.com>,
S.P.Harvey <sha...@interaccess.com> wrote:

>To me, part of the enjoyment of playing IF are the games that have the
>following question and answer set as an integral part of the design:
>
>Q: What am I supposed to do in this game?
>A: You're supposed to discover what you're supposed to do.
>
>This seems a good method of immersing the player in the world you're
>creating. The player will find bits and pieces of the plot scattered
>about in different places - solving puzzles, talking to NPC's, examining
>objects - and slowly but surely, all the bits will fall into place and
>the jigsaw puzzle finally shows a complete picture.

I will never, ever try to do this again. That was what was supposed
to happen in _Inmate_ -- the player has all sorts of gradual revelations
of increasing horror before he can figure out who he is, why he's there,
and what is going on (all things he had known before, but forgotten).
The result was that everyone who ever played the game lost interest
and gave up before they figured out what was going on.

Phil

Felix Lee

unread,
Oct 10, 1994, 9:50:07 PM10/10/94
to
Phil Goetz:

>I will never, ever try to do this again. That was what was supposed
>to happen in _Inmate_ -- the player has all sorts of gradual revelations
>of increasing horror before he can figure out who he is, why he's there,
>and what is going on (all things he had known before, but forgotten).
>The result was that everyone who ever played the game lost interest
>and gave up before they figured out what was going on.

I think I agree. It's hard to control the pace of this type of slow
revelation. Unless the player has other factors to keep him playing,
then it's likely to be just frustrating.
--

Felix Lee

unread,
Oct 10, 1994, 2:14:23 PM10/10/94
to
Paul Munn:

>I haven't uploaded the package yet because whenever I look at it, there
>are always little improvements I can make. It's not the best code
>around, and it could be written more compactly (but then it wouldn't be
>as clear as it is now).

David Baggett:


>Phooey! Upload it anyway! :)

ditto. Perfection's overrated.

more thoughts on plot-modelling. Structured design and an adaptive
hint system are just two applications of plot-modelling. I've also
been wondering about using it in narrating the story to the player.

at the moment, narrative is often written into location descriptions.
This tends to be annoyingly repetitive and often obscures features of
the location. If you're just moving around or trying to solve a
puzzle, it gets in the way.

What I'd like to do is separate the narration from the description,
probably by sticking them in different windows. The narrative should
never repeat, and it's better if it resembles linear story text.
Probably better. I'll have to see, if I get that far..
--

S.P.Harvey

unread,
Oct 11, 1994, 3:02:01 AM10/11/94
to
Phil Goetz (go...@cs.buffalo.edu) wrote:

: I will never, ever try to do this again. That was what was supposed


: to happen in _Inmate_ -- the player has all sorts of gradual revelations
: of increasing horror before he can figure out who he is, why he's there,
: and what is going on (all things he had known before, but forgotten).
: The result was that everyone who ever played the game lost interest
: and gave up before they figured out what was going on.

Well, to defend my point a bit, I wasn't planning to drop a player in the
middle of nowhere and make them figure out everything. Well, maybe
someday I'll do an IF-port of Kafka's "The Trial" to take this to an extreme.

My game sets up the player as a very specific "person", with a very vague
"goal". The goal will become more and more concrete and precise as the
game progresses, while the player tries to reach that goal. This is
rather hard to explain without an example.

The game begins. You've been assigned to find out what happened on the
secret space base. At this point, we don't know what happened or how to
fix it. We've got to find out. You arrive at the eerily-silent base and
begin to explore. Your first order of business will be finding your way
around, then figuring out what happened. After several incidents which
lead you closer to the truth (which has been becoming slowly clearer),
you manage to track down the evil aliens which have popped in from a
nearby dimension no one even knew about and decided the space station
will make a nice doorstop.

The trick is to make the goal far-removed yet still intimately linked
with the opening of the game. This way, the goal gets discovered as the
game goes on. The real trick is to make sure none of the puzzles are
gratutious, they must all add small fragments to the storyline, or the
entire effect is ruined. Tricky, but done well, could be great.

David Baggett

unread,
Oct 11, 1994, 3:19:43 AM10/11/94
to
In article <FLEE.94O...@simula.cse.psu.edu>,
Felix Lee <fl...@cse.psu.edu> wrote:

>umm, if you really can solve the puzzles in any order, then it doesn't
>matter which puzzle you work on next.

Yeah, yeah; I knew someone would say that. :)

If you can solve the puzzles *absolutely* in any order, yes, you're right.
In general, I don't think it's possible to write a nontrivial game like
this, so we're left with a situation where many of the puzzles can be
solved in any order. Since some still have to be solved in a particular
order, you get the problem.

>The game becomes hard when:
>- you can see puzzles that are currently unsolvable, and
> - you can't tell if a puzzle is unsolvable or not, or
> - you don't notice a solvable puzzle because it's camouflaged.

Right on.

>If the game had an internal model of its own plot, then it could guarantee
>a useful hint.

Right again; I think it works out well in practice. You'll be able to try
it out for yourself soon, when the _Legend_ demo is out. Or you could try
it *right now* by grabbing David Allen's hintified Ditch Day Drifter from
ftp.gmd.de. (Look for if-archive/programming/tads/examples/adhint.zip)

>I think you have to throw away the traditional conception of plot.

Milan Kundera touches on this with regard to music in _The Book of Laughter
and Forgetting_. In essence, he expresses utter amazement at the fact that
one generation of composers could be so arrogant as to assume that the
whole of music before them was worthless, and that to do anything new and
interesting they had to start over from scratch. (In fact the motivations
of the serialists were not so plain, but he makes the point well.)

If you are going to throw out thousands of years of literary tradition, you
had better offer something damn good to replace it. Simulationist IF shows
absolutely no signs of having the potential to come within a stone's throw
of even _The Wizard of Oz_ in artistic merit.

This may change; I'd be a fool to say it couldn't. But I'd argue that the
technology required is well beyond our reach, so worrying about it now is
kind of silly.

>Simulationist IF will probably be quite different from literary IF.

Yes, it will suck. :) (Hey, you baited me!)

>There's no particular reason a player's role in a simulation has to
>have a "good" plot in a literary sense. It just has to have the
>potential of having a good plot.

I doubt I'll ever agree with this. I don't think story- or novel-length
fiction can have any real merit without plotting. But that's just me...

>There may be a coherent overarching story involving the kingdom, like
>the return of the king or the breaking of the world. But the player
>does not have to be at the center of this, not all the time.

In my view, it comes down to this: either humans design the plot, and the
plot must be followed by the player, or a machine designs the plot and the
plot is lousy (I predict this will be true for our lifetimes at least), or
the player "finds his own plot" -- i.e., there is no predetermined plot,
and the player does whatever he darn well feels like -- killing his evil
uncle who murdered his father, or, hell, just blowing it off and eating
lots of fully-simulated donuts. Again, lousy plot -- most people are
simply incapable of inventing an exciting plot on the fly, and those who
*could* do it probably wouldn't enjoy it like they'd enjoy going along with
another author's carefully-crafted plot. (It's hard work!). It's much like
the difference between writing a text adventure and playing the same text
adventure. I don't like playing my own games -- why bother? I already
know all there is to know about them. (Like Hindemith once said: "Did you
think you could play it better than I imagined it in my head?!")

I'm not saying a simulation wouldn't be "fun". That's another argument.
I'm saying a simulation can't support real plot (sort of by definition,
actually), and hence that it can't be "good literature".

There may be something that simulationist IF can be that is not "good
literature" but is nonetheless interesting from an artistic standpoint.
But I've seen no evidence of it, and can't personally imagine it.

>It's certainly not something I'm going to produce; I'm still stuck at
>trying to design puzzle-adventures.

My advice: don't worry about making something fun; worry about making
something that says something that you care about. If you care about it,
chances are someone else will too, and you will make a connection. That is
what art is all about. Art is communication so effective it's a thing of
wonder.

(IMHO, as always.)

David Baggett

unread,
Oct 11, 1994, 2:16:41 PM10/11/94
to
In article <FLEE.94Oc...@simula.cse.psu.edu>,
Felix Lee <fl...@cse.psu.edu> wrote:

>But that's not really the point. Simcity isn't a work of literary/artistic
>merit, and it can't be evaluated on those terms.

What terms can I evaluate it on, then? It's fun ... for a while. But does
it offer anything but fleeting gratification? Plenty of things are like
Simcity -- they're enjoyed and forgotten. It sounds like you're taking the
"I'm not talking about art" way out. OK. But I don't see why anyone
should take simulationist IF seriously if it's just going to produce fadish
diversions.

As much as I like video games, I can't think of a single one that people
will be talking about in a hundred years. Is this a problem with the
medium? Is the video game form one that cannot sustain really
thought-provoking works?

Think about Rubik's Cube. It's certainly a challenging puzzle, and fun,
and it was a fabulously successful product in its day. But does it
communicate anything to future generations?

Of course things do not have to be "art" to be worthwhile. But it is
important to get straight what has artistic potential and what doesn't,
because in the long run works of art are much more important to the society
than things like Rubik's Cube.

Felix Lee

unread,
Oct 11, 1994, 4:42:35 PM10/11/94
to
Dave Baggett:

>What terms can I evaluate it on, then? It's fun ... for a while. But does
>it offer anything but fleeting gratification? Plenty of things are like
>Simcity -- they're enjoyed and forgotten. It sounds like you're taking the
>"I'm not talking about art" way out. OK. But I don't see why anyone
>should take simulationist IF seriously if it's just going to produce fadish
>diversions.

Not every novel is a work of art either. Most books are faddish
diversions. :)

I don't really know if it's possible to consider a simulation a work
of art. It seems to me more like a medium for doing art than art
itself. Perhaps it's just too soon to tell; the technology is just a
couple decades old, and not particularly easy to work with.

>As much as I like video games, I can't think of a single one that people
>will be talking about in a hundred years. Is this a problem with the
>medium? Is the video game form one that cannot sustain really
>thought-provoking works?

Probably. Anything that has a substantial realtime component that
involves hand-eye coordination becomes sport, not art. People don't
worry much about the deep meaning of tennis or football.

>Think about Rubik's Cube. It's certainly a challenging puzzle, and fun,
>and it was a fabulously successful product in its day. But does it
>communicate anything to future generations?

How about group theory? Rubik's Cube is a standard example of a
simple but interesting group. It's also an elegant physical design.
It's science and technology, and art in the sense of design.

actually, design is a good point. I'm an occasional fan of design in
general, and I think you can argue that Rubik's Cube occupies a place
in design similar to, say, one of Bach's two-part inventions in music.

>Of course things do not have to be "art" to be worthwhile. But it is
>important to get straight what has artistic potential and what doesn't,
>because in the long run works of art are much more important to the society
>than things like Rubik's Cube.

heh. this could become a long argument. but I'm not the one to argue
this, because I'm basically agnostic about the role and importance of
art. I know what I like when I see it. :)

(but offhand, much as I admire James Joyce, I'd say that television is
more important to society in the long run.)
--

Felix Lee

unread,
Oct 11, 1994, 5:42:49 PM10/11/94
to
Molley the Mage:
>I mean, consider the paradigm
>here -- how many times do you want to type
> >Shovel the horse manure

well, heroic fantasy doesn't dwell much on mucking out stalls either.
clearly that's not what the game is going to be about, any more than
you spend all your time typing:

>lift left leg
Okay.
>move left leg forward
Okay.
>drop left leg
You take a step. You need to breathe soon.
>breathe
The cool air is refreshing.

What the game *will* be about, well, I don't know. Maybe it'll be a
little like nethack. Maybe it'll be based on interaction with NPCs.
Maybe in a few years I'll take a stab at it.
--

S.P.Harvey

unread,
Oct 11, 1994, 5:39:40 PM10/11/94
to
David Baggett (d...@case.ai.mit.edu) wrote:

: Think about Rubik's Cube. It's certainly a challenging puzzle, and fun,


: and it was a fabulously successful product in its day. But does it
: communicate anything to future generations?


Here's a spin:

Think about the original mainframe Adventure. It's certainly full of
challenging puzzles and it was a fabulously successful product in its

day. But does it communicate anything to future generations?

I think it does. And I think Rubik's Cube communicates many things.
Rubik's Cube has evolved into a sort of 20th-century mythologic puzzle.
We all know the story of the twisted mathematicin Erno Rubik and his
brain-wrenching creations which have allowed untold millions to exercise
their minds.

Put simply, folks, IF is an entertainment form, not something you're
going to win a Nobel prize in literature for. I agree strongly with the
concept of making games with a definite plot, theme, and setting, just
like any other work of fiction. However, there must be some
entertainment value inherent in the product; in this case, it's
brain-teasers and puzzles. In "serious" literature, the entertainment
value is experiencing the story and interpreting the author's
intentions. IF can combine both. If we lean too far to one side, it'll
get tedious.

If I was interested in writing the Great American Novel, I'd do so. But
I'm not. I'm interested in writing a game of exploration and puzzle
solving with a really good plot stringing it all together in coherent
fashion.

Scott


--
----------------------| S.P. Harvey |--------------------------

"Most of the world was mad. And the part that wasn't mad was angry.
And the part that wasn't mad or angry was just stupid.
I had no chance. I had no choice." - Charles Bukowski, 'Pulp'
----------------------| sha...@interaccess.com |--------------------------

S.P.Harvey

unread,
Oct 11, 1994, 5:47:14 PM10/11/94
to
Felix Lee (fl...@cse.psu.edu) wrote:

: Maybe simulation-IF will necessarily be literary, but I don't see any
: particular reason for this to be true. (other than the way it relies
: on words, that is. :)
: --

Okay, but here's a point. All works of fiction (save the truly
experimental) involve some nature of conflict. We all remember these
from sixth-grade English lessons:

Man vs. man
Man vs. environment
Man vs. self

In IF, the player is dropped into the body of "man" who has to then
overcome the conflict brought on by the story. In straight fiction, the
author manipulates the character in order to overcome the obstacles.

Extrapolating this concept: Simulation-IF cannot work; There's no place
for real interaction. Or if there is, it's no longer a straight
simulation. It's a typical "Adventure" game with all the puzzles taken
out and nothing but locations and non-functional objects.

May as well connect to an E-Text gopher server and read online fiction,
simply hitting "return for next page".

Personally, I'm not too keen on the concept of spending my time with the
electronic equivalent of a "Choose Your Own Adventure" paperback, no
matter how good the story is.

Scott

--
----------------------| S.P. Harvey |--------------------------

"I do not know which to prefer, / The beauty of inflections /
Or the beauty of innuendoes / The blackbird whistling /
Or just after." - Wallace Stevens
----------------------| sha...@interaccess.com |--------------------------

Felix Lee

unread,
Oct 11, 1994, 9:27:39 AM10/11/94
to
Dave Baggett:

> Since some still have to be solved in a particular
>order, you get the problem.

but if it's blatently obvious which puzzles are unsolvable, then you
don't have that problem.

>If you are going to throw out thousands of years of literary tradition, you
>had better offer something damn good to replace it. Simulationist IF shows
>absolutely no signs of having the potential to come within a stone's throw
>of even _The Wizard of Oz_ in artistic merit.

heh. right. But that's not really the point. Simcity isn't a work


of literary/artistic merit, and it can't be evaluated on those terms.

There seems to be three different modes of IF gameplay: puzzle-IF,
story-IF, and simulation-IF. If you want to push the boundaries of
IF, pick one direction and run with it. I think the directions have
to diverge. No one game is likely to be the ultimate everything.

(I'm interested in all three directions, so my point-of-view is
sometimes a little schizophrenic. :)

Maybe you can argue that simulation-IF shouldn't be called interactive
*fiction*, but by that argument I don't think I'd call Zork IF either.
--

Felix Lee

unread,
Oct 12, 1994, 2:02:25 AM10/12/94
to
S.P.Harvey:

>Extrapolating this concept: Simulation-IF cannot work; There's no place
>for real interaction. Or if there is, it's no longer a straight
>simulation. It's a typical "Adventure" game with all the puzzles taken
>out and nothing but locations and non-functional objects.

umm, you've lost me. I don't see how you reached this conclusion.
Doesn't nethack work? doesn't it have conflict and functional
objects?

I've just started playing Enhanced, and so far I'd categorize it as
puzzle/simulation/story, in that order. The Multidimensional Thief is
also puzzle/simulation/story. With a little work, I think it wouldn't
be too hard to make a text game that's simulation/puzzle/story.

simulation/story/puzzle or story/simulation/puzzle is something else
though. I don't know what these would look like.

hmm.. this a/b/c categorization doesn't really work; it's not very
accurate. oh well. I think it illustrates a point.

I've probably exhausted my soapbox on this topic...
--

S.P.Harvey

unread,
Oct 12, 1994, 12:26:31 PM10/12/94
to
Gerry Kevin Wilson (whiz...@uclink.berkeley.edu) wrote:
: <pained look> All right. I give up. For over a week I've followed this
: thread, dropped comments supporting plot, etc. Now, I fear that someone
: is going to have to explain to me, as if to a little child, what the
: difference between simulation and current IF is. The lack of puzzles?
: Doesn't sound like it. The lack of traditional puzzles, maybe, but all
: puzzles? The laack of any non-obvious interaction in the world setting,
: is that what you mean? Please, help a senile old teenager out here.

Gerry:

I won't explain it to you as if you were a little child, because you
aren't. I'll explain it to you like a rational adult, which you are! :)

Anyway, this isn't the definitive explanation, just my
interpretation. <disclaimer mode:OFF>

What I understood as the debate between Sims and IF was more an analogy
to the earlier comparison: "Schindler's List" as SIF (Simulation
Interactive Fiction, tm).

To me the term "simulation" implies doing everything involved in keeping
a complex system functioning: fly an airplane, build a city, etc. The
challenges in this type of setting are strictly functional challenges,
there are not artificial "puzzle" constructs in the system.

Thinking more about this now, after phrasing the above paragraph the way
I did... (sound FX of neurons firing, light bulb going on)

We may actually be on to something here. Gerry, sorry if I'm going to
blur the lines yet again.

I suppose you could call a work of IF a simulation if all the puzzles in
the game are >strictly< out-growths of the world the game is set in.
That would mean discovering the use of objects, finding keys, etc. If
the game is not magical, no mystical puzzles. No "abstract" puzzles.

If we want to extrapolate (why not?), I suppose we could fail the
original Zork triology (in places) on this rule. Zork was essential a
string of non-related puzzles in a common setting. Don't misunderstand
me, Zork was (and is) a true milestone and I love and respect it.

Again, the line blurs whenever we cross the road into a "speculative"
environment, when there is no definitive set of rules on how things behave.

I admit I'm floundering, but I'm still trying!

Gerry Kevin Wilson

unread,
Oct 12, 1994, 1:46:34 PM10/12/94
to
In article <37h2nn$g...@nntp.interaccess.com>,
S.P.Harvey <sha...@interaccess.com> wrote:

>Gerry:

>I won't explain it to you as if you were a little child, because you
>aren't. I'll explain it to you like a rational adult, which you are! :)

Nice that someones thinks so, anyways.

>Anyway, this isn't the definitive explanation, just my
>interpretation. <disclaimer mode:OFF>

#insert "disclaim.h"

>What I understood as the debate between Sims and IF was more an analogy
>to the earlier comparison: "Schindler's List" as SIF (Simulation
>Interactive Fiction, tm).

A Schindler's List Sim? I'd have to say it can't be done without
stripping away the good bits of the movie. The true gemstone in the
movie was Schindler, and the changes he undergoes. "Why did I keep the
car? I could've saved 10 lives!" You could set up a simulation to let
someone just watch what goes on and take no part, but then you've created
a book with a moveable camera, so to speak. Interesting the first time.

>To me the term "simulation" implies doing everything involved in keeping
>a complex system functioning: fly an airplane, build a city, etc. The
>challenges in this type of setting are strictly functional challenges,
>there are not artificial "puzzle" constructs in the system.

Hmm, this describes an ordinary old well-written piece of IF to me. I'm
not saying that there have been many, but they exist, all the same.

>Thinking more about this now, after phrasing the above paragraph the way
>I did... (sound FX of neurons firing, light bulb going on)
>
>We may actually be on to something here. Gerry, sorry if I'm going to
>blur the lines yet again.

That's ok. I must confess that my post was more an attempt to get folks
thinking "What exactly are we arguing about here?" This argument has
cropped up 3 or 4 times since I've been reading r.a.i-f, and it never
reaches a satisfactory conclusion. I think a lot of it is that we are
sitting in the dark, shouting at one another, without bothering to wait
for the neurons to fire. So a simulation is IF w/o artificial puzzles.
This does not disallow puzzles that are a side effect of the
environment. So, is Starcross a simulation? How about Suspended? If
not, why not? I could argue that everything in both these games that is
a 'puzzle', logically follows from the setting.

>I suppose you could call a work of IF a simulation if all the puzzles in
>the game are >strictly< out-growths of the world the game is set in.
>That would mean discovering the use of objects, finding keys, etc. If
>the game is not magical, no mystical puzzles. No "abstract" puzzles.

:) Just what I thought you said.

>If we want to extrapolate (why not?), I suppose we could fail the
>original Zork triology (in places) on this rule. Zork was essential a
>string of non-related puzzles in a common setting. Don't misunderstand
>me, Zork was (and is) a true milestone and I love and respect it.

Ah yes, but maybe Zork simulates perfectly the Great Underground Empire.
Maaybe GUE is a world filled with puzzles and traps for the unwary
adventurer....

>Again, the line blurs whenever we cross the road into a "speculative"
>environment, when there is no definitive set of rules on how things behave.
>I admit I'm floundering, but I'm still trying!

Don't worry, Scott. The whole concept is slippery, and kind of sidles
away when you look directly at it. We have yet to seperate what really
differentiates a simulation puzzle from a normal puzzle. No odd items
that bewilder you with their meaning? No tower of hanoi logic puzzles?
No spellcasting puzzles? Again, very slippery. Once I have a clear
definition, I'll argue, but for now, I'll pass.

--
<~~~~~E~~~G~~~SIGHT~UNSEEN~~~LOST~IN~THE~FOG~~~CYBER~CHESS~~~SPAG~~~|~~~~~~~>
< V R I O Software. We bring words to life! | ~~\ >
< T "We at Vertigo apologize for the delay. Sorry." | /~\ | >
<_WATCH for Avalon in early AUGUST!___wh...@uclink.berkeley.edu__|_\__/__>

David Baggett

unread,
Oct 12, 1994, 3:30:51 PM10/12/94
to
In article <JAMIE.94O...@akeake.its.vuw.ac.nz>,
Jamieson Norrish <ja...@akeake.its.vuw.ac.nz> wrote:

>Does that make it clearer what I mean? The character can wander
>around, doing something here, something there, and getting their own
>story built out of what is done.

Is it different from "plot graph" IF I described? In my mind, "doing
something here and there" would correspond to choosing an arbitary path in
the plot graph. So long as the author has ensured that every path will be
a good plot, this works. Otherwise, the player may do things here and
there and end up having followed a lousy plot -- one that does not work out
at all, or one that has no real climax; a plot that seems pointless
overall.

>[Having good plots] both for the little stories in a game, and also


>the overall story that is created by what the character does and
>experiences.

I'm not sure a sequence of good plots sums to a good overall plot. I don't
think that's likely. But you could always impose some grand plot, in which
the hand-crafted subplots are simply local changes. (This is a more
sophisticated form of checkpointing, I guess.)

>I think that it should be possible to do something along these lines
>without having to sit down and generate an exlicit plot for each path.

I'm skeptical that plots that are in any way generated (using near-future
technology) can be high-quality. But it remains to be seen.

>Also, since there might be many plot trees in a given game (rather like
>your option 3, perhaps), a leaf may be an ending, but not *the* ending.

Right.

>That's a very simply example of course, but I hope it serves to illustrate

>both that stories might spring from events ... and also that plots don't


>need to be totally designed beforehand.

I would never argue that good plots could not *arise*; just that the
arising plots won't *usually* be good.

Jamieson Norrish

unread,
Oct 13, 1994, 8:12:11 AM10/13/94
to
In article <37g5pc$8...@agate.berkeley.edu>

whiz...@uclink.berkeley.edu (Gerry Kevin Wilson) writes:

<pained look> All right. I give up. For over a week I've followed
this thread, dropped comments supporting plot, etc. Now, I fear
that someone is going to have to explain to me, as if to a little
child, what the difference between simulation and current IF is.
The lack of puzzles? Doesn't sound like it. The lack of
traditional puzzles, maybe, but all puzzles? The laack of any
non-obvious interaction in the world setting, is that what you
mean? Please, help a senile old teenager out here.

Okay, I'll have a go, although I think I'm as confused as you are. I
don't quite know where the sudden discussion of simulation came from.
But anyway...

I don't see simulation as being a lack of puzzles, although it rather
sounds as though other people do. The difference between simulation
and what IF is now is roughly this: simulation does not limit the
world on the basis of an enforced (author written) plot. Once the
extent of the world is defined, then the only limitations on player
movement/freedom are those introduced by the limits of the parser,
mechanics, and world. This is in contrast to IF as now where the
author's plot limits what can be done.

Does that help? I don't actually see where simulation fits into any of
the current discussions, but maybe when more posts come through (I
tend to get them in the wrong order), I'll understand.

Jamie

Molley the Mage

unread,
Oct 11, 1994, 2:59:22 PM10/11/94
to

> In article <yeehaw> fl...@cse.psu.edu (Felix Lee) writes:

> That, basically, sums up what I meant. That the plot isn't imposed on
> the player to follow, but there is a plot or plots inherent in the
> setting, in the NPCs, in actions, and the players is left to find them
> and make of them what they will.

I didn't want to get in on this thread, but I can't hold myself back
any longer. Please remember that everything which follows is IMHO.
:)

I can think of a million reasons *not* to write a game in this way.
And, when you think about it, what is the real difference between a
game such as you describe (where the plots are "hidden") and a regular
game (where the plot is stated or implied). The end result of the
game still involves the player following the plotline written and
directed by the author. The only difference is that in your game, the
player has to flounder around until he finds one or more of the plots,
while in the traditional IF mode he already has some idea what he's
supposed to be doing.

Additionally, have you simulationist people ever considered the fact
that this type of thing is *boring*? Take the example of a game where
you start out as as stable boy and rise to become the saviour of the
world (or whatever) based on incumbent plots that you either discover
or which appear at timed intervals. What's your character going to be
doing in the meantime? Mucking out stalls? Exercising horses?
Knitting a warm blanket for his Auntie? I mean, consider the paradigm


here -- how many times do you want to type

>Shovel the horse manure

Or perform the point-n-click equivalent -- before you get tired of it?
The simple fact is that "real life" is tedious, in that we repeat
tasks over and over again in the course of a day. Sure, real life
isn't "boring" as such, because each day is a little different. But
try sitting in front of your computer screen playing a game where you
go through a regular day in the life of a medieval peasant. Get up
with the sun, work the fields, bow and scrape whenever a Lord or
Knight happens by, give up 90% of your bounty to the Baron, come home
for a crust of bread, and go to sleep. Repeat 364 times. Is this
really the makings of an epic game?

And think about the author. If I am going to write a game, I want
to write a *story*. Is there any value in my writing a story and
then concealing it so that the player might never discover it? Of
course not! I'm investing my time and energy into this game; I want
it to be played in its entirety. I might make some things more
difficult than others, through the use of puzzles, but eventually I
want every player of my game to get out of it the entire story that I
put into it. Otherwise, it's just an exercise in typing (or mousing,
or whatever). Imagine a novel that was nothing more than the
descriptions of everyday activities of a typical American family.
Nothing out of the ordinary happens; they just get up in the morning,
eat breakfast, go to work or school, come home, etc. There is
absolutely no conflict or story whatsoever, and the characters don't
mature in any way. Now extend the idea to a computer game. See what
I mean? The novel wouldn't be any more fun to write than it would be
to read. You notice that there are no titles in the biography section
called "Joe Average: Just Another Ordinary Guy."

Simulationist IF is a bad idea. Plain and simple. Simulations are
great, in their own way -- I love SimCity and Railroad Tycoon more
than almost any other games. But directionless simulation of
"reality" (or someone's imaginary world, or whatever) is not only
impossible with today's technology, but also fairly pointless. Out of
conflict rises plot; out of plot rises story; out of story rises
enjoyment. Without enjoyment, there is no game. QED. And, you'll
note, the "simulation" games have "plots" as well -- in SimCity,
events happen. Disasters occur. Crises crop up. You have to deal
with these things. What's that except for plot? Sure, the events are
quasi-random. But the author of the game decided under which set of
circumstances each event would be possible. Isn't that analagous to
plot in an IF game?

Sure, games like the Ultima series have lots of sub-plots and quests
which are hidden and which the player must discover. Some of them
aren't even required to finish the game. I have no problem with this.
I can even see something like that working in IF. But try playing an
Ultima game WITHOUT going on any of the quests. Fun for a while, but
it gets old quick. Even a game like Ultima 6, where almost everything
"works" like it should, and where you can get a job to earn a few gold
coins, is boring unless you follow the story.

I realize that I've actually strayed from the original article I was
responding to; I've sort of degenerated into an attack on
simluationist IF. Sorry about that; this post is sort of the
combination of several other posts I've been meaning to make but never
did. But I hope my points remain coherent. This isn't really an
attack on simluations at all -- but I remain firm in the belief that
IF games without an author-driven plot would be directionless and no
fun to play or write. Therefore, they would be pointless. Show me a
game that proves me wrong, and I'll gladly eat this post. :)

> Jamie

Sean
--
M. Sean Molley, CS Department, Western Kentucky University, Bowling Green, KY
Internet: mol...@wkuvx1.wku.edu | That is not dead which can eternal lie....
--
"Yes, Ytalk -- the next best thing to the next best thing to being there."
-- David Hammett

Jamieson Norrish

unread,
Oct 13, 1994, 8:50:38 AM10/13/94
to
Note: I followed up to this without reading it all the way to the end
first. Mistake! However, I'm not going to take the time to change all
my responses. Just try to ignore the slightly exasperated tone, and
look at the content. Sorry Sean.

In article <1994Oct11...@wkuvx2.wku.edu> mol...@wkuvx2.wku.edu
(Molley the Mage) writes:

And, when you think about it, what is the real difference between a
game such as you describe (where the plots are "hidden") and a
regular game (where the plot is stated or implied). The end result
of the game still involves the player following the plotline
written and directed by the author. The only difference is that in
your game, the player has to flounder around until he finds one or
more of the plots, while in the traditional IF mode he already has
some idea what he's supposed to be doing.

There are certainly large differences. In character-based IF, the plot
is forced onto the player. There is a plot, and it will be followed,
or the game does not work. If the stories come from the world, then
the player is not forced to follow them, and the game can still
succeed.

Additionally, have you simulationist people ever considered the fact
that this type of thing is *boring*?

Whoah. Where did this "simulationist" thing come from? I am not
talking about simulation. I am talking about having author crafted
stories imbedded in the objects, NPCs, and events of the world, rather
than being intrinsically connected to the character.

But if we were talking about simulation, then it comes down to a
question of degree. Many novels or films are termed realistic, yet
don't show hum-drum boring existence. Why should this be the case in
simulation IF? Of course you can't model everything, and you almost
certainly wouldn't want to.

Is this really the makings of an epic game?

Are we all trying to create epic games?

And think about the author. If I am going to write a game, I want
to write a *story*. Is there any value in my writing a story and
then concealing it so that the player might never discover it?

Misconception alert! Misconception alert! Who said anything about
"concealing" it? Sure, some stories might be harder to find than
others. But how about coming across someone who's car has broken down.
The potential stories in that are hardly concealed, yet the story is
not intimately linked with the character, who can walk off without
doing anything, or not even come across the person - without the game
failing as it would in traditional IF where it is necessary to solve
all the puzzles to "win".

I'm investing my time and energy into this game; I want it to be
played in its entirety. I might make some things more difficult
than others, through the use of puzzles, but eventually I want
every player of my game to get out of it the entire story that I
put into it.

And why does it matter that the player might require several games to
explore the entire story? Or is a one-shot game better?

There is absolutely no conflict or story whatsoever, and the
characters don't mature in any way. Now extend the idea to a
computer game. See what I mean?

Why do you assume that there is no conflict and no story? I think you
are talking about something quite different to what I was advocating.
But to keep on this "simulation" tack - is there no conflict in real
life? No character development? I think not. As an interesting side
note, there is a soap of sorts in Britain which simply follows the
lives of some people living in a flat. No script, just these people
being filmed in various parts of their everyday lives. It's a really
popular program, and it has conflict and character development.

And, you'll note, the "simulation" games have "plots" as well -- in
SimCity, events happen. Disasters occur. Crises crop up. You
have to deal with these things. What's that except for plot?

So what's your problem?! If you've just said you can have events,
crises, and plots in a simulation game, then why is that not possible
in IF?

Sure, games like the Ultima series have lots of sub-plots and
quests which are hidden and which the player must discover. Some
of them aren't even required to finish the game. I have no problem
with this. I can even see something like that working in IF. But
try playing an Ultima game WITHOUT going on any of the quests.

Now, at last, we get onto something which replies to what I said. It
is not necessary to remove all plot which requires the character,
although I don't think the player should be forced into it to the
extent that he or she is now.

Even a game like Ultima 6, where almost everything "works" like it
should, and where you can get a job to earn a few gold coins, is
boring unless you follow the story.

Which is, I think, why I said I'd like to see stories in the objects,
characters, and events in the world.

I realize that I've actually strayed from the original article I
was responding to; I've sort of degenerated into an attack on
simluationist IF. Sorry about that; this post is sort of the
combination of several other posts I've been meaning to make but
never did.

Ooops. :)

Show me a game that proves me wrong, and I'll gladly eat this post.
:)

You're on!

Jamie

Felix Lee

unread,
Oct 11, 1994, 4:50:57 PM10/11/94
to
Oh, I missed a point. I said that Simcity couldn't be evaluated on
the basis of "literary/artistic" merit. "literary" is important. It
may have artistic merit, but it wouldn't be a literary one, any more
than a Brahms concerto would have literary merit.

Gerry Kevin Wilson

unread,
Oct 12, 1994, 8:04:53 PM10/12/94
to
In article <JAMIE.94O...@akeake.its.vuw.ac.nz>,
Jamieson Norrish <ja...@akeake.its.vuw.ac.nz> wrote:
>In article <37g5pc$8...@agate.berkeley.edu>
>whiz...@uclink.berkeley.edu (Gerry Kevin Wilson) writes:
>
>Okay, I'll have a go, although I think I'm as confused as you are. I
>don't quite know where the sudden discussion of simulation came from.
>But anyway...
>
>I don't see simulation as being a lack of puzzles, although it rather
>sounds as though other people do. The difference between simulation
>and what IF is now is roughly this: simulation does not limit the
>world on the basis of an enforced (author written) plot. Once the

Ok, you're taking the plotless simulation approach. Gotcha. Note, to be
a true simulation, this means you desire a complete lack of any string of
events that forms a coherant story. If a plot forms from your actions,
the author anticipated these actions, and designed a situational
response. Or, the author may have set up a system of rules for all
objects in the simulation and then let them loose to see what happens. I
believe this is what everyone is aiming at. Also note, no coherant plot
will arise from such a system. You cannot create order from chaos
without intervening. Either the rules bash you over the head with a
story, or there is no story, and to be quite frank, no audience. Just to
stay in practice, here's a numbered list.

1. The number of rules involved in such an endeavor are such that no one
person is going to be able to predict the outcome of their interaction.

2. Debugging such a product is a logical nightmare. You've heard of
Hell, well welcome to Hell's big brother.

3. More than 50% of your plots are likely going to be gibberish.

In conclusion, simulation is a toy unless it has a very specific, very
limited scope. Ask any statistics major what happens to a simulation as
you add more and more variables to it, such as you would have to with a
game.

>extent of the world is defined, then the only limitations on player
>movement/freedom are those introduced by the limits of the parser,
>mechanics, and world. This is in contrast to IF as now where the
>author's plot limits what can be done.
>
>Does that help? I don't actually see where simulation fits into any of
>the current discussions, but maybe when more posts come through (I
>tend to get them in the wrong order), I'll understand.

People were arguing in favor of simulation over traditional IF.
Simulation, in my opinion, is going to require a much larger number of
man hours, create a less logical, less satisfying product, and in general
just give you less bang for your buck. I would rather entrust the
creation of a story to a human than a set of rules any day. Humans have
imagination, computers don't. You say, "Ah ha! But a human designed
those rules!" So what? A human mind can only grasp so many logical
connections at once before it flees screaming into the woods. But, I'll
give everyone arguing for sim-IF the benefit of the doubt. Produce a
better product than Avalon, in less or equal amount of time, and show me
how to repeat your feat, and I will embrace sim-IF. $5 will get you $10
that I get a lot of theoretical rebuttals, but not one person takes up
the challenge. I'm a very down to earth kind of guy. I hear, "Hey, it's
better to write a set of rules than a story." I think, "Funny, then why
don't writers do that to write books? There are certainly enough grammar
programs and such out there to write something coherant, (I never said
logical, mind you.) In fact, there have been several attempts at
sim-FICTION. Odd that I don't see any of them on the bestsellers list.
My my, how peculiar indeed."

Now comes the nugget of the argument, why is a set of rules for a
computer to follow less useful than a human author? Answer: The human
author is a better computer, without realizing the rules he follows.

So, the final bit that remains for me to chew at is this. sim-IF almost
certainly has a use. An entire game based on it is going to be illogical
crap. On the other hand, why don't the simulationists get together and
decide where it is most efficient to use simulation within the context of
a normal IF product? Show me some new and better methods of coding
realistic NPCs. Give me a way to generate a realistic landscape
following rules I create, but don't actually have to run through myself.
Create a better physical model of gravity to use in space games. Make me
a system of magic that is logical, endlessly variable, and bounded by
only a few rules. (Hell, I've done this one myself. I haven't
programmed it into anything, but I don't have a game to use it in yet
either.) Create me a realistic animal simulation, such as a bear. Hell,
make me a good brown bear and I'll put it in Avalon. Don't get caught
into thinking that nearly 4,000 years of human authors taught us diddly
shit about how to create IF. Instead, contribute to the pool of
knowledge rather than endlessly bickering that 'My system's better than
yours.' Hell, that's what I've been meaning to say throughout this
post. Don't argue at us, that doesn't accomplish anything, SHOW, DON'T TELL.
I rest my case, and my fingers.


--
<~~~~T~~~~~~~~~~~~~~~~LOST~IN~THE~FOG~~~~~~NO~RELEASE~DATE~~~~~~~~~~~|~~~~~~~>
< R I A lonely shipwreck survivor is swallowed by a mysterious | ~~\ >
< E G fog bank in the Bermuda Triangle, and meets his destiny... | /~\ | >
<_V____SOFTWARE___MEET_Y...@uclink.berkeley.edu__|_\__/__>

Gerry Kevin Wilson

unread,
Oct 12, 1994, 8:32:11 PM10/12/94
to

Oh great, sure, fine! Just post a new standpoint on the 'simulation'
argument right after I post a big message oon how we can integrate the
two. Suure, be that way. *grumble* Now I've got to respond to this new
explanation in order to avoid looking like a goober. Ok, why force
the player into the plot? Well, why not? That's why the player has
chosen my game to play, on the basis of its advertised plot. Certainly
not on his endless confidence in my yet unseen abilities as a writer. I
just don't see the point of making the plot to my game optional. The
player already gets to wander freely. The things around Avalon almost
never aggressively involve the player. Hell, the player can fart around
for eternity for all I care. It's not going to get them to the end of
the game, mind you. The only way that's gonna happen is through activity
on the part of the player. Many games are like this. The NPCs, in
Avalon, will try to draw the player into the plot, if the player refuses,
and walks away, the NPC mentally shrugs and then ignores the player until
he changes his mind. Certain situations, such as ignoring chivalry's
rules in front of a knight, will miff the NPC, and reduce his opinion of
the player, but usually the NPC won't do anything about it.

On the other hand, I have certain pipe-lines in Avalon. Once you
enter a certain area, you WILL do this certain part of the adventure, or
you will die. Why? Because this is where I deem it appropriate to put a
sink or swim spot. I make some of the basic points of Avalon in these
spots. I want the player to see these thing, because >I< have put two
years of work into Avalon, and >I< have decided the player will visit
these spots. Like it or lump it, it's the author's prerogative to have
prerogatives(sp?). One of my prerog's is that I wish to say something
about the Vietnam war. Since I'm writing the game, no one can stop me.
You can refuse to buy the game, of course, boycotts have been quite
effective in the past. But then, you'll not only miss what I have to
say, you'll also miss the story that I've wrapped the point in.
Basically, what I'm saying here is that enforced plot or not, the author
has a reason for the way he does things. If you write games that are
amusing, with no point, there is no reason to place pipe-lines in your
game. It's quite easy to design around them. On the other hand, if
there is something important that you want to say, the only way to be
sure of it getting said is to maneuver the player into a situation where
the relevant text is bound to show up. You can try sprinkling less
obvious hints here and there about your point, sure, go ahead. I don't
do things like that. If I want to say something, I goddamn say it. That
is why I choose to use the constraints of plot.

Jamieson Norrish

unread,
Oct 12, 1994, 8:39:50 AM10/12/94
to
In article <37a9k0$i...@nntp.interaccess.com> sharvey@interaccess (
S.P.Harvey) writes:

True, and an excellent point made and well-taken. However, we
cannot (as designers) allow so much openness in our games that the
player is able to form their own conclusions as to what is going
on.

Why? I mean, what are the reasons for restricting things in this
fashion? People have pointed out some already, mostly how
unimplementable it is. However, I don't think it is. I think it's more
a matter of mindset. In fact, believe it or not, it did link in with
the title of the post I made - Attitudes to playing. Like I said
there, I don't play to particularly solve everything, learn all the
things in the game, and put it back on the shelf.

Rather, I want to be able to wander about, without being forced into a
rigid scheme. To make the distinction I made earlier, I want the
stories to be in the objects, the events, and the other characters.
Not *just* in my character. By which I mean (since I haven't been
clear about this yet :) that instead of having the stories all
directly and totally relate to my character, I'd like to be able to
intrude onto a story that doesn't have my character as a *necessary*
component.

This, as far as I can see, is not simulation - merely an extension of
IF in a different plane.

To me, part of the enjoyment of playing IF are the games that have
the following question and answer set as an integral part of the
design:

Q: What am I supposed to do in this game?
A: You're supposed to discover what you're supposed to do.

I think that my approach doesn't so much invalidate this, but extend
it. The answer is not discover what you're supposed to, but discover
what's there. That isn't going to take one game (even with saves and
undos and other bells and whistles :), because by the character's
actions, the world state will change, eliminating various outcomes.

Jamie

Jason Noble

unread,
Oct 12, 1994, 8:58:00 PM10/12/94
to

Felix Lee writes:

>|> What I'd like to do is separate the narration from the description,
>|> probably by sticking them in different windows.


Greg Ewing writes:

>I'm not sure I like the idea of multiple windows.

[deletia]

>I'm happier when the action and the narrative are
>somehow merged into a single stream of consciousness.
>Perhaps the narrative could be displayed in a
>distinctive font? That might help to signal to
>your brain "here comes a block of narrative, go into
>sit-back-and-listen mode".


It's interesting how people's tastes differ on issues like this. I agree
with Greg; I would hate to see multiple windows in an IF piece.

I strongly agree with Greg's point about it being desirable for action and
narrative to be "merged together into a single stream of consciousness". It
seems to me that this is what's so good about old-fashioned, non-interactive
novels: a single one-directional stream of a limited number of typographical
symbols is capable of inducing an almost transcendent experience in the
reader.

Novels are generally typeset in some unobtrusive font, and generally do not
have pictures and other distractions. IMHO, the idea is for the reader to
*not notice the medium*, and spend their time *absorbing the message*. I
think that a complicated, multi-window IF environment will be a medium
that's difficult to ignore, and will thus reduce the impact of the message.
(By "message" I mean no more than the content of the work: I don't assume
that all IF must have some deathless, deep message).

Certainly it's a problem for IF to have narrative inserted into room
descriptions, so that each time you look at the room again you see the same
long body of text that describes something you should only have seen/
felt/experienced the first time.

However, I feel that liberal use of techniques like the firstseen method in
TADS can get around this problem. If the designer takes the time to write
an initial, rich room description, a shorter room description for repeated
use, plus extra blocks of narrative text that get called when certain
conditions are true (such as when first entering, or when first in the room
at the same time as a certain actor), the player's experience will be richer
and undisturbed by silly repetitive text.

It's my goal to try and do all this *without* multiple windows, and
*without* relatively subtle things like a different font for narrative text.

(Don't take this as an attack, Felix: your recent comments have been
extremely interesting and thought-provoking).

Regards,


---------------------------------------------------------------------------
Jason Noble | jno...@bunyip.bhs.mq.edu.au
National Centre for HIV Social Research | jno...@laurel.ocs.mq.edu.au
Macquarie University, Sydney, Australia | ph. (61 2) 850 8667
---------------------------------------------------------------------------

Jamieson Norrish

unread,
Oct 12, 1994, 8:46:39 AM10/12/94
to
In article <37am3o$k...@psuvax1.cse.psu.edu>

So IMHO: that's what a game needs -- _motive_. Not necessarily
a big or a fancy one, just enough of one to get the player off
his or her duff and get moving.

This is an excellent point, and one which also ties in with my
approach to games - only using another solution. Given no particular
motivation at the beginning (but perhaps a desire to explore),
presenting a wide array of events, people, and objects offers the
character a myriad of things to become interested in. And since the
puzzles are not puzzles - to be solved and then left behind) - but
situations, where there is the option of walking away, there hopefully
wouldn't be a lack of potential motivation.

Hmph, one day I'll be able to talk coherently about this. :)

Jamie

Felix Lee

unread,
Oct 13, 1994, 12:25:07 AM10/13/94
to
Greg Ewing:
>I'm not sure I like the idea of multiple windows. I've
>played games where there are separate windows for
>narrative, inventory, map, etc. and I didn't really
>like them. I found it too distracting looking between
>different windows all the time to really get involved
>with the story.

yah, I have qualms about that too. But I'll see how it goes. Is
looking at an inventory window really any more distracting than having
your screen filled with your inventory list?

>I'm happier when the action and the narrative are
>somehow merged into a single stream of consciousness.

I don't want action separate from narrative. action is part of the
narrative. but I think description is not. Doesn't anyone find it
annoying when every time you "LOOK" it tells you stuff like, "You
think to yourself, 'So this is what a breathtaking view is like', and
you almost asphyxiate in wonder as a cloud of iridescent butterflies
gently aspirates from the flowers, dances a pattern of joy around your
dizzied head, and disappears into the gentle morning mist that
blankets the valley like the warm woolen sweater your Aunt Martha gave
to you last year on your birthday---ah, what a birthday that was."

Most of this I'd call narrative, not description. Half the time I'm
looking just to find an exit.
--

Felix Lee

unread,
Oct 12, 1994, 10:41:27 PM10/12/94
to
Gerry Kevin Wilson:

><pained look> All right. I give up. For over a week I've followed this
>thread, dropped comments supporting plot, etc. Now, I fear that someone
>is going to have to explain to me, as if to a little child, what the
>difference between simulation and current IF is. The lack of puzzles?
>Doesn't sound like it. The lack of traditional puzzles, maybe, but all
>puzzles? The laack of any non-obvious interaction in the world setting,
>is that what you mean? Please, help a senile old teenager out here.

ah, sorry. I think I brought up "simulation", and I'll admit I only
have a vague idea what I'm talking about. :)

One way of approaching it is, there are three ways of playing an IF
game:
- puzzle-oriented: the game is a series of obstacles to overcome.
- story-oriented: the game is a narrative you read.
- simulation-oriented: the game is a virtual world you run around in.

another way of approaching it is looking at where game design starts:

- puzzle-based: you start with a couple puzzles, build the objects to
support the puzzle, and attach a story to it.

- story-based: you start with a story, build some puzzles to support
the story, and build the objects to support the puzzles and the story.

- simulation-based: you start by building objects that have
interesting behavior. Then you come up with puzzles involving these
objects, and attach a story to it.

(Of course, this isn't how game design really works; design tends to
be a messy process.)

Here's one example of the distinction I make between simulation and
puzzle in IF:

The Multidimensional Thief has a "portable hole" object, and you can
stick it pretty much anywhere to make a hole you can walk through.
There are a few critical places where you need it to solve certain
puzzles, but the portable hole is more versatile than that. This is a
"simulation-level object".

In contrast, many text adventures have a shovel object that doesn't do
anything except in one or two places. Most of the time you get
something like, "The ground's not suitable for digging." The shovel
was added to support a particular puzzle or two, so this is a
"puzzle-level object".

Of course, there are good reasons for restricting the usefulness of
shovels. But if you start out by building a simulation-level shovel,
then you'd probably end up with a game that looks very different.
Maybe something like Lode Runner :). (A shovel is probably a poor
choice for a simulation-level object in a pure-text game.)

lightsources and containers are simulation-level objects that are
commonly used. Offhand, I can't think of anything else commonly used
that I'd consider a simulation-level object. Oh, I forgot keys. Keys
and locks are perfectly simulated :). Many puzzle-level objects are
keys in disguise.

hmm. Maybe this isn't so vague an idea after all. writing things out
is great for clarifying thought..
--

Felix Lee

unread,
Oct 12, 1994, 11:20:09 PM10/12/94
to
just a few small things, then I'll shut up :).

Gerry Kevin Wilson:


> You cannot create order from chaos without intervening.

Yes you can. Order can arise spontaneously from chaos given the right
framework. It doesn't have to be a complicated framework either.
This is what the whole field of self-organizing systems is all about.

> On the other hand, why don't the simulationists get together and
>decide where it is most efficient to use simulation within the context of
>a normal IF product

Umm, I don't know about anyone else, but yes, that's one of the things
I'm interested in.

Maybe the ideal plotless sim-IF will turn out to be "illogical crap",
but it's a useful concept to play with. It's a dream; I'm a dreamer.
But I do have a reasonable grasp of reality and of what can and can't
be done. It's the boundary that fascinates me.

>Don't get caught
>into thinking that nearly 4,000 years of human authors taught us diddly
>shit about how to create IF

but 4000 years of human authors have taught us nothing about computers
and complexity. It's all new. Noone even had the vaguest idea that
something like the Mandlebrot Set could exist. Very few authors write
anything remotely realistic about computers, even in SF.

>Don't argue at us, that doesn't accomplish anything, SHOW, DON'T TELL.

Working on it. Gotta warn you, my followthrough tends to be pretty
poor. Occupational hazard of being a dreamer :). Too many things
catch my attention..
--

Greg Ewing

unread,
Oct 12, 1994, 1:26:23 AM10/12/94
to

In article <FLEE.94Oc...@simula.cse.psu.edu>, fl...@cse.psu.edu (Felix Lee) writes:
|>
|> What I'd like to do is separate the narration from the description,
|> probably by sticking them in different windows.

I'm not sure I like the idea of multiple windows. I've


played games where there are separate windows for
narrative, inventory, map, etc. and I didn't really
like them. I found it too distracting looking between
different windows all the time to really get involved
with the story.

I'm happier when the action and the narrative are


somehow merged into a single stream of consciousness.

Perhaps the narrative could be displayed in a
distinctive font? That might help to signal to
your brain "here comes a block of narrative, go into
sit-back-and-listen mode".

Greg Ewing, Computer Science Dept, +--------------------------------------+
University of Canterbury, | A citizen of NewZealandCorp, a |
Christchurch, New Zealand | wholly-owned subsidiary of Japan Inc.|
gr...@cosc.canterbury.ac.nz +--------------------------------------+

David Baggett

unread,
Oct 12, 1994, 12:36:55 AM10/12/94
to
In article <37f0ms$9...@nntp.interaccess.com>,
S.P.Harvey <sha...@interaccess.com> wrote:

>We all know the story of the twisted mathematicin Erno Rubik and his
>brain-wrenching creations which have allowed untold millions to exercise
>their minds.

I'm sure there's quaint story behind the hula hoop, too, but no one
remembers it because it's just not very interesting. There is a colossal
difference between these two kinds of communication. One is trivial
historical footnoting; the other is a tremendous achievement that people
can appreciate deeply even hundreds of years later.

>Put simply, folks, IF is an entertainment form, not something you're going
>to win a Nobel prize in literature for.

Admitting defeat before we're out of the starting gate?!

It remains to be seen what artistic potential interactive fiction has.
I've put in my two cents on the simulation issue already. That's not to
say that other kinds of IF can't be artistic.

>However, there must be some entertainment value inherent in the product; in
>this case, it's brain-teasers and puzzles.

Who ever said that art isn't entertaining? I never said that IF should be
dreary and obtuse; on the contrary, I think IF can offer much more
fulfilling entertainment than brain-teasers.

>In "serious" literature, the entertainment value is experiencing the story
>and interpreting the author's intentions. IF can combine both. If we lean
>too far to one side, it'll get tedious.

Do you mean to imply that serious literature is tedious? I'll admit that
not every book is for everyone, but I can think of few things as enriching
and thought-provoking as reading a top-notch "serious" literary work. I'll
resuse an example from a previous post: _Of Mice and Men_ is short, very
easy to read, is technically virtuosic, and makes a tremendous impact on
the reader. It's serious literature, but it has none of the pomposity or
elitism that I fear you are alluding to. There is plenty of silliness put
between author and reader by people who wish to complicate rather than
simplify; just ignore the criticsm and read the books! (I'm on _Of Mice
and Men_ because I recently re-read it and was re-amazed by it.)

>If I was interested in writing the Great American Novel, I'd do so. But
>I'm not. I'm interested in writing a game of exploration and puzzle
>solving with a really good plot stringing it all together in coherent
>fashion.

Nothing wrong with that, and you may produce the best puzzle game ever
made, and people will love you for it. Personally, I've already written a
puzzle game and a "half & half" game, and doing so has sort of turned me
off to puzzle-centric IF and on to looking to do something more rewarding
(for both author and reader). This is my bias; you do not have to make it
yours for me to be happy! :)

Felix Lee

unread,
Oct 13, 1994, 4:05:20 AM10/13/94
to
Jason Noble:

>IMHO, the idea is for the reader to
>*not notice the medium*, and spend their time *absorbing the message*. I
>think that a complicated, multi-window IF environment will be a medium
>that's difficult to ignore, and will thus reduce the impact of the message.

yup. Keeping the interface uncluttered and unobtrusive will be
tricky. In a story-oriented game, the narrative window should be the
primary focus, except when you're just navigating or browsing.

>However, I feel that liberal use of techniques like the firstseen method in
>TADS can get around this problem.

Oh, probably.

It's a little strange that stuff like this isn't easier to do. Like,
one construct I've been using in my pseudo-code is a "once" clause:
once {
say "You find a cedarwood box"
score 5
move box to player
}
This clause protects itself from being executed more than once during
a game. It's essentially just
if (! didthisthing) {
didthisthing := 1
...
}
but easier to read and write. Syntactic sugar. One side benefit is I
don't have to invent a variable name.

This could probably be generalized to a clause that expresses a
sequence of alternatives. Like, for the action "get fire":
first {
"You start to reach for the fire, but the beauty of the";
" flames moves your spirit, and you decide the fire would";
" be better off living free and undisturbed.";
} then {
"The fire has a happy life on its own, remember? You"
" don't really want to mess with it.";
} then 2 times {
"You can't have it."
} then {
"I give up. You're burned by the flames. Happy now?"
} then forever {
"You're burned by the flames."
}

It's more syntactic sugar. I'm probably going to implement something
like this in my Tcl/Tk prototype, because it'll be easy to do.

>(Don't take this as an attack, Felix: your recent comments have been
>extremely interesting and thought-provoking).

thanks. I'm just having fun, but it's nice to know I'm being useful
too :).
--

Gerry Kevin Wilson

unread,
Oct 12, 1994, 4:12:28 AM10/12/94
to
In article <FLEE.94Oc...@simula.cse.psu.edu>,
Felix Lee <fl...@cse.psu.edu> wrote:
>
>I've just started playing Enhanced, and so far I'd categorize it as
>puzzle/simulation/story, in that order. The Multidimensional Thief is
>also puzzle/simulation/story. With a little work, I think it wouldn't
>be too hard to make a text game that's simulation/puzzle/story.
>
>simulation/story/puzzle or story/simulation/puzzle is something else
>though. I don't know what these would look like.

<pained look> All right. I give up. For over a week I've followed this

thread, dropped comments supporting plot, etc. Now, I fear that someone
is going to have to explain to me, as if to a little child, what the
difference between simulation and current IF is. The lack of puzzles?
Doesn't sound like it. The lack of traditional puzzles, maybe, but all
puzzles? The laack of any non-obvious interaction in the world setting,
is that what you mean? Please, help a senile old teenager out here.

--
<~~~VERTIGO~~~~~~~~~~~~THE~BRASS~LANTERN~~~~~~ISSUE~1~INCL~W/AVALON~~|~~~~~~~>
< In the irreverent tradition of _The New Zork Times_ comes The | ~~\ >
< Brass Lantern, an informative newsletter from Vertigo Software. | /~\ | >
<___SOFTWARE____________...@uclink.berkeley.edu__|_\__/__>

Jamieson Norrish

unread,
Oct 14, 1994, 6:17:34 AM10/14/94
to
In article <37htj5$3...@agate.berkeley.edu>

whiz...@uclink.berkeley.edu (Gerry Kevin Wilson) writes:

I said:

>I don't see simulation as being a lack of puzzles, although it
>rather sounds as though other people do. The difference between
>simulation and what IF is now is roughly this: simulation does not
>limit the world on the basis of an enforced (author written) plot.

Ok, you're taking the plotless simulation approach. Gotcha. Note,


to be a true simulation, this means you desire a complete lack of
any string of events that forms a coherant story.

Except that you can imbed certain things into the characters and
objects. Sure, things may turn out in wierd and wonderfully
unpredicted ways, but that doesn't mean that there is no story.

Also note, no coherant plot will arise from such a system. You
cannot create order from chaos without intervening.

I'm not sure I totally agree with this. I think, as I've said, that
stories can be present which are not forced down the character's
throat, but can in fact lead to coherent and interesting stories.

Now comes the nugget of the argument, why is a set of rules for a
computer to follow less useful than a human author? Answer: The
human author is a better computer, without realizing the rules he
follows.

I think you've moved onto something else again here. The difference is
between having a plot forced onto the character, through which that
character must proceed in order for the player to get any value from
the game at all, and a game in which there are plots/stories, but they
are not individually vital to the game - either on the meta level of
enjoyment, or for the "completion" of the game. IF today is written
with very strict goals in mind - you aim to "win", that is the point,
and that is where the enjoyment comes from. But how about widening the
horizon a bit, and increase the number of options for the character?

Instead, contribute to the pool of knowledge rather than endlessly
bickering that 'My system's better than yours.' Hell, that's what
I've been meaning to say throughout this post. Don't argue at us,
that doesn't accomplish anything, SHOW, DON'T TELL.

Hey, calm down here, Kevin. I haven't seen any post, by myself or
anyone else, doing a comparison of systems. I haven't seen a single
person (except possible myself) arguing that simulation is good in
itself. I admit that I've proposed a different way of doing IF (which
is *not* simulation pure and simple) and have defended this view
against other people's criticisms of it, and tried to explain what I
mean better. But I'm certainly not arguing at anyone, nor do I think
that all other IF is useless or even not as good, even in theory.

Jamie

Mike Threepoint

unread,
Oct 13, 1994, 11:26:33 AM10/13/94
to
The Felix Lee writes:
=> >But what happens if you plant the tree somewhere else?

> In the bad history, the tree was planted at X-
> and caused event P-. In the good history, the tree gets planted at X+
> and causes event P+. If you plant the tree anywhere else, the result
> is a neutral event P0 which doesn't accomplish your goal.

True, and that's how most adventure games would handle it (if they
don't straightjacket you into only being able to plant the tree in
the place the author wants it). But how much richer to add mostly
neutral but potentially useful effects from planting it elsewhere...

For example, planting the tree next to the house itself in the
distant past provides a way to get out the window and escape the
burning house later.

Or if you plant it a little further from the house, the house will still
be destroyed, but the tree will still have an old tree house or a
swing hanging from a branch, a poignant reminder. And perhaps you
could take the rope ladder and use it as another way to get out the
window of the burning house.