Default parser responses: how do they affect the gaming / authorship experience?

16 views
Skip to first unread message

Dennis G Jerz

unread,
Jan 3, 2002, 6:04:21 AM1/3/02
to
In Inform, the programmer signals the end of the game by setting a
variable named "deadflag". Hypothesis: The name of this variable
encourages new authors to come up with antagonistic IF scenarios, in
which the PC must fight to survive. (I don't have my TADS manual with
me... what's the equivalent way to end a TADS game?)

Was the name "deadflag" suitably applied because IF authors had
already come to the realization that killing the PC was a convenient
storytelling shortcut? Is it a way to make the game world seem
responsive enough to recognize and respond to the player's desire to
make a radical change to the "ideal" narrative, but since the change
was fatal, it was short-lived, and hence required the programmer to
invest little time in order to implement it -- at least until such
time as we an instruct the Holodeck to improvise a story based on
parameters that we provide.

I've been collecting stories & interviews with Will Crowther. Most
people know that, during the time he created "Colossal Cave
Adventure," he was an avid caver in the 1970s, and also a D & D
player. I've also found references to him playing chess and posing
logic puzzles for the amusement of his companions, having a great
sense of direction, and being interested in natural language
processing.

Consider this: If Crowther's hobby was something other than caving,
would the history of gaming have included so many dungeon crawls?

If you're with me so far, and agree that this one decision that
Crowther made did affect the development of computer games for a long
time, then it's probably not too much of a stretch to suggest that
other decisions Crowther (and others) made in the history of IF, or
the decisions that the current generation of IF storytellers are
making now, may have tremendous unforeseen effects.

All the things that IF was good at doing in 1975 (manipulating
objects, language games), graphic computer games are still good at
doing. The things that IF was not good at -- the things that Crowther
didn't implement in his original simulated universe (complex NPCs,
dynamic plots) -- are still elusive.

Scott Adams says that when Adventure International began adding
graphics to their titles, fans wrote in to say, "The pictures I made
in my head were better than yours." A company that puts resources
into graphic and sound design can get richer game play that can be
generated dynamically. Plot is finite, and many a fine game has
gotten poor reviews due to low "replayability".

Yet the animators for Monsters Inc. worked in a complex simulated
universe that handled such things as the motion of fabric, the effect
of wind on hair, water absorption, etc. That meant that animators
didn't have to animate every ripple of fabric or lock of hair.
Monsters Inc. is, of course, a linear narrative.

Will plot-creation tools, or chatterbots with the complexity of
Galatea as NPCs, bring new audiences to games that have traditionally
expected a different kind of gameplay? Is "plot" a resource that can
be manipulated to this degree? I don't think many gaming companies
have much faith in that concept. I recently read a transcript of a
speech by the lead designer for Deus Ex, who noted that, in order to
advance the story, there were certain events that had to happen all
the time -- a particular alarm being triggered, for example. Gamers
often responded with hostility to that scripted plot point, since
elsewhere in the game it was possible for stealthy characters to
disable alarms.

When I was doing my English Ph.D., I heard quite a bit about
post-structuralism and the "death of the author," along with praise of
hypertext as a "writerly" genre -- a text that not only invited reader
participation, but demanded it. While I'm not sure how productive it
will be to apply those concepts directly to IF (it certainly wasn't
productive to apply those concepts to hyperfiction), the goal of a
game designer is generally to give more control to the player. In IF,
I think a more realistic goal is improved writing, that encourages the
player to behave in a way that will unlock the most aesthetically
satisfying narratives that the author has chosen to code.

I think that this means "authorship" will, in the future, be
democratized... shared among the members of a design team, and also
taken on by the players in future IF. Even solitary authors will use
software packages that reflect the experience of multiple designers,
playtesters, and upgrades driven by user requests.

(I haven't played "Majestic" yet... it sounds promising, but is it
really interactive? Or just a linear story dished out via faxes,
e-mails, fake web sites, chatbots, and phone calls?)

There's been much recent discussion on r*if on NPC interaction. Of
course the parser has gotten much better, but the core problems in IF
are still the same, and won't be solved by the Holy Grail of FPS and
MMORPG designers -- an ever more accurate simulation of physics.
Location, containment, proximity, illumination, etc. are being
developed with ever-further complexity, permitting players of graphic
action games to improvise solutions to problems that the designers
didn't foresee.

Along with textual IF's motions towards a more robust parser, we have
probably seen the expectations of the player rising
disproportionately. When the player expects to be able to perform
more actions, the player probably has more exposure to the game
world's default responses for actions that the designer didn't
foresee.

While a graphic game designer might tweak the gravity-simulation
subroutine, IF authors who want special textual effects will sometimes
painstakingly re-write the library files. "Ad Verbum" comes to mind
("No! Nibbling nappies not normal!"), as does any game that involves a
point of view other than second person present tense ("Muse") or the
tone of the narration ("Varicella"). The PC in chapters 1 and 3 of
"Fine Tuned" has a very different world view than the PC in chapters 2
and 4, and various default messages would have been inappropriate for
one or the other characters.

But what of the countless Inform games that include these gems, which
are part of the default library?

>take me
You are always self-possessed.

>examine me
As good looking as ever.

Is there a particular style to Inform parser responses -- a style that
influences the experience of writing such games?

Is the Inform parser really more polite and gentlemanly than the
Infocom parser? Does anybody find some of the Inform responses just a
bit too... precious? ("Real adventurers do not use such language."
[nudge-nudge, wink-wink.])

I find that at least some of the flavor of the original Advent is
affected markedly by the Inform behavior added to the game as part of
the 1993/4 port to Inform.

For example, ">enter building" should probably work, but since the
building object doesn't have the flag "enterable", you get the
standard Inform response, "That's not something you can enter."

(More on this at http://www.uwec.edu/jerzdg/orr/articles/IF/canon/Adventure.htm
)

The Crowther/Woods version would occasionally say something like, "I'm
not allowed to tell you any more about that, so I will repeat the room
description." In a similar situation, the 1994 Z-code version will
say something like "That's not a word you need to use in the course of
the game." In Return to Pirate's Island 2, Scott Adams has the
parser say something like, "This is terribly embarrassing, but I don't
understand the word 'X'."
But if someone were to play Kent Tessman's HUGO version of Adventure,
which was based on Graham Nelson's Inform version, which in turn was
based on Dave Bagget's TADS version, which was in turn based on Don
Ekman's DOS version, which was based on Wood's expansion of Crowther's
original, then... uh... I think I lost my train of thought while
trying to track down that geneology.
(See http://www.rickadams.org/adventure/e_downloads.html)
My point is that what is essentially the same game can change
markedly, based on the language used to create it. Roger Firth's
"Cloak of Darkness" website is an interesting case study.

Non-programmers who don't understand why the game can print out a
word, but not accept it as input, often find the experience disruptive
enough to damage their ability to enjoy the game. After I first
introduce IF to my students, I have them do a little fill-in-the-blank
coding exercise, and then the next time they play IF they are
generally much more appreciative of the work that goes into coding.

Scott Adams has predicted that within the next 5 years, there will be
sophisticated IF story generating tools on the market, designed with a
much lower learning curve. I think X is right when he suggested, in a
thread on commercializing IF, that story-generation tools will
probably play a large role.

While the Usenet community often laments what happened when
technologically naive AOL users began flooding the public newsgroups
that had previously "belonged" to the technological elites, the
Internet really emerged as a social force when it became accessible to
a wide range of people outside the traditional group of programmers
and researchers.

If that's the case, then it's probably safe to say that new IF authors
will probably lean very heavily on default responses. And if that's
the case, then the default responses, may affect the future of
interactive narrative as strongly as Crowther's decision to create an
underground treasure hunt.

Dennis G. Jerz


P.S.

As an amusing side-note, I was just re-reading "The Two Towers." In
the chapter "The Road to Isengard", Gimli the Dwarf and Legolas the
Elf argue about which are better -- forests or caves. Gimli has a
wonderful set of speeches about the wonders of exploring caves, about
how horrible it is to plunder their natural treasures. Considering
the function and fate of the dwarves in "Adventure," it would almost
seem the game advocates the Elvish world view.

Magnus Olsson

unread,
Jan 3, 2002, 7:26:25 AM1/3/02
to
In article <792c6202.02010...@posting.google.com>,

Dennis G Jerz <jer...@uwec.edu> wrote:
>In Inform, the programmer signals the end of the game by setting a
>variable named "deadflag". Hypothesis: The name of this variable
>encourages new authors to come up with antagonistic IF scenarios, in
>which the PC must fight to survive.

I don't think the *name* of the flag is very influential compared
to the fact that there are two default ways to end an Inform game:
by winning or dying; the default end-of-game messages are
"You have won" and "You have died".

But scenarios where the PC must fight to survive are IMHO rather
rare; death in IF usually seems to occur by walking into traps,
drinking poisonous potions, falling off cliffs and similar things.

--
Magnus Olsson (m...@df.lth.se, m...@pobox.com)
------ http://www.pobox.com/~mol ------

Adam Cadre

unread,
Jan 3, 2002, 7:47:53 AM1/3/02
to
Dennis Jerz wrote:
> The Crowther/Woods version would occasionally say something like, "I'm
> not allowed to tell you any more about that, so I will repeat the room
> description." In a similar situation, the 1994 Z-code version will
> say something like "That's not a word you need to use in the course of
> the game."

I believe that that message is reserved for when the player attempts to
use a word listed in the name field of the current room. Words that don't
occur in the game's dictionary at all produce the infamous "You can't see
any such thing." The same message is produced if the word *is* in the
game's dictionary but the object to which it refers is out of scope.
This was Graham's way of evading this little trick:

>EAT APPLE
I don't know the word "apple".

>EAT BANANA
You see no banana here.

(Player: "Ah ha! So there's a banana SOMEWHERE in the game!")

The problem with Graham's solution is that it assumes that everything the
player can see will be implemented, at least to the extent of producing a
"That's not important."-type message. So you end up with stuff like this:

>I
You are carrying a pair of gloves (being worn).

>LOOK AT MY HANDS
You can't see any such thing.

(Player: "Eeeagh! I have no hands! So where'm I wearing the gloves, on
my ears?")

Another problem with "You can't see any such thing." is that it appears
to be diegetic discourse -- ie, part of the story, rather than a
nondiegetic error message. Take the following exchange:

>EAT BANANA
You can't see any such thing.

To my mind, what has just happened in the story is that the character
in question has decided, "Hey, a banana sure would hit the spot," looked
around for one to eat, and sadly concluded, "Darn, it doesn't look like
there's a banana in this room." Me, I've come to prefer to make it fairly
explicit to the player that the command is erroneous, bracketing off the
response as nondiegetic:

>EAT BANANA
[That is not an object with which you can currently interact.]

This, by contrast, says, "Hey, player -- game here. You can't do that."
Nothing happens in the actual story. Now, some might call this an
unfortunate interruption of mimesis. But I tend to think of bracketed
messages as being essentially invisible: since they're not part of the
story I'm building in my mind as I play, they're less harmful than
messages which insert a lot of trying-to-find-objects-and-failing into
the story. Plus, it resolves this nicely:

>I
You're wearing a pair of gloves.

>LOOK AT MY HANDS
[That is not an object with which you can currently interact.]

(Player: "Hrm, not implemented. Okay, moving on then.")

I wouldn't take this approach to *every* game, but as LOCK & KEY will
attest, it is my preferred recourse as of January of '02.

-----
Adam Cadre, Brooklyn, NY
http://adamcadre.ac

joh

unread,
Jan 3, 2002, 8:12:27 AM1/3/02
to
I'm sure I'm going to hit a lot of things that have already been
discussed on this group, so if that bothers you, read no further.

As a newcomer to IF, I've already considered many of the topics you
bring up. The angle I'm proceeding from, though, is "Is it worth my
while to get into this as an author, or am I better off spending the
time on noninteractive short fiction?" And what that boils down to for
me is solving this dilemma: "Is IF just a gimmick for taking a
traditional story, chopping it up, and making it 'fun' to read, with
the side benefit that the audience is less judgmental about some of
the standard criteria for other prose (e.g., if the gameplay is
original, the plot need not be)?"

Re the influence of "deadflag," I think A Change in the Weather
provides counterexamples here. You need not kill the character to end
the story suddenly and with finality. On the other hand, once any game
goes into an unwinnable state, you could argue that everything from
there on out is a slow loss.

[I've snipped a lot, without always marking where. Sorry.]
>>>>> "Dennis" == Dennis G Jerz <jer...@uwec.edu> writes:

Dennis> Consider this: If Crowther's hobby was something other
Dennis> than caving, would the history of gaming have included so
Dennis> many dungeon crawls?

If not that, then something similar, just because it's easiest to
implement either a constrained space or a relatively empty open
space. The raif thread on depth of room description is exploring this
topic and coming up with some interesting ideas.

Dennis> All the things that IF was good at doing in 1975
Dennis> (manipulating objects, language games), graphic computer
Dennis> games are still good at doing. The things that IF was not
Dennis> good at -- the things that Crowther didn't implement in
Dennis> his original simulated universe (complex NPCs, dynamic
Dennis> plots) -- are still elusive.

Isn't that just because they're harder? I mean, as far as I've seen,
nobody's programmed a really convincing human character, in IF or any
other branch of programming. And multiple plots mean a multiplication
of effort. It seems logical in most cases to spend that effort on
making one plot more satisfying.

Dennis> A company that puts resources into graphic and sound
Dennis> design can get richer game play that can be generated
Dennis> dynamically. Plot is finite, and many a fine game has
Dennis> gotten poor reviews due to low "replayability".

Here's the crux. If good prose is an important part of the IF
experience, you can't really have "dynamic" plots. You can only have
multiple plots. Unless you think Eliza generates good prose.

Dennis> Yet the animators for Monsters Inc. worked in a complex
Dennis> simulated universe that handled such things as the motion
Dennis> of fabric, the effect of wind on hair, water absorption,
Dennis> etc....

And they flubbed it just as bad as most big-budget studios that work
in a nonsimulated universe that handles the motion of fabric,
etc. Because the limit of their vision was to clothe a standard
Hollywood plot in wacky costumes.

Dennis> ... Gamers often
Dennis> responded with hostility to that scripted plot point,
Dennis> since elsewhere in the game it was possible for stealthy
Dennis> characters to disable alarms.

Dennis> .... In IF, I think a more
Dennis> realistic goal is improved writing, that encourages the
Dennis> player to behave in a way that will unlock the most
Dennis> aesthetically satisfying narratives that the author has
Dennis> chosen to code.

That's the other crux :) In a "game" we want control. In a "story," we
have to be led. In IF, I think we want a convincing illusion of
control, while still being led. At least, I think that's what I want.

Dennis> I think that this means "authorship" will, in the future,
Dennis> be democratized... shared among the members of a design
Dennis> team, and also taken on by the players in future IF. Even
Dennis> solitary authors will use software packages that reflect
Dennis> the experience of multiple designers, playtesters, and
Dennis> upgrades driven by user requests.

I've very often flirted with the idea of trying to organize a team of
writers for a story. I've never seen more than two authors' names on a
work of prose fiction, but look at TV shows like the Simpsons, which
uses the wide knowledge of a _lot_ of people to create an incredibly
densely textured story.

I'm running out of time to get ready for work, so let me just agree
that the standard responses from Inform are totally
distracting. Especially when you're just starting a new
game. Everything is fresh and new, until you try something that
doesn't work. Suddenly you feel like you're playing Curses again.

joh

J. D. Berry

unread,
Jan 3, 2002, 10:48:47 AM1/3/02
to
jer...@uwec.edu (Dennis G Jerz) wrote:

> In Inform, the programmer signals the end of the game by setting a
> variable named "deadflag". Hypothesis: The name of this variable
> encourages new authors to come up with antagonistic IF scenarios, in
> which the PC must fight to survive. (I don't have my TADS manual with
> me... what's the equivalent way to end a TADS game?)

"Encourages" is too strong, but I would agree that it at least plants
the seed. I think most new authors already have in mind what they
want to do. If they're Doom oriented, they're already going the
death ending route. If, like Alan DeNiro with "Isolato", they have
a story concept in mind that they feel tranlates well to the medium,
they're not even thinking of a death ending (unless it truly fits
the story, obviously.)

> Was the name "deadflag" suitably applied because IF authors had
> already come to the realization that killing the PC was a convenient
> storytelling shortcut?

I don't think this was a storytelling shortcut. The deadflag then
was a combination of the philosophy of gaming at the time--kill or
be killed (chess and D&D, from your Crowther references)--and the purpose
of creating the games in the first place--they were fun diversions
designed to challenge (drive nuts) their colleagues. "Ha ha! I've stymied
you!"

While Crowther's prose is competent, I'm sure he wasn't after a
Nobel prize in literature. I don't think he was bypassing story,
because (just my speculation, of course) the story wasn't the point.

Noteworthy, too, was the computer's lack of memory. This affected
the story in many ways. The author hadn't room to write a complete
background on everything--the little things that gradually add up to make
today's IF a story experience.

> Is it a way to make the game world seem
> responsive enough to recognize and respond to the player's desire to
> make a radical change to the "ideal" narrative, but since the change
> was fatal, it was short-lived, and hence required the programmer to
> invest little time in order to implement it -- at least until such
> time as we an instruct the Holodeck to improvise a story based on
> parameters that we provide.

I think we're at the stage where authors are going
the less ideal narrative route, rather than the death route. Look
at Cadre, Short and Fischer games, for example. Sure, sometimes their
non-deaths are the "death" equivalent, but they always fit the situation
and story.

From your own words above, we are talking about a GAME world. The
Costikyan articles (via Brass Lantern (although the links aren't
working, currently)) on what makes a game a game are apporpriate here.
Taking from that, 1) there must be competition on some level, and 2)
there must be decision making. And it is here, I feel, where there
are shortcut temptations.

The danger of #1, above, is "OK, it's you the player versus me
the author." In #2, "OK, make a wrong decision and die." These
satisfy the GAME conditions, but they (may) seriously neglect the
STORY.

This again returns to the "NEW" author discussion. There's just
so much an IF author has to do. To a new author, way too much.
Integrating competition is complex. Incorporating
subtle decision-making is complex. Don't forget original storylines
and depth of world, too. Oh, and cool NPCs, please.

Thus, even if DEADFLAG were not the syntax, authors (especially
new ones) would still take this route. That's why I feel that
the "deadflag effect" is only a seed, not a full encouragement.
The seed will grow, but it's up to the author to shape its
future. There can and should be pruning.

> I've been collecting stories & interviews with Will Crowther. Most
> people know that, during the time he created "Colossal Cave
> Adventure," he was an avid caver in the 1970s, and also a D & D
> player. I've also found references to him playing chess and posing
> logic puzzles for the amusement of his companions, having a great
> sense of direction, and being interested in natural language
> processing.
>
> Consider this: If Crowther's hobby was something other than caving,
> would the history of gaming have included so many dungeon crawls?
>

As mentioned in other threads, a desolate, similarly
constructed landscape lent itself perfectly to the gaming style and
computer capacity of the 20th century. If Crowther didn't
write CC, someone else would have.


> If you're with me so far, and agree that this one decision that
> Crowther made did affect the development of computer games for a long
> time, then it's probably not too much of a stretch to suggest that
> other decisions Crowther (and others) made in the history of IF, or
> the decisions that the current generation of IF storytellers are
> making now, may have tremendous unforeseen effects.
>

Perhaps there's a gray area between the "natural progression"
(that is, SOMEONE would have done what X, Y or Z did, if X, Y or Z
hadn't) that I suggest above, and the "cause and irreversable effect
of X, Y or Z (that only they could have done)" that you do.

> All the things that IF was good at doing in 1975 (manipulating
> objects, language games), graphic computer games are still good at
> doing. The things that IF was not good at -- the things that Crowther
> didn't implement in his original simulated universe (complex NPCs,
> dynamic plots) -- are still elusive.
>

Elusively elusive. :)

I think they are do-able but for the time constraints placed
on humans.

> Scott Adams says that when Adventure International began adding
> graphics to their titles, fans wrote in to say, "The pictures I made
> in my head were better than yours." A company that puts resources
> into graphic and sound design can get richer game play that can be
> generated dynamically. Plot is finite, and many a fine game has
> gotten poor reviews due to low "replayability".
>
> Yet the animators for Monsters Inc. worked in a complex simulated
> universe that handled such things as the motion of fabric, the effect
> of wind on hair, water absorption, etc. That meant that animators
> didn't have to animate every ripple of fabric or lock of hair.
> Monsters Inc. is, of course, a linear narrative.
>

A thought-provoking paragraph, indeed.

The "automated" world of Monsters Inc. was still defined at some
level by humans. If I took their tools to develop my own Pixar
rip-of^H^H^H^H^H^H translation movie, it would still have the Monsters
look and feel. Of course, I could redefine the physics of
the world, change some of the libraries...

Well, you see what I'm getting at (you mention plot-creation
tools below).

Extrapolate to game-design tools, you'd get Doom clones, Myst clones,
etc...

Extrapolate to current IF tools (and your theory), you'd get
Cave Crawls where death ended the story at every turn.

Thus, we could have an IF tool that mirrored the Monsters inc.
abilities. But after the first few games, send in
the clones. They'd be slick-as-snot clones, but they would
still bore us if there wasn't Quality (as Pirsig defines it
in "Zen...").

It always comes down to author originality and how he/she
creates entertainment (in its many forms) artistically.

> Will plot-creation tools, or chatterbots with the complexity of
> Galatea as NPCs, bring new audiences to games that have traditionally
> expected a different kind of gameplay? Is "plot" a resource that can
> be manipulated to this degree? I don't think many gaming companies
> have much faith in that concept. I recently read a transcript of a
> speech by the lead designer for Deus Ex, who noted that, in order to
> advance the story, there were certain events that had to happen all
> the time -- a particular alarm being triggered, for example. Gamers
> often responded with hostility to that scripted plot point, since
> elsewhere in the game it was possible for stealthy characters to
> disable alarms.
>

Again, this is where human creativity and TIME come in to play
With both of those elements, that alarm situation could have been
alleviated. There's always the subjective factor, though. There's
going to be SOMEONE that doesn't like it, however you do it.

> When I was doing my English Ph.D., I heard quite a bit about
> post-structuralism and the "death of the author," along with praise of
> hypertext as a "writerly" genre -- a text that not only invited reader
> participation, but demanded it. While I'm not sure how productive it
> will be to apply those concepts directly to IF (it certainly wasn't
> productive to apply those concepts to hyperfiction), the goal of a
> game designer is generally to give more control to the player. In IF,
> I think a more realistic goal is improved writing, that encourages the
> player to behave in a way that will unlock the most aesthetically
> satisfying narratives that the author has chosen to code.
>

"Control" is a tricky word, anyway.

I fully agree with you about improving the writing.

> I think that this means "authorship" will, in the future, be
> democratized... shared among the members of a design team, and also
> taken on by the players in future IF. Even solitary authors will use
> software packages that reflect the experience of multiple designers,
> playtesters, and upgrades driven by user requests.

Sure. I don't think there's any way to avoid this. It's the
time issue, again.

My first recommendation to a new IF author is "change the stock
responses to reflect your world." Mimesis is made or broken when the
the player doesn't follow the script. "Oh no, I just got slapped
from the girl I touched" versus "Oh, I'm just playing a game."



> But what of the countless Inform games that include these gems, which
> are part of the default library?
>
> >take me
> You are always self-possessed.
>
> >examine me
> As good looking as ever.
>
> Is there a particular style to Inform parser responses -- a style that
> influences the experience of writing such games?
>
> Is the Inform parser really more polite and gentlemanly than the
> Infocom parser? Does anybody find some of the Inform responses just a
> bit too... precious? ("Real adventurers do not use such language."
> [nudge-nudge, wink-wink.])
>

Right. They won't fit too many games. I think of them as place-holders.

This is a good place to emphasize beta-testing. Having a diverse
group of testers can find these trouble spots and you can
minimize these barriers.

> After I first
> introduce IF to my students, I have them do a little fill-in-the-blank
> coding exercise, and then the next time they play IF they are
> generally much more appreciative of the work that goes into coding.
>

:) Yes, although this is true with just about any craft.
For example, children baking their first cake learn, "wow, what a lot
of work."

> Scott Adams has predicted that within the next 5 years, there will be
> sophisticated IF story generating tools on the market, designed with a
> much lower learning curve. I think X is right when he suggested, in a
> thread on commercializing IF, that story-generation tools will
> probably play a large role.
>

I still stand by what I said, above. The products will be more
sophisticated, but there will still be the same ratio of Quality.

> While the Usenet community often laments what happened when
> technologically naive AOL users began flooding the public newsgroups
> that had previously "belonged" to the technological elites, the
> Internet really emerged as a social force when it became accessible to
> a wide range of people outside the traditional group of programmers
> and researchers.
>

> If that's the case, then it's probably safe to say that new IF authors
> will probably lean very heavily on default responses. And if that's
> the case, then the default responses, may affect the future of
> interactive narrative as strongly as Crowther's decision to create an
> underground treasure hunt.

And when they do rely heavily, they will produce the same mediocre
stuff. And those who realize it takes a certain "caring" (OK,
enough with the motorcycle maintenance stuff!), will put the extra
efforts in.

>
>
> Dennis G. Jerz
>

Great post, Dennis.

>
> P.S.
>
> As an amusing side-note, I was just re-reading "The Two Towers." In
> the chapter "The Road to Isengard", Gimli the Dwarf and Legolas the
> Elf argue about which are better -- forests or caves. Gimli has a
> wonderful set of speeches about the wonders of exploring caves, about
> how horrible it is to plunder their natural treasures. Considering
> the function and fate of the dwarves in "Adventure," it would almost
> seem the game advocates the Elvish world view.

Perhaps there's one IF tool to bind us all? ;D

Jim

Peter Seebach

unread,
Jan 3, 2002, 11:40:25 AM1/3/02
to
In article <792c6202.02010...@posting.google.com>,
Dennis G Jerz <jer...@uwec.edu> wrote:
>In Inform, the programmer signals the end of the game by setting a
>variable named "deadflag". Hypothesis: The name of this variable
>encourages new authors to come up with antagonistic IF scenarios, in
>which the PC must fight to survive. (I don't have my TADS manual with
>me... what's the equivalent way to end a TADS game?)

I disagree - it's just that, if you aren't dead, why shouldn't you be allowed
to keep playing?

-s
--
Copyright 2001, all wrongs reversed. Peter Seebach / se...@plethora.net
$ chmod a+x /bin/laden Please do not feed or harbor the terrorists.
C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon!
Consulting, computers, web hosting, and shell access: http://www.plethora.net/

Magnus Olsson

unread,
Jan 3, 2002, 11:57:13 AM1/3/02
to
In article <3c348979$0$79557$3c09...@news.plethora.net>,

Peter Seebach <se...@plethora.net> wrote:
>In article <792c6202.02010...@posting.google.com>,
>Dennis G Jerz <jer...@uwec.edu> wrote:
>>In Inform, the programmer signals the end of the game by setting a
>>variable named "deadflag". Hypothesis: The name of this variable
>>encourages new authors to come up with antagonistic IF scenarios, in
>>which the PC must fight to survive. (I don't have my TADS manual with
>>me... what's the equivalent way to end a TADS game?)
>
>I disagree - it's just that, if you aren't dead, why shouldn't you be allowed
>to keep playing?

Because the story is over? Or because you've lost the game?

I understand your point, but I think this depends on where the game is
on the simulation/story scale, i.e. how much control the author wants
over the player.

Passenger Pigeon

unread,
Jan 3, 2002, 12:54:33 PM1/3/02
to
In article <792c6202.02010...@posting.google.com>,
jer...@uwec.edu (Dennis G Jerz) wrote:

> In Inform, the programmer signals the end of the game by setting a
> variable named "deadflag". Hypothesis: The name of this variable
> encourages new authors to come up with antagonistic IF scenarios, in
> which the PC must fight to survive. (I don't have my TADS manual with
> me... what's the equivalent way to end a TADS game?)

I tend to agree with Magnus in that the more influential fact is that
the two default ways to finish a game are winning and dying.

> Consider this: If Crowther's hobby was something other than caving,
> would the history of gaming have included so many dungeon crawls?
>
> If you're with me so far, and agree that this one decision that
> Crowther made did affect the development of computer games for a long
> time, then it's probably not too much of a stretch to suggest that
> other decisions Crowther (and others) made in the history of IF, or
> the decisions that the current generation of IF storytellers are
> making now, may have tremendous unforeseen effects.
>
> All the things that IF was good at doing in 1975 (manipulating
> objects, language games), graphic computer games are still good at
> doing. The things that IF was not good at -- the things that Crowther
> didn't implement in his original simulated universe (complex NPCs,
> dynamic plots) -- are still elusive.

Powerful points.

> When I was doing my English Ph.D., I heard quite a bit about
> post-structuralism and the "death of the author," along with praise of
> hypertext as a "writerly" genre -- a text that not only invited reader
> participation, but demanded it. While I'm not sure how productive it
> will be to apply those concepts directly to IF (it certainly wasn't
> productive to apply those concepts to hyperfiction), the goal of a
> game designer is generally to give more control to the player. In IF,
> I think a more realistic goal is improved writing, that encourages the
> player to behave in a way that will unlock the most aesthetically
> satisfying narratives that the author has chosen to code.

Yes.

> I think that this means "authorship" will, in the future, be
> democratized... shared among the members of a design team, and also
> taken on by the players in future IF. Even solitary authors will use
> software packages that reflect the experience of multiple designers,
> playtesters, and upgrades driven by user requests.

How this follows I don't see. Writing careful enough and effective
enough to shunt the player unconsciously in one direction seems least
likely to result from committee. The latter point, though, seems
obvious.



> (I haven't played "Majestic" yet... it sounds promising, but is it
> really interactive? Or just a linear story dished out via faxes,
> e-mails, fake web sites, chatbots, and phone calls?)

The latter, except in that you can interact with other humans, who will
be unhelpful, and could probably be identified by their ability to
understand what you are talking about, should you feel the desire to
find out who's part of the game and who's not.



> There's been much recent discussion on r*if on NPC interaction. Of
> course the parser has gotten much better, but the core problems in IF
> are still the same, and won't be solved by the Holy Grail of FPS and
> MMORPG designers -- an ever more accurate simulation of physics.
> Location, containment, proximity, illumination, etc. are being
> developed with ever-further complexity, permitting players of graphic
> action games to improvise solutions to problems that the designers
> didn't foresee.

Emily Short's simulationist work actually could lead to puzzles with
unintended solutions, and her conversational systems similarly could
pave the way for "neural net" NPC simulations which could display
emergent behavior; that is, sophisticated unprogrammed behavior arising
from simple programmed principles.

The problem is that, in a field tending towards leading with writing and
creating predefined stories, an unintended solution is called a bug.
Simulationist libraries are, obviously, enormously important for
simulations, but fiction is not simulation, it's narrative with cultural
realism. Neural nets and the like perhaps are more hopeful, but at low
levels of development and complexity the same objections apply; an NPC
who thinks for himself is unlikely to be helpful in plot development,
doing pernickety things like not going into the haunted house alone.



> Is there a particular style to Inform parser responses -- a style that
> influences the experience of writing such games?
>
> Is the Inform parser really more polite and gentlemanly than the
> Infocom parser? Does anybody find some of the Inform responses just a
> bit too... precious? ("Real adventurers do not use such language."
> [nudge-nudge, wink-wink.])

I can only speak for my own experiences; in my case, the answer to the
first question is a resounding yes. Even a one-room game, to my eye and
pen, would require rewriting of library messages to eliminate the by
turns flip and quaint "flavour" redolent of Curses (unsurprisingly,
since they share authors). While I certainly appreciate the original
responses, they would be inappropriate, I feel, for my writing.

You may note, though, the mild preponderance of Inform games which
correspond surprisingly in tone to the Inform default messaging; a
surprising amount of what might be termed high farce can be found in the
IF archive. Is this a result of unconsciously following the leader, or
does it perhaps result from the peculiar ability the IF writer has to
trap not merely the characters but the audience itself in ludicrous
situations, even as the protagonist?



> While the Usenet community often laments what happened when
> technologically naive AOL users began flooding the public newsgroups
> that had previously "belonged" to the technological elites, the
> Internet really emerged as a social force when it became accessible to
> a wide range of people outside the traditional group of programmers
> and researchers.
>
> If that's the case, then it's probably safe to say that new IF authors
> will probably lean very heavily on default responses. And if that's
> the case, then the default responses, may affect the future of
> interactive narrative as strongly as Crowther's decision to create an
> underground treasure hunt.
>
>
>
> Dennis G. Jerz

How very interesting.

The existence of default responses allows new authors to begin their
artistic process immediately, without carefully redesigning all their
tools to fit their personal preferences, just as the existence of MIDI
sequencers allows any bored college student to create techno music. A
committed and determined artist in both cases, though, will eventually
reach of their own the desire to redesign and customize their tools to
adapt them to the work they plan or prefer to create. It seems to me,
then, that the future-shapers, and thus the future, will follow their
own particular path regardless of the messaging they are supplied with
to start. It is interesting to contemplate the fate of interactive
nations hanging in the balance with a precious backtalking parser, but
"That's not how life works"; or so I contend, anyway.

--
William Burke, passeng...@email.com if you say so
"Many people include in their signatures contact information, and perhaps
a joke or quotation." -- Simon Fraser Go Slugs!
http://www.passengerpigeon.net (not com, not org)

Andrew Plotkin

unread,
Jan 3, 2002, 1:04:19 PM1/3/02
to
In rec.games.int-fiction Passenger Pigeon <passeng...@email.com> wrote:
> In article <792c6202.02010...@posting.google.com>,
> jer...@uwec.edu (Dennis G Jerz) wrote:

>> In Inform, the programmer signals the end of the game by setting a
>> variable named "deadflag". Hypothesis: The name of this variable
>> encourages new authors to come up with antagonistic IF scenarios, in
>> which the PC must fight to survive. (I don't have my TADS manual with
>> me... what's the equivalent way to end a TADS game?)

> I tend to agree with Magnus in that the more influential fact is that
> the two default ways to finish a game are winning and dying.

But I also think that the design of the Inform library is an *effect*
of this fact, not in any way a cause of it.

--Z

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
* Make your vote count. Get your vote counted.

Alexandre Owen Muniz

unread,
Jan 3, 2002, 1:44:18 PM1/3/02
to
jer...@uwec.edu (Dennis G Jerz) wrote in message news:<792c6202.02010...@posting.google.com>...

> Consider this: If Crowther's hobby was something other than caving,
> would the history of gaming have included so many dungeon crawls?

Yes.

First, because everyone who wrote a fantasy game would still have read
Tolkein.
Second, because an underground dungeon or cave is inherently an
excellent solution to the problem of how to keep a player on the map.

When I play a game set in an outdoor setting, I'm always trying to
wander off, and the best that a game can tell me to keep me from doing
so is to say, "Nah, the vegetation is too thick that way," or "Nah, it
doesn't look like there's anything interesting over there." Physical
barriers like cliffs can be sprinkled in, but putting physical
barriers everywhere is highly unrealistic. In a game set underground
there is a very good reason you can go only where the author wants you
to go: all other directions are solid rock! For this same reason
buildings and space stations are common in games.

I do think Crowther may have set a standard in the amount of
naturalism with which caves in games are depicted. (Even if that
standard was mostly ignored.)

I know that you've done a good bit of scholarship on Crowther and
Adventure, and it's certainly appreciated, but perhaps your proximity
to the subject is causing you to overestimate its influence.

**Owen

Adam Thornton

unread,
Jan 3, 2002, 2:01:02 PM1/3/02
to
In article <passengerpigeon-F5...@news.la.sbcglobal.net>,

Passenger Pigeon <passeng...@email.com> wrote:
>The existence of default responses allows new authors to begin their
>artistic process immediately, without carefully redesigning all their
>tools to fit their personal preferences, just as the existence of MIDI
>sequencers allows any bored college student to create techno music. A
>committed and determined artist in both cases, though, will eventually
>reach of their own the desire to redesign and customize their tools to
>adapt them to the work they plan or prefer to create. It seems to me,
>then, that the future-shapers, and thus the future, will follow their
>own particular path regardless of the messaging they are supplied with
>to start. It is interesting to contemplate the fate of interactive
>nations hanging in the balance with a precious backtalking parser, but
>"That's not how life works"; or so I contend, anyway.

However, this is precisely why any Adventure-Builder sorta system which
allows for quick-and-easy creation of IF *MUST* have the ability to rip
off the covers and start rebuilding the engine, to mix a metaphor.
Anyone sufficiently "committed and determined" must not find that the
tool puts an implacable roadblock in their way.

Adam

Andrew Plotkin

unread,
Jan 3, 2002, 2:18:30 PM1/3/02
to
In rec.arts.int-fiction Alexandre Owen Muniz <mun...@xprt.net> wrote:
> jer...@uwec.edu (Dennis G Jerz) wrote in message news:<792c6202.02010...@posting.google.com>...

>> Consider this: If Crowther's hobby was something other than caving,
>> would the history of gaming have included so many dungeon crawls?

> Yes.

> First, because everyone who wrote a fantasy game would still have read
> Tolkein.
> Second, because an underground dungeon or cave is inherently an
> excellent solution to the problem of how to keep a player on the map.

I disagree -- I think a lot of what's "obvious" about it is only long
habit.

Tolkien didn't write any dungeon crawls. The caves in _The Hobbit_ are
one dark orc lair, one dark Gollum lair (with lake), and one dark
dragon lair (with hoard). They're quite vaguely described -- there's
absolutely nothing like Crowther's detailed cave system, with its
dozens of rooms, hundreds of passages, mazes, pits, chasms, ravines,
crawls, etc.

The Mines of Moria are somewhat clearer, and they have a chasm (with
bridge), but it's one piece of a huge story -- whose primary attribute
is that it has *so many* settings. And it's still much smaller, on its
own, than even Zork 1.

You just couldn't do a dungeon crawl without including all the cave
tropes that Crowther used -- you'd run out of stuff.

Now, if Crowther hadn't been a caver, I'm not sure what other hobby
could have led to _Adventure_ being created at all. (I'm presuming
that there would have been such a game, and it would have had the
detail and richness of the Colossal Cave game -- otherwise it would
have fallen into the same obscurity as Hunt the Wumpus, and we would
have no field of IF today.)

Your point about caves restricting movement is true, but that could
have come out quite differently, while still being IF. If the
archetypical IF game were a walk in the wilderness (a la _The Hobbit_,
or most of _Alice in Wonderland_ or _The Wizard of Oz_ for that
matter), we would have paths. Perhaps a genre convention of getting
lost if you stray from the path, or being unwilling to do so, or
quickly stumbling into some annoying (but non-fatal) fate.

> I do think Crowther may have set a standard in the amount of
> naturalism with which caves in games are depicted. (Even if that
> standard was mostly ignored.)

Ignored, or badly imitated.

Eric Mayer

unread,
Jan 3, 2002, 2:36:15 PM1/3/02
to
On 3 Jan 2002 03:04:21 -0800, jer...@uwec.edu (Dennis G Jerz) wrote:

>In Inform, the programmer signals the end of the game by setting a
>variable named "deadflag". Hypothesis: The name of this variable
>encourages new authors to come up with antagonistic IF scenarios, in
>which the PC must fight to survive. (I don't have my TADS manual with
>me... what's the equivalent way to end a TADS game?)
>
>Was the name "deadflag" suitably applied because IF authors had
>already come to the realization that killing the PC was a convenient
>storytelling shortcut? Is it a way to make the game world seem
>responsive enough to recognize and respond to the player's desire to
>make a radical change to the "ideal" narrative, but since the change
>was fatal, it was short-lived, and hence required the programmer to
>invest little time in order to implement it -- at least until such
>time as we an instruct the Holodeck to improvise a story based on
>parameters that we provide.
>

I really think this is over analysis. Avoiding death is
probably the most popular plot of all time, and with good reason
seeing what an important part remaining alive plays in most of our
lives. Now if people who had decided to write games didn't often have
their PCs avoiding death, there would be an anomally worth analyzing.
I might add that this is not only true of what some might call
popular literature either. For example I've just been rereading some
Joseph Conrad and his protagonists and other characters are always
avoiding death or failing to do so. The possibility of death is just a
very compelling element in a story for both authors and readers.
Always has been and always will be, regardless of the medium involved.

Of course, artists also, from time immemorial, have tended to
imitate what's gone before. So if the best early example of a computer
adventure game features lots of ways to die then that naturally
becomes a model of what that sort of game should be for people who are
inspired and want to produce something in the same vein. But that is
hardly a revelation.
The fact that death happens to obviate the necessity to
endlessly expand the game into multiple stories depending on the
player's choices (if that is what is meant) is certainly interesting
and conveniant but, I would think, a happy coicidence given the
popularity of stories about fights to survive.

--
Eric Mayer
Web Site: <http://home.epix.net/~maywrite>

"The map is not the territory." -- Alfred Korzybski

Sean T Barrett

unread,
Jan 3, 2002, 3:51:55 PM1/3/02
to
Dennis G Jerz <jer...@uwec.edu> wrote:
[snip much material I agree with]

I note that, w.r.t. deadflag, I almost always change the
message from "You have died" to "You have lost", assuming
the game allows losing at all (most of mine do not).

I have suggested before that we should have multiple default
library message sets, since changing all of them is too much
work for very small games (e.g. minicomp games), and it
doesn't seem like it would be too much work, but nobody's
ever followed up on it. I've also had to rewrite them
all several times, enough that I've published a document
designed to help people who are changing ALL library
messages--http://nothings.org/games/if/lm.txt

With respect to cave crawls, I've also suggested in the past
that many of the properties which seem inherent to IF are
really accidental properties derived from history. The choice
of which things we simulate (rooms, containers, doors, locks)
tops my list.

>Yet the animators for Monsters Inc. worked in a complex simulated
>universe that handled such things as the motion of fabric, the effect
>of wind on hair, water absorption, etc. That meant that animators
>didn't have to animate every ripple of fabric or lock of hair.
>Monsters Inc. is, of course, a linear narrative.

Just a general-purpose correction here: computer animators do not
work in simulated universes. What you see on the screen is generally
a set; things are not happening offscreen; often characters are
rendered independently of the background and composited, rather than
being rendered in the combined scene. They had *tools* which would
run simulations of cloth and hair, but these simulations didn't
constitute a universe; the animator might attach the clothes to a
body and animate that body and even have the simulation run automatically
to animate the cloth, but there's generally no sense of a single
coherent universe; e.g. in a typical animation package the cloth and
the hair would not interact correctly.

3d computer GAMES do generally have a simulated universe, of course;
the analogy between 3d games and movie CGI maps reasonably well onto
IF and fiction written with a word processor.

SeanB

Dennis G Jerz

unread,
Jan 3, 2002, 4:42:45 PM1/3/02
to
Andrew Plotkin <erky...@eblong.com> wrote in message news:<a12aq6$43p$1...@news.panix.com>...

> In rec.arts.int-fiction Alexandre Owen Muniz <mun...@xprt.net> wrote:
> > jer...@uwec.edu (Dennis G Jerz) wrote in message news:<792c6202.02010...@posting.google.com>...
>
> >> Consider this: If Crowther's hobby was something other than caving,
> >> would the history of gaming have included so many dungeon crawls?
>
> > Yes.
>
> > First, because everyone who wrote a fantasy game would still have read
> > Tolkein.
> > Second, because an underground dungeon or cave is inherently an
> > excellent solution to the problem of how to keep a player on the map.
>
> I disagree -- I think a lot of what's "obvious" about it is only long
> habit.
>
> Tolkien didn't write any dungeon crawls. The caves in _The Hobbit_ are
> one dark orc lair, one dark Gollum lair (with lake), and one dark
> dragon lair (with hoard). They're quite vaguely described -- there's
> absolutely nothing like Crowther's detailed cave system, with its
> dozens of rooms, hundreds of passages, mazes, pits, chasms, ravines,
> crawls, etc.
>
> The Mines of Moria are somewhat clearer, and they have a chasm (with
> bridge), but it's one piece of a huge story -- whose primary attribute
> is that it has *so many* settings. And it's still much smaller, on its
> own, than even Zork 1.

If I may play devil's advocate here... consider the last three
chapters of "The Two Towers" ("The Stairs of Cirith Ungol", "Shelob's
Lair," and "The Choices of Master Samwise") include interactions with
inventory management, NPCs, and focuses on a choice of how to continue
the quest. (Spoilers for those who haven't read this book, the middle
of the Lord of the Rings trilogy.)

Sam deiciding to wear the ring, NPC interaction (of a sort -- Sam
learns plot points by overhearing 2 guards talking, etc.), having to
choose between following Frodo's body or continuing with his quest.

Dennis G. Jerz

Andrew Plotkin

unread,
Jan 3, 2002, 4:58:19 PM1/3/02
to
In rec.arts.int-fiction Dennis G Jerz <jer...@uwec.edu> wrote:
> Andrew Plotkin <erky...@eblong.com> wrote in message news:<a12aq6$43p$1...@news.panix.com>...
>>
>> Tolkien didn't write any dungeon crawls. The caves in _The Hobbit_ are
>> one dark orc lair, one dark Gollum lair (with lake), and one dark
>> dragon lair (with hoard). They're quite vaguely described -- there's
>> absolutely nothing like Crowther's detailed cave system, with its
>> dozens of rooms, hundreds of passages, mazes, pits, chasms, ravines,
>> crawls, etc.
>>
>> The Mines of Moria are somewhat clearer, and they have a chasm (with
>> bridge), but it's one piece of a huge story -- whose primary attribute
>> is that it has *so many* settings. And it's still much smaller, on its
>> own, than even Zork 1.

> If I may play devil's advocate here... consider the last three
> chapters of "The Two Towers" ("The Stairs of Cirith Ungol", "Shelob's
> Lair," and "The Choices of Master Samwise") include interactions with
> inventory management, NPCs, and focuses on a choice of how to continue
> the quest. (Spoilers for those who haven't read this book, the middle
> of the Lord of the Rings trilogy.)

I agree with your examples, in a rough way, but I don't understand how
it's a reply to what I said, or to your original contention about
Crowther and cave-crawls.

(One could cast the entire LOTR trilogy as a guess-the-verb puzzle. :)
DROP RING. "I don't know how to do that." THROW RING. "What do you
want to throw the ring into?" VOLCANO. "You see no volcano here...")

L. Ross Raszewski

unread,
Jan 3, 2002, 5:00:39 PM1/3/02
to
On 3 Jan 2002 12:26:25 GMT, Magnus Olsson <m...@df.lth.se> wrote:
>In article <792c6202.02010...@posting.google.com>,
>Dennis G Jerz <jer...@uwec.edu> wrote:
>>In Inform, the programmer signals the end of the game by setting a
>>variable named "deadflag". Hypothesis: The name of this variable
>>encourages new authors to come up with antagonistic IF scenarios, in
>>which the PC must fight to survive.
>
>I don't think the *name* of the flag is very influential compared
>to the fact that there are two default ways to end an Inform game:
>by winning or dying; the default end-of-game messages are
>"You have won" and "You have died".

Which is, in turn, because, in general, the easiest fail-state to
detect is death; if you've just reached a stopping point but not won,
there's often no reason you couldn't just go back and reach the wining
conclusion (or, if there is a reason, it may be difficult for the
program to track this)

Adam Thornton

unread,
Jan 3, 2002, 5:02:36 PM1/3/02
to
In article <GpDrA...@world.std.com>,

Sean T Barrett <buz...@TheWorld.com> wrote:
>I note that, w.r.t. deadflag, I almost always change the
>message from "You have died" to "You have lost", assuming
>the game allows losing at all (most of mine do not).

Once upon a time and place, the Death Message of Fashion was, uh,
(called, of course, with GOSUB 50000):

50000 FLASH
50010 PRINT "YOU ARE ";
50020 FOR I = 1 TO 100
50030 PRINT "DEAD! ";
50040 NEXT I
50050 PRINT
50060 RETURN

Thank goodness *that* meme didn't take.

Adam

Kevin Forchione

unread,
Jan 3, 2002, 5:26:28 PM1/3/02
to
"Adam Thornton" <ad...@fsf.net> wrote in message
news:a129pe$6g9$1...@news.fsf.net...

> However, this is precisely why any Adventure-Builder sorta system which
> allows for quick-and-easy creation of IF *MUST* have the ability to rip
> off the covers and start rebuilding the engine, to mix a metaphor.
> Anyone sufficiently "committed and determined" must not find that the
> tool puts an implacable roadblock in their way.

Welcome to TADS 3!

--Kevin


GV

unread,
Jan 3, 2002, 5:41:19 PM1/3/02
to
Adam Cadre

[...]

> >I
> You're wearing a pair of gloves.
>
> >LOOK AT MY HANDS
> [That is not an object with which you can currently interact.]
>
> (Player: "Hrm, not implemented. Okay, moving on then.")
>

> I wouldn't take this approach to *every* game [...]

What about LOOK AT SDFGSG?

Dennis G. Jerz

unread,
Jan 3, 2002, 5:49:56 PM1/3/02
to
"L. Ross Raszewski" <lrasz...@loyola.edu> wrote in message
news:a12ka7$hlu$1...@foobar.cs.jhu.edu...

Good point, but why is Inform set up so that death is so easy to detect?
I'm curious... I also wonder about stuff like why do I shut down Windows by
clicking the Start button, and why do I eject my disk on a Mac by dragging
it to the trash can?

What do you mean by "go back"? Do you mean "type undo"? Or, without erasing
the game's memory of the narrative dead-end you took, turn around and try
something else? What if those dead-ends were morally signfiicant choices
(give the Ring back to Gollum and go back to the Shire).

--
Dennis G. Jerz, Ph.D.; (715)836-2431
Dept. of English; U Wisc.-Eau Claire
419 Hibbard, Eau Claire, WI 54702
------------------------------------
Literacy Weblog: www.uwec.edu/jerzdg


Dennis G. Jerz

unread,
Jan 3, 2002, 5:44:53 PM1/3/02
to
"Magnus Olsson" <m...@df.lth.se> wrote in message
news:a11ilh$4a5$1...@news.lth.se...

> In article <792c6202.02010...@posting.google.com>,
> Dennis G Jerz <jer...@uwec.edu> wrote:
> >In Inform, the programmer signals the end of the game by setting a
> >variable named "deadflag". Hypothesis: The name of this variable
> >encourages new authors to come up with antagonistic IF scenarios, in
> >which the PC must fight to survive.
>
> I don't think the *name* of the flag is very influential compared
> to the fact that there are two default ways to end an Inform game:
> by winning or dying; the default end-of-game messages are
> "You have won" and "You have died".
>
> But scenarios where the PC must fight to survive are IMHO rather
> rare; death in IF usually seems to occur by walking into traps,
> drinking poisonous potions, falling off cliffs and similar things.

In such cases, then the "fight" isn't against any NPCs, but rather against
the limitations of the story. It's still a fight, in a kind of poetic sense,
isn't it?

--
Dennis G. Jerz, Ph.D.; (715)836-2431
Dept. of English; U Wisc.-Eau Claire
419 Hibbard, Eau Claire, WI 54702
------------------------------------
Literacy Weblog: www.uwec.edu/jerzdg

Dennis G. Jerz

unread,
Jan 3, 2002, 6:05:09 PM1/3/02
to
"Adam Cadre" <gri...@drizzle.com> wrote in message
news:a11jtp$8kk$1...@drizzle.com...

> Dennis Jerz wrote:
> > The Crowther/Woods version would occasionally say something like, "I'm
> > not allowed to tell you any more about that, so I will repeat the room
> > description." In a similar situation, the 1994 Z-code version will
> > say something like "That's not a word you need to use in the course of
> > the game."
>
> I believe that that message is reserved for when the player attempts to
> use a word listed in the name field of the current room. Words that don't
> occur in the game's dictionary at all produce the infamous "You can't see
> any such thing."

Right. That's why Adventure repeated the room description when the player
tried to interact with part of the scenery.

[Excellent examples deleted.]

>
> >EAT BANANA
> You can't see any such thing.
>
> To my mind, what has just happened in the story is that the character
> in question has decided, "Hey, a banana sure would hit the spot," looked
> around for one to eat, and sadly concluded, "Darn, it doesn't look like
> there's a banana in this room." Me, I've come to prefer to make it fairly
> explicit to the player that the command is erroneous, bracketing off the
> response as nondiegetic:
>
> >EAT BANANA
> [That is not an object with which you can currently interact.]

Good point. I seem to remember a scene in Hitchhiker's Guide, when Ford
whispers advice to the player, explaining how to look up stuff in the guide.
The narrator of Adventure frequently addresses the player directly ("With
what, your bare hands?"), which handles the problem of how to handle the
interruptions efficiently. Does the need to hide this bit of informatin
(whether the object is "not here" or "not anywyere" (as "For a Change"
phrased it) invoke the necessity to involve the narrator in the action, thus
leading to the gentle mocking of the player's efforts that characterized the
generic Infocom narrator? Few IF authors are going to be as comfortable as
Adam when it comes to making conscious choices regarding innovations, while
being fully conscious of the effect that a particular interface modification
will have on the aestheic impact of the narrative.

I found it rather maddening, for instance, to deal with "say yes" "yes"
"answer yes" and "tell NPC yes". Once I figured it out, I was more
comfortable putting a few yes/no interations in the game world, but when I
first tried it, I used the old technique of (Question? Y|N), which I found
disruptive when it was an NPC asking the question, not the game itself.

I must say that, as a player, I do appreciate knowing which word is the one
causing the problem. The response for "Put X on Y," for example, doesn't
differentiate whether there is no match for X or no match for Y. While
thinking of myself as an author, I didn't want to give too much away.

[...]

Dennis G. Jerz

unread,
Jan 3, 2002, 6:24:20 PM1/3/02
to
"joh" <j...@bigblueheron.com> wrote in message
news:87g05n8...@hammurabi.foo.bar...

> I'm sure I'm going to hit a lot of things that have already been
> discussed on this group, so if that bothers you, read no further.
>
> As a newcomer to IF, I've already considered many of the topics you
> bring up. The angle I'm proceeding from, though, is "Is it worth my
> while to get into this as an author, or am I better off spending the
> time on noninteractive short fiction?" And what that boils down to for
> me is solving this dilemma: "Is IF just a gimmick for taking a
> traditional story, chopping it up, and making it 'fun' to read, with
> the side benefit that the audience is less judgmental about some of
> the standard criteria for other prose (e.g., if the gameplay is
> original, the plot need not be)?"
>
> Re the influence of "deadflag," I think A Change in the Weather
> provides counterexamples here. You need not kill the character to end
> the story suddenly and with finality. On the other hand, once any game
> goes into an unwinnable state, you could argue that everything from
> there on out is a slow loss.

Good point. But Plotkin has shown in many ways that he's completely
comfortable making changes to the default game world, in order to achieve
some effect. Since the learning curve for playing IF is steep enough, I'm
very interested in hearing what aspects of the programming model encourage
or discourage certain kinds of stories/gameplay.

To posit another random thought... if Inform came with a default "diagnose"
verb, then perhaps more gamers would try to expand the RPG elements of Zork.
If Inform came with a spellbook simulation library, then spellcasting games
would be easier.

The original authors of Zork published an IEEE article about what they
called a Fantasy Roleplaying Simulation system... idea was that if there was
only one thing in the inventory tagged as "weapon", and the player typed,
"attack NPC," the game would automatically equip the weapon and use it.

> If not that, then something similar, just because it's easiest to
> implement either a constrained space or a relatively empty open
> space.

Good point. I noticed in the early graphics games that lots of games
continued to be set in dark places, so that the graphics designers were
excused from rendering anything more than 3 or 4 "tiles" away. "Lands of
Lores" had a pleasant variation on this, by setting a few scenes in
mist-shrouded areas, so that distant objects faded to white.

> Dennis> [...] The things that IF was not


> Dennis> good at -- the things that Crowther didn't implement in
> Dennis> his original simulated universe (complex NPCs, dynamic
> Dennis> plots) -- are still elusive.
>
> Isn't that just because they're harder? I mean, as far as I've seen,
> nobody's programmed a really convincing human character, in IF or any
> other branch of programming. And multiple plots mean a multiplication
> of effort. It seems logical in most cases to spend that effort on
> making one plot more satisfying.

Could be. Eliza was a simple program that offered some fairly complex
interaction possibilities, but I suppose in the mid 60s there just wasn't
the critical mass of players who decided they wanted to improve the model.

But IF isn't all about plot.

> Dennis> A company that puts resources into graphic and sound
> Dennis> design can get richer game play that can be generated
> Dennis> dynamically. Plot is finite, and many a fine game has
> Dennis> gotten poor reviews due to low "replayability".
>
> Here's the crux. If good prose is an important part of the IF
> experience, you can't really have "dynamic" plots. You can only have
> multiple plots. Unless you think Eliza generates good prose.

What do you mean, unless I think Eliza generates good prose ? :)

>
> Dennis> Yet the animators for Monsters Inc. worked in a complex
> Dennis> simulated universe that handled such things as the motion
> Dennis> of fabric, the effect of wind on hair, water absorption,
> Dennis> etc....
>
> And they flubbed it just as bad as most big-budget studios that work
> in a nonsimulated universe that handles the motion of fabric,

Good point! I never thought of it quite that way before.


[...]


> That's the other crux :) In a "game" we want control. In a "story," we
> have to be led. In IF, I think we want a convincing illusion of
> control, while still being led. At least, I think that's what I want.

The willing suspension of disbelief. Coleridge. Right on.

[...]

>
> I'm running out of time to get ready for work, so let me just agree
> that the standard responses from Inform are totally
> distracting. Especially when you're just starting a new
> game. Everything is fresh and new, until you try something that
> doesn't work. Suddenly you feel like you're playing Curses again.
>

Hmm. Has anybody put much thought into advice for how to systematically
revise the default responses for tone? I'm sure the DM4 chapter on
non-English would provide all the technical details (which really amounts to
editing a file, though that could cause problems in the event of a library
update). But what about other languages besides Inform? How extensible are
they?

Dennis G. Jerz

unread,
Jan 3, 2002, 6:39:54 PM1/3/02
to
"J. D. Berry" <ber...@earthlink.net> wrote in message
news:ff102855.02010...@posting.google.com...

I imagine also that, like Scott Adams, whose games are so textually sparse
that they're hardly ever mentioned in literary studies of IF, despite the
new ground they broke by expanding IF into new genres, the original games
may have had texutal limits to deal with. A significant number of people
were probably also playing games on printer terminals, which probably led to
the "brief" command... but in Inform games, "brief" is the default mode.
Personally, I think that people who would rather get less information should
have to ask to turn off the stuff they don't want. A new player (and a new
author) is probably not going to think of something like setting lookmode.
(Again, I'm being specific to Inform.)

>
> I think we're at the stage where authors are going
> the less ideal narrative route, rather than the death route. Look
> at Cadre, Short and Fischer games, for example. Sure, sometimes their
> non-deaths are the "death" equivalent, but they always fit the situation
> and story.

We do have good models. Absolutely.

[Ref to Costikyan deleted]

> Taking from that, 1) there must be competition on some level, and 2)
> there must be decision making. And it is here, I feel, where there
> are shortcut temptations.
>
> The danger of #1, above, is "OK, it's you the player versus me
> the author." In #2, "OK, make a wrong decision and die." These
> satisfy the GAME conditions, but they (may) seriously neglect the
> STORY.

Even if good authors manage to avoid obviously awkward compomises to handle
the above ploarities, those authors are going to be affected by the
conditions that shaped their game worlds. For instance, the PC in Galatea
comments on the title character's limited conversation ability. That's part
of the storyline.

> This again returns to the "NEW" author discussion. There's just
> so much an IF author has to do. To a new author, way too much.
> Integrating competition is complex. Incorporating
> subtle decision-making is complex. Don't forget original storylines
> and depth of world, too. Oh, and cool NPCs, please.
>
> Thus, even if DEADFLAG were not the syntax, authors (especially
> new ones) would still take this route. That's why I feel that
> the "deadflag effect" is only a seed, not a full encouragement.
> The seed will grow, but it's up to the author to shape its
> future. There can and should be pruning.

I just sort of picked deadflag as a starting point, but I'm sure there are
others. Which ones do you see?

[...]

True, but you'd also need to duplicate the way the characters do a
double-take, or the timing of the physical comedy.

[...]


> > Gamers
> > often responded with hostility to that scripted plot point, since
> > elsewhere in the game it was possible for stealthy characters to
> > disable alarms.
> >
>
> Again, this is where human creativity and TIME come in to play
> With both of those elements, that alarm situation could have been
> alleviated. There's always the subjective factor, though. There's
> going to be SOMEONE that doesn't like it, however you do it.

True.

[...]

Thanks! It may take me a while to reply to everyone's posts... but I'll
certainly take to heart all the responses in this thread.


Dennis G. Jerz

unread,
Jan 3, 2002, 6:45:02 PM1/3/02
to
"Magnus Olsson" <m...@df.lth.se> wrote in message
news:a122h9$811$1...@news.lth.se...

> In article <3c348979$0$79557$3c09...@news.plethora.net>,
> Peter Seebach <se...@plethora.net> wrote:
> >In article <792c6202.02010...@posting.google.com>,
> >Dennis G Jerz <jer...@uwec.edu> wrote:
> >>In Inform, the programmer signals the end of the game by setting a
> >>variable named "deadflag". Hypothesis: [...]

> >I disagree - it's just that, if you aren't dead, why shouldn't you be
allowed
> >to keep playing?
>
> Because the story is over? Or because you've lost the game?
>
> I understand your point, but I think this depends on where the game is
> on the simulation/story scale, i.e. how much control the author wants
> over the player.
>

Bingo. After I gave the last red page to the guy in the red book, and then
restored and gave the last blue page to the guy in the blue book, Atrus
suggested that I could wander around in the Myst world if I wanted to. Why?
The story was over.

But in Daggerfall, I think I only tried following the over arching plot for
a few game-months, and then veered off on my own quest. Very different
modes of gaming.

In something like Photopia, where the game environment is secondary to the
story, the environment changes radically in order to account for plot
developments. As Mr. Wu says in "Herbie The Love Bug," "When story comes
to last page, close the book."

(Unless you've got the Star Trek computer to synthesize stories for you.)
DGJ


Dennis G. Jerz

unread,
Jan 3, 2002, 6:58:09 PM1/3/02
to
"Passenger Pigeon" <passeng...@email.com> wrote in message
news:passengerpigeon-F5...@news.la.sbcglobal.net...

> In article <792c6202.02010...@posting.google.com>,
> jer...@uwec.edu (Dennis G Jerz) wrote:
>
> > In Inform, the programmer signals the end of the game by setting a
> > variable named "deadflag". Hypothesis: The name of this variable
> > encourages new authors to come up with antagonistic IF scenarios, in
> > which the PC must fight to survive. (I don't have my TADS manual with
> > me... what's the equivalent way to end a TADS game?)
>
> I tend to agree with Magnus in that the more influential fact is that
> the two default ways to finish a game are winning and dying.
>

OK... I suppose my point would still be the same if the flag were
"liveflag", but what if it were "story_done" or "plot_advancing"? Would
that influence the way people planned and created IF?


[...]

> > I think that this means "authorship" will, in the future, be
> > democratized... shared among the members of a design team, and also
> > taken on by the players in future IF. Even solitary authors will use
> > software packages that reflect the experience of multiple designers,
> > playtesters, and upgrades driven by user requests.
>
> How this follows I don't see. Writing careful enough and effective
> enough to shunt the player unconsciously in one direction seems least
> likely to result from committee. The latter point, though, seems
> obvious.

A committe might be more open to extensive beta-testing... if nobody "owns"
a story, perhaps a committee would be more tuned in to what the beta-testers
want to do, and giving it to them. That's exactly the reason why it seems
that the real innovation in interactive storytelling is happening with the
dedicated amateurs, rather than the deep-pocketed corporations.

[...]

> Emily Short's simulationist work actually could lead to puzzles with
> unintended solutions, and her conversational systems similarly could
> pave the way for "neural net" NPC simulations which could display
> emergent behavior; that is, sophisticated unprogrammed behavior arising
> from simple programmed principles.
>
> The problem is that, in a field tending towards leading with writing and
> creating predefined stories, an unintended solution is called a bug.

Not quite... it's a bug if there's no way to adjust the storyline to account
for that unintended solution. Careful plotting and abstracted plot elements
would help.

> Simulationist libraries are, obviously, enormously important for
> simulations, but fiction is not simulation, it's narrative with cultural
> realism.

True. But IF is neither simulation nor fiction. What's right for IF, as we
know it now, and as we imagine it could be in the future?

> Neural nets and the like perhaps are more hopeful, but at low
> levels of development and complexity the same objections apply; an NPC
> who thinks for himself is unlikely to be helpful in plot development,
> doing pernickety things like not going into the haunted house alone.

True, but it would open up new realms of interactive possibilities. Think
of a RPG-style campaign game, where, as your exploits grow, stories of your
deeds start getting told in taverns and around campfires, and your repuation
spreads thus. In such a game, interacting with story-spinning NPCs might be
quite rewarding. (Think of it... if AI advances to the point where it's
possible for the computer to improvise story elements, why shouldnt NPCs
have access to the same abilities?)

[...]

>
> You may note, though, the mild preponderance of Inform games which
> correspond surprisingly in tone to the Inform default messaging; a
> surprising amount of what might be termed high farce can be found in the
> IF archive. Is this a result of unconsciously following the leader, or
> does it perhaps result from the peculiar ability the IF writer has to
> trap not merely the characters but the audience itself in ludicrous
> situations, even as the protagonist?

Yup, that's pretty much my point. I suppose also, it's easier to create a
goofy world, like the NPCs of Starship Titanic, which only rarely
"understand" you, but frequently respond with in-character, humorous
responses.

[...]

> The existence of default responses allows new authors to begin their
> artistic process immediately, without carefully redesigning all their
> tools to fit their personal preferences, just as the existence of MIDI
> sequencers allows any bored college student to create techno music.

Good analogy. Karaoke IF, anyone?

[...]


> It is interesting to contemplate the fate of interactive
> nations hanging in the balance with a precious backtalking parser, but
> "That's not how life works"; or so I contend, anyway.

Interactive nations? I wasn't thinking about "I Have No Mouth and I Must
Scream" here. :)


>
> --
> William Burke, passeng...@email.com if you say so
> "Many people include in their signatures contact information, and perhaps
> a joke or quotation." -- Simon Fraser Go Slugs!
> http://www.passengerpigeon.net (not com, not org)

--

Dennis G. Jerz

unread,
Jan 3, 2002, 7:00:30 PM1/3/02
to
"Kevin Forchione" <Ke...@lysseus.com> wrote in message
news:oU4Z7.2039$MK.21326@rwcrnsc54...

Kevin, or someone, please expand on this! I have some idea of what you
mean, but I'd like to see someone else's take on TADS, since mine will be
too colored by the first IF language I learned..

Dennis G. Jerz

unread,
Jan 3, 2002, 7:04:12 PM1/3/02
to
Zarf writes:
> I agree with your examples, in a rough way, but I don't understand how
> it's a reply to what I said, or to your original contention about
> Crowther and cave-crawls.
>
> --Z

Er... sorry for being vague. I meant that those chapters of LOTR come even
closer to being a dungeon crawl. I also happened to have just read those 3
chapters this weekend.

Dennis G. Jerz

unread,
Jan 3, 2002, 7:15:28 PM1/3/02
to
"Eric Mayer" <emay...@epix.net> wrote in message
news:3c34b0e9...@newsserver.epix.net...
[Overanalysis deleted]


> I really think this is over analysis.

Well, it generated some good discussion, which I think makes it worthwhile.


> Avoiding death is
> probably the most popular plot of all time, and with good reason
> seeing what an important part remaining alive plays in most of our
> lives.

I'm reminded of the Brooke Shields quote: "If smoking kills you,then you've
lost a very important part of your life." :)

Sure, survival plays an important part in adventure stories, and war epics
and the like... but I don't think that the Brontes or Jane Austen felt that
day-to-day survival was all that central to the stories they were writing
about.


> Now if people who had decided to write games didn't often have
> their PCs avoiding death, there would be an anomally worth analyzing.
> I might add that this is not only true of what some might call
> popular literature either. For example I've just been rereading some
> Joseph Conrad and his protagonists and other characters are always
> avoiding death or failing to do so. The possibility of death is just a
> very compelling element in a story for both authors and readers.
> Always has been and always will be, regardless of the medium involved.

Of course, but see Bruntiere's definitions of literary conflict:
a.. the individual vs. fatality (that is, a fight for survival)
b.. the individual vs. social law (justice, morality, etc.)
c.. the individual vs. another person
d.. the individual vs. himself
e.. the individual vs. "the ambitions, the interests, the prejudices, the
folly, the malevolence of those who surround him"
http://www.uwec.edu/jerzdg/ORR/handouts/Style/crisis-vs-conflict.htm

>
> Of course, artists also, from time immemorial, have tended to
> imitate what's gone before. So if the best early example of a computer
> adventure game features lots of ways to die then that naturally
> becomes a model of what that sort of game should be for people who are
> inspired and want to produce something in the same vein. But that is
> hardly a revelation.

True, but by what definition is it "best"? My point is that early
computer games featured survival and exploration, which just happened to be
important themes in Crowther's gaming and caving life. I'm not so much
announcing the "deadflag" hypothesis for its own sake, but rather hoping to
see what people think of it.

Thanks, Eric.

Dennis G. Jerz

unread,
Jan 3, 2002, 7:33:04 PM1/3/02
to
"Sean T Barrett" <buz...@TheWorld.com> wrote in message
news:GpDrA...@world.std.com...

> I have suggested before that we should have multiple default
> library message sets, since changing all of them is too much
> work for very small games (e.g. minicomp games), and it
> doesn't seem like it would be too much work, but nobody's
> ever followed up on it. I've also had to rewrite them
> all several times, enough that I've published a document
> designed to help people who are changing ALL library
> messages--http://nothings.org/games/if/lm.txt

Ah! I had thought there was such a document out there. Will check it out
tomorrow.


--


Dennis G. Jerz, Ph.D.; (715)836-2431
Dept. of English; U Wisc.-Eau Claire
419 Hibbard, Eau Claire, WI 54702
------------------------------------
Literacy Weblog: www.uwec.edu/jerzdg
>

L. Ross Raszewski

unread,
Jan 3, 2002, 7:46:48 PM1/3/02
to
On Thu, 3 Jan 2002 16:49:56 -0600, Dennis G. Jerz <Jer...@uwec.edu> wrote:
>"L. Ross Raszewski" <lrasz...@loyola.edu> wrote in message
>news:a12ka7$hlu$1...@foobar.cs.jhu.edu...
>> On 3 Jan 2002 12:26:25 GMT, Magnus Olsson <m...@df.lth.se> wrote:
>> >I don't think the *name* of the flag is very influential compared
>> >to the fact that there are two default ways to end an Inform game:
>> >by winning or dying; the default end-of-game messages are
>> >"You have won" and "You have died".
>>
>> Which is, in turn, because, in general, the easiest fail-state to
>> detect is death; if you've just reached a stopping point but not won,
>> there's often no reason you couldn't just go back and reach the wining
>> conclusion (or, if there is a reason, it may be difficult for the
>> program to track this)
>
>Good point, but why is Inform set up so that death is so easy to detect?
>I'm curious... I also wonder about stuff like why do I shut down Windows by
>clicking the Start button, and why do I eject my disk on a Mac by dragging
>it to the trash can?

Because death is a simple binary state. YOu are either dead or you
aren't. It is probably impossible to determine with total accuracy
"you can't win from here", in the general case. It's not a matter of
"Death is the default way to end the game in a nonwin-condition", but
"As long as the PC isn't dead, there might be a chance for him to pull
this off".

>
>What do you mean by "go back"? Do you mean "type undo"? Or, without erasing
>the game's memory of the narrative dead-end you took, turn around and try
>something else? What if those dead-ends were morally signfiicant choices
>(give the Ring back to Gollum and go back to the Shire).

I mean, backtrack and try something else. Suppose the player has missed
a criticial event. Perhaps this means the game is over, but there's
no generic way to know this; the author may know that in *this
particular instance*, the game is now unwinnable, and the player has
lost, but it's possible that there might be another way around it. If
the player has died, however, the system can be sure (barring
reincarnation) that the player really has lost the game.

Kevin Forchione

unread,
Jan 3, 2002, 8:09:06 PM1/3/02
to
"Dennis G. Jerz" <Jer...@uwec.edu> wrote in message
news:a12mt5$mbk$1...@wiscnews.wiscnet.net...

> "Magnus Olsson" <m...@df.lth.se> wrote in message
> news:a11ilh$4a5$1...@news.lth.se...
> > In article <792c6202.02010...@posting.google.com>,
> > Dennis G Jerz <jer...@uwec.edu> wrote:
> > >In Inform, the programmer signals the end of the game by setting a
> > >variable named "deadflag". Hypothesis: The name of this variable
> > >encourages new authors to come up with antagonistic IF scenarios, in
> > >which the PC must fight to survive.
> >
> > I don't think the *name* of the flag is very influential compared
> > to the fact that there are two default ways to end an Inform game:
> > by winning or dying; the default end-of-game messages are
> > "You have won" and "You have died".
> >
> > But scenarios where the PC must fight to survive are IMHO rather
> > rare; death in IF usually seems to occur by walking into traps,
> > drinking poisonous potions, falling off cliffs and similar things.
>
> In such cases, then the "fight" isn't against any NPCs, but rather against
> the limitations of the story. It's still a fight, in a kind of poetic
sense,
> isn't it?

Conflict is probably more conotatively accurate than fight.

However, it seems more appropriate to me to view "death" as another "reward"
in the process of solving a puzzle. The death of an actor may be a narrative
device, but the death of the PC is generally not viewed that way, but in
conjunction with failing to accomplish a given task.

--Kevin


Peter Seebach

unread,
Jan 3, 2002, 11:35:53 PM1/3/02
to
In article <a12n6k$mc2$1...@wiscnews.wiscnet.net>,

Dennis G. Jerz <Jer...@uwec.edu> wrote:
>Good point, but why is Inform set up so that death is so easy to detect?

Because "the game has ended" is a very useful state to have the game
intrinsically aware of.

-s
--
Copyright 2001, all wrongs reversed. Peter Seebach / se...@plethora.net
$ chmod a+x /bin/laden Please do not feed or harbor the terrorists.
C/Unix wizard, Pro-commerce radical, Spam fighter. Boycott Spamazon!
Consulting, computers, web hosting, and shell access: http://www.plethora.net/

Adam Cadre

unread,
Jan 4, 2002, 1:39:26 AM1/4/02
to
> What about LOOK AT SDFGSG?

That is most definitely not an object with which you can currently
interact. In that it doesn't exist and all. But the reason I like
this phrasing is that it doesn't *specify* that the object doesn't
exist, and so is equally applicable to unimplemented scenery. So
if the player tries >TOUCH MY UVULA, you're basically saying, okay,
without getting into the question of whether you've got one or not,
you're not allowed to play with it, so try something else. And
similarly, without getting into the question of whether there's
a sdfgsg around or not, there's not one that you're allowed to
do anything with, and that's all you really need to know.

-----
Adam Cadre, Brooklyn, NY
http://adamcadre.ac

Watch out, Amazon! http://adamcadre.ac/content/books-for-sale.html

Magnus Olsson

unread,
Jan 4, 2002, 4:48:17 AM1/4/02
to
In article <a12n6k$mc2$1...@wiscnews.wiscnet.net>,
Dennis G. Jerz <Jer...@uwec.edu> wrote:
>"L. Ross Raszewski" <lrasz...@loyola.edu> wrote in message
>news:a12ka7$hlu$1...@foobar.cs.jhu.edu...
>> Which is, in turn, because, in general, the easiest fail-state to
>> detect is death; if you've just reached a stopping point but not won,
>> there's often no reason you couldn't just go back and reach the wining
>> conclusion (or, if there is a reason, it may be difficult for the
>> program to track this)
>
>Good point, but why is Inform set up so that death is so easy to detect?

With all respect, Dennis, but I think you're putting the cart before
the horse here.

The purpose of deadflag in Inform isn't really to detect death, but
to detect the end of the game; the logic is that if deadflag is
different from zero at the end of the game, then the game has ended,
and the value of deadflag indicates which end-of-game message should
be displayed.

Now, Graham chose to implement only two end-of-game messages,
corresponding to a win or a PC death. But I'd say that there's nothing
in the design of Inform to make "death easy to detect". What there is
is an assumption that the most common way of losing the game would be
death.

Why? I suppose you'd have to ask Graham, but the convention was surely
firmly entrenched long before Inform and goes all the way back to
ADVENT. I suppose one way of reasoning would be that only the PC's dying
would put a definite end to the game; even if the player has put the game
into a completely unwinnable state, as long as the PC is alive the game
could just as well allow him to thrash about for a while.

>What if those dead-ends were morally signfiicant choices
>(give the Ring back to Gollum and go back to the Shire).

I think this is a good example: in a (hypothetical) LOTR game, where
you play Frodo, should giving the ring to Gollum, or Boromir, or
Saruman for that matter, be implemented, and if it is, should doing
so end the game (since Frodo will probably be able to do nothing of
consequence after that)?

But it must be noted that Inform does not prohibit or even discourage
this; it just requires the game author to read the DM a little to find
out how to replace the "You have died" message with, say, "The fate of
the world is out of your hands."

But you have a point, albeit a rather weak one. I suppose that if I
had been able to go back in time and influence Graham's design
decision, I would have asked him to have at least three default
end-of-game messages, viz. "You have won"/"You have died"/"You have
lost"; or perhaps just one: "Game over".

Richard Bos

unread,
Jan 4, 2002, 4:27:13 AM1/4/02
to
"GV" <gosta_va...@hotmail.com> wrote:

> Adam Cadre

Well, from a computer's limited world-knowledge, there's very little
difference between a sdfgsg and an earlobe.

Richard

Richard Bos

unread,
Jan 4, 2002, 4:27:14 AM1/4/02
to
mun...@xprt.net (Alexandre Owen Muniz) wrote:

> jer...@uwec.edu (Dennis G Jerz) wrote in message news:<792c6202.02010...@posting.google.com>...
>
> > Consider this: If Crowther's hobby was something other than caving,
> > would the history of gaming have included so many dungeon crawls?
>
> Yes.
>
> First, because everyone who wrote a fantasy game would still have read
> Tolkein.

Nitpick: nobody reads Tolkein. A lot of people have read Tolkien,
though, and enjoyed his books thoroughly.

Yes, it's just a typo, but it bugs me that so many people can't even
spell _his_ name correctly.

Richard

Magnus Olsson

unread,
Jan 4, 2002, 6:52:25 AM1/4/02
to
In article <a13tp1$l9j$1...@news.lth.se>, Magnus Olsson <m...@df.lth.se> wrote:
>In article <a12n6k$mc2$1...@wiscnews.wiscnet.net>,
>Dennis G. Jerz <Jer...@uwec.edu> wrote:
>>"L. Ross Raszewski" <lrasz...@loyola.edu> wrote in message
>>news:a12ka7$hlu$1...@foobar.cs.jhu.edu...
>>> Which is, in turn, because, in general, the easiest fail-state to
>>> detect is death; if you've just reached a stopping point but not won,
>>> there's often no reason you couldn't just go back and reach the wining
>>> conclusion (or, if there is a reason, it may be difficult for the
>>> program to track this)
>>
>>Good point, but why is Inform set up so that death is so easy to detect?
>
>With all respect, Dennis, but I think you're putting the cart before
>the horse here.
>
>The purpose of deadflag in Inform isn't really to detect death, but
>to detect the end of the game;

(snip)

Hmm. It seems I got my logic confused a bit here; the rest of my
post discusses different things from what I'm responding to.

To reply directly to what Ross and Dennis wrote, rather than to
my confused impression fo fwhat they were discussing :-), it's
not Inform that's set up to easily detect death; it's just that
the death of the PC is normally a certain show-stopper.

Of course, there are exceptions; interestingly enough, both ADVENT
and Zork I offer resurrection of dead PC's, so PC death does not
end the game directly. In the case of ADVENT, I think this may have
something to do with the fact that mainframe version of the game
enforces a time delay before you can restart a game, so offering the
player a chance to continue after the death of the PC would be a
nice gesture towards the player.

Magnus Olsson

unread,
Jan 4, 2002, 7:04:51 AM1/4/02
to
In article <a12mt5$mbk$1...@wiscnews.wiscnet.net>,

Dennis G. Jerz <Jer...@uwec.edu> wrote:
>"Magnus Olsson" <m...@df.lth.se> wrote in message
>news:a11ilh$4a5$1...@news.lth.se...
>> In article <792c6202.02010...@posting.google.com>,
>> Dennis G Jerz <jer...@uwec.edu> wrote:
>> >In Inform, the programmer signals the end of the game by setting a
>> >variable named "deadflag". Hypothesis: The name of this variable
>> >encourages new authors to come up with antagonistic IF scenarios, in
>> >which the PC must fight to survive.
>>
>> I don't think the *name* of the flag is very influential compared
>> to the fact that there are two default ways to end an Inform game:
>> by winning or dying; the default end-of-game messages are
>> "You have won" and "You have died".
>>
>> But scenarios where the PC must fight to survive are IMHO rather
>> rare; death in IF usually seems to occur by walking into traps,
>> drinking poisonous potions, falling off cliffs and similar things.
>
>In such cases, then the "fight" isn't against any NPCs, but rather against
>the limitations of the story. It's still a fight, in a kind of poetic sense,
>isn't it?

Yes; but once we're over on the metaphorical plane, the actual death
of the PC isn't as important anymore - it's more a matter of winning
or losing.

And here I think you have a point: Inform makes an assumption that
a piece of IF is a game which can be won or lost, and where the
player by default is assigned a score. Come to think of it, I think
the notion of score is at least as influential as the notion of death
in making authors think in game-like terms.

The gme aspects of IF have been criticized before on this newsgroup,
but then the discussion has mostly been on the micro-level, regarding
puzzles and so on. But I think it's more important what's going on on
the macro-level: is the entire work seen as a game that can be won
or lost, or as a work of art to be experienced?

Of course, there are many examples of IF that are works of art at the
same time as having the game nature. But I think the idea that a piece
of IF should have a score and a winning ending may have made authors
more reluctant to write, say, the IF Art Show type of works.

Magnus Olsson

unread,
Jan 4, 2002, 7:32:42 AM1/4/02
to
In article <a12r6h$bmg$1...@wiscnews.wiscnet.net>,

Dennis G. Jerz <Jer...@uwec.edu> wrote:
>OK... I suppose my point would still be the same if the flag were
>"liveflag", but what if it were "story_done" or "plot_advancing"? Would
>that influence the way people planned and created IF?

In this case I don't think so; the actual name of the variable carries
very little persuasive power compared to the actual default messages.

However, I've seen cases where the name of a variable or a system
command was very suggestive of what it did, in a way that didn't
really map to reality. For example, suppose that, rather than writing

deadflag = 1;

you had to write

KillPlayer(1);

to end the game. Would that give authors the idea that the only way
to end the game would be to kill the PC? Perhaps, though I doubt it.

Here's how my brain works in cases like this: I'm influenced by the
*behaviour* of the Library - if it had been very hard to change the
"You have died" message, I might be more inclined to include death in
my games.

And it is, of course, essential that the Library *allow* default
behaviours to be changed, or otherwise we might have situations like
in GAGS (IIRC) where all NPC's would develop unnaturally large teeth
and bite you, or, less drastically, games where all doors *must* have
a lock and a key.

I'm also influenced by the way the DM, and other authors, discuss
things. For example, what the Inform library calls doors really don't
have ot be doors, but they can be any kind of obstacle that have to be
traversed to move from one location to another (the bridge in "Troll
Bridge" is a door from a programming point of view). If the DM talked
exclusively about (Inform) doors as (real-world) doors, this would
perhaps limit my games (the DM doesn't, of course).

But the actual names of variables and so on matter very little,
because once I get to the stage of actually doing something to
deadflag, I will have read about it in the DM and would know that it
is possible to change the default behaviour of the game.

To summarize, terminology matters a lot, but documentation overrides
terminology.

What matters much, much more are genre conventions, such as "every game
I've played had a score and endings where the PC died".

Magnus Olsson

unread,
Jan 4, 2002, 7:44:27 AM1/4/02
to
In article <a12aq6$43p$1...@news.panix.com>,
Andrew Plotkin <erky...@eblong.com> wrote:

>In rec.arts.int-fiction Alexandre Owen Muniz <mun...@xprt.net> wrote:
>> jer...@uwec.edu (Dennis G Jerz) wrote in message
>news:<792c6202.02010...@posting.google.com>...
>
>>> Consider this: If Crowther's hobby was something other than caving,
>>> would the history of gaming have included so many dungeon crawls?
>
>> Yes.
>
>> First, because everyone who wrote a fantasy game would still have read
>> Tolkein.
>> Second, because an underground dungeon or cave is inherently an
>> excellent solution to the problem of how to keep a player on the map.
>
>I disagree -- I think a lot of what's "obvious" about it is only long
>habit.
>
>Tolkien didn't write any dungeon crawls. The caves in _The Hobbit_ are
>one dark orc lair, one dark Gollum lair (with lake), and one dark
>dragon lair (with hoard). They're quite vaguely described -- there's
>absolutely nothing like Crowther's detailed cave system, with its
>dozens of rooms, hundreds of passages, mazes, pits, chasms, ravines,
>crawls, etc.

(snip discussion of other caves in Tolkien).

I don't think the dungeon crawls entered IF directly from Tolkien,
or any other literary source, but rather from gaming, most notably
D&D. Of course, ADVENT was an important precedent here, but if dungeon
crawls hadn't been popular elsewhere we might have seen a few dungeon
crawl works of IF (ADVENT, Zork, a few more) and then authors would
have turned to other genres.

This of course leads to the question where the dungeon crawls in
D&D and other games came from.

D&D was inspired by Tolkien (most notably in the concepts of races)
and Jack Vance (magic, of which Tolkien has very little) but perhaps
mostly by Fritz Leiber (the notion of adventuring party, the thief
class). None of these authors have written very much about dungeon
crawls.

Perhaps it's just that dungeon crawls make good game settings, and
that a number of successful games used it at the time IF took off.

Dan Schmidt

unread,
Jan 4, 2002, 9:33:28 AM1/4/02
to
"Dennis G. Jerz" <Jer...@uwec.edu> writes:

| "joh" <j...@bigblueheron.com> wrote in message
| news:87g05n8...@hammurabi.foo.bar...
|

| > I'm running out of time to get ready for work, so let me just agree
| > that the standard responses from Inform are totally
| > distracting. Especially when you're just starting a new
| > game. Everything is fresh and new, until you try something that
| > doesn't work. Suddenly you feel like you're playing Curses again.
| >
|
| Hmm. Has anybody put much thought into advice for how to
| systematically revise the default responses for tone? I'm sure the
| DM4 chapter on non-English would provide all the technical details
| (which really amounts to editing a file, though that could cause
| problems in the event of a library update). But what about other
| languages besides Inform? How extensible are they?

I do know that TADS 3 is specifically addressing the 'most games tend
to have the same tone of default responses' issue and has pluggable-in
sets of library messages.

You can fairly systematically revise the default responses in Inform -
almost all the text is in one place - but it's kind of a pain, and
it's easy to miss things, as I did in For A Change.

--
http://www.dfan.org

Daryl McCullough

unread,
Jan 4, 2002, 10:09:01 AM1/4/02
to
gri...@drizzle.com (Adam Cadre?) says...

>Me, I've come to prefer to make it fairly
>explicit to the player that the command is erroneous, bracketing off the
>response as nondiegetic:
>
>>EAT BANANA

>[That is not an object with which you can currently interact.]

I looked up the word "diegetic" with the online dictionary
"dictionary.com". It didn't give me a definition, but it did
helpfully suggest the 10 most popular websites.

For those not up on film terminology (it seems to be mostly
used in discussing films, although it can apply to books and
IF as well)

diegetic sound : sound in a movie that comes from (or is meant to
come from) a source internal to the story on screen. So the sounds
of people talking, the sounds of gunfire, etc. are all diegetic
sounds (from the Greek word "diegesis" meaning story, or something
like that).

non-diegetic sound : sound in a movie that comes from a source
external to the story. A voice-over narrator, background music,
etc.

>>I
>You're wearing a pair of gloves.
>
>>LOOK AT MY HANDS
>[That is not an object with which you can currently interact.]

One problem with this is response is that it may be misleading
in a case where the object *can* be interacted with, but through
a different synonym. Of course, the author should try to anticipate
all synonyms that the player would be likely to try, but the players
will always find another one.

--
Daryl McCullough
CoGenTex, Inc.
Ithaca, NY

Daryl McCullough

unread,
Jan 4, 2002, 10:11:24 AM1/4/02
to
Dennis says...

>Good point, but why is Inform set up so that death is so easy to detect?

I don't think it is so much that it is easy to detect, but that
it requires special treatment (restarting the game).

Matthew Russotto

unread,
Jan 4, 2002, 11:25:04 AM1/4/02
to
In article <a14gi...@drn.newsguy.com>,

Daryl McCullough <da...@cogentex.com> wrote:
>
>For those not up on film terminology (it seems to be mostly
>used in discussing films, although it can apply to books and
>IF as well)
>
> diegetic sound : sound in a movie that comes from (or is meant to
> come from) a source internal to the story on screen. So the sounds
> of people talking, the sounds of gunfire, etc. are all diegetic
> sounds (from the Greek word "diegesis" meaning story, or something
> like that).
>
> non-diegetic sound : sound in a movie that comes from a source
> external to the story. A voice-over narrator, background music,
> etc.

Though some movies will screw with you by having background music that
is later revealed to be audible to the characters.

--
Matthew T. Russotto mrus...@speakeasy.net
=====
Dmitry is free, but the DMCA survives. DMCA delenda est!
"Extremism in defense of liberty is no vice, and moderation in pursuit
of justice is no virtue."

Peter Seebach

unread,
Jan 4, 2002, 12:01:57 PM1/4/02
to
In article <u3blr0h...@corp.supernews.com>,

Matthew Russotto <russ...@wanda.pond.com> wrote:
>Though some movies will screw with you by having background music that
>is later revealed to be audible to the characters.

I believe there was a very good example of this in _So I Married an Ax
Murderer_.

Peter Seebach

unread,
Jan 4, 2002, 12:05:28 PM1/4/02
to
In article <a145p3$n69$1...@news.lth.se>, Magnus Olsson <m...@df.lth.se> wrote:
>Of course, there are many examples of IF that are works of art at the
>same time as having the game nature. But I think the idea that a piece
>of IF should have a score and a winning ending may have made authors
>more reluctant to write, say, the IF Art Show type of works.

I dunno about that. I do know that it took me a while to realize that I
really wanted to have a separate score for "interesting tidbits" from the
one for "winning the game" in my game, although I haven't implemented it yet.

David Thornley

unread,
Jan 4, 2002, 12:40:32 PM1/4/02
to
In article <a1483b$nl3$2...@news.lth.se>, Magnus Olsson <m...@df.lth.se> wrote:
>In article <a12aq6$43p$1...@news.panix.com>,
>Andrew Plotkin <erky...@eblong.com> wrote:
>>In rec.arts.int-fiction Alexandre Owen Muniz <mun...@xprt.net> wrote:
>>> jer...@uwec.edu (Dennis G Jerz) wrote in message
>>news:<792c6202.02010...@posting.google.com>...
>>
>>>> Consider this: If Crowther's hobby was something other than caving,
>>>> would the history of gaming have included so many dungeon crawls?
>>
>>> Yes.
>>
>>> First, because everyone who wrote a fantasy game would still have read
>>> Tolkein.

Not relevant, as pointed out elsewhere.

>>> Second, because an underground dungeon or cave is inherently an
>>> excellent solution to the problem of how to keep a player on the map.
>>
>>I disagree -- I think a lot of what's "obvious" about it is only long
>>habit.
>>

Yes, but the habit is from D&D. People have talked about how a dungeon
crawl is good at limiting players, but this was not lost on early
D&D players.

>This of course leads to the question where the dungeon crawls in
>D&D and other games came from.
>

I should ask one of my friends, who was part of the Greyhawk circle
(the second D&D campaign - Blackmoor was the first). He may have
some memories.

>D&D was inspired by Tolkien (most notably in the concepts of races)
>and Jack Vance (magic, of which Tolkien has very little) but perhaps
>mostly by Fritz Leiber (the notion of adventuring party, the thief
>class). None of these authors have written very much about dungeon
>crawls.
>

Correct.

>Perhaps it's just that dungeon crawls make good game settings, and
>that a number of successful games used it at the time IF took off.
>

To nitpick, they make easy game settings. It's easy to write up
a dungeon and write monsters and treasures in it, and back then we
thought it was a lot of fun. Heck, I still think so, although I
find that it has been computerized as roguelike games. Putting
some sort of plot in these games is hard, and the players are
likely to break out of it anyway.

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

Kevin Forchione

unread,
Jan 4, 2002, 1:12:06 PM1/4/02
to
"Richard Bos" <in...@hoekstra-uitgeverij.nl> wrote in message
news:3c356cc0....@news.tiscali.nl...

You think that's annoying, most people can't pronounce mine correctly.

--Kevin


L. Ross Raszewski

unread,
Jan 4, 2002, 1:17:01 PM1/4/02
to
On 4 Jan 2002 09:48:17 GMT, Magnus Olsson <m...@df.lth.se> wrote:

>But you have a point, albeit a rather weak one. I suppose that if I
>had been able to go back in time and influence Graham's design
>decision, I would have asked him to have at least three default
>end-of-game messages, viz. "You have won"/"You have died"/"You have
>lost"; or perhaps just one: "Game over".

Note: AGT has all of these, or at least three. I think.

However, if the standard library provided only a "Game Over" message,
I think *that* would have influenced game design a lot more than "the
flag that signals end-of-game is caled "deadflag"".

Dennis G. Jerz

unread,
Jan 4, 2002, 1:18:13 PM1/4/02
to
"Daryl McCullough" <da...@cogentex.com> wrote in message
news:a14gm...@drn.newsguy.com...

> Dennis says...
>
> >Good point, but why is Inform set up so that death is so easy to detect?
>
> I don't think it is so much that it is easy to detect, but that
> it requires special treatment (restarting the game).
>


But... does this business of asking the author to deal with "deadflag" bring
the concept of "death" to the forefront of a beginning author's mind?


--
Dennis G. Jerz, Ph.D.; (715)836-2431
Dept. of English; U Wisc.-Eau Claire
419 Hibbard, Eau Claire, WI 54702
------------------------------------
Literacy Weblog: www.uwec.edu/jerzdg

> --