player agency and Victor Gijbers's essay

13 views
Skip to first unread message

Jacek Pudlo

unread,
May 3, 2007, 8:22:12 AM5/3/07
to
Victor:

"Most radically, there have been games that no longer challenge the
player to reach the ideal ending: sometimes, choosing which ending is the
best
one is itself the most important act of the game"

Choice is meaningful in real life only if it is excluding and irrevocable.

An example of exclusion. John offers Tom a choice between an apple and a
plum. Tom chooses the plum. Once Tom has eaten the plum, John tells Tom that
John doesn't want the apple any more and that Tom can have it as well. By
offering the apple, John has invalidated Tom's choice. Tom's choice was a
meaningful choice only because it excluded the apple.

An example of irrevocability. A murderer can not unkill his victim.

When you apply the concept of choice to interactive fiction, you'll notice
that, from a *player's* point of view, choice is never excluding and always
revocable. Thanks to RESTART, UNDO, SAVE and RESTORE, choice in IF is
meaningless.

A game like _Slouching Toward Bedlam_ offers not choices, but a garden of
forking paths. When a player says that he has played _Slouching Toward
Bedlam_, what he means is not that he has made *one* "choice" and quit the
game, but that he has explored *all* the "choices," walked *all* the paths
in the garden. From the point of view of player agency, the order in which
the "choices" (paths) were made is of no consequence. If a game about the
Holocaust offers the player the "choice" of becoming either Adolf Eichmann
or Oskar Schindler, the player will explore both "choices," he will walk
both paths. The order in which the player "chooses" to walk the Eichmann
path has no moral relevance, because interactive fiction is not about
choice. Interactive fiction is about exploration, be it pseudo-moral
exploration (forking paths), or spatial exploration (mapping rooms).


albtraum

unread,
May 3, 2007, 11:17:08 AM5/3/07
to
You know, I think I agree with you.

I'm not such a big fan of "choice" or multiple endings as a major
concern in IF. I think choice is as big a red herring as "mimesis". If
I wanted several equally interesting choices which lead to potentially
different outcomes, I'd play chess. When I play a text adventure, I
want to go on the one main worthwhile narrative journey that the
author has planned for me, but with stops along the way for
exploration, maybe some branching paths here and there, some diversion
and some puzzle-solving. I don't want to replay the same game a
thousand times, trying a different door and getting a different ending
each time. Again, that's more like chess.

This is one reason why I like games which can't be put in an
unwinnable state. I don't need to be slapped on the hand and made to
restore for exploring an IF world. Just let me continue down the
narrative path. By starting up the interpreter and playing a certain
game, I've already made all the excluding and irrevocable choices I
want to make for that playing session. If you gave me a choice between
playing Rameses and playing Galatea, or a game where I keep missing
timed events and/or dying like Cutthroats, I'd always pick Rameses.

Jeff Nyman

unread,
May 3, 2007, 11:30:17 AM5/3/07
to
"albtraum" <toh...@gmail.com> wrote in message
news:1178205428.1...@y80g2000hsf.googlegroups.com...

>
> I wanted several equally interesting choices which lead to potentially
> different outcomes, I'd play chess. When I play a text adventure, I
> want to go on the one main worthwhile narrative journey that the
> author has planned for me, but with stops along the way for
> exploration, maybe some branching paths here and there, some diversion
> and some puzzle-solving.

I tend to agree with this as well. Replay value on most games for me is zero
and that's simply because I really don't want to. Once I finish something, I
generally want to move on. I don't want to have to go through the same thing
hunting for little bits I might have not experienced. I would rather the
story be told in a way that allowed me to experience the relevant bits. When
I read a novel, I want the story to be told in a way that the author
effectively covers the "other paths" and narrative branches such that I can
glean all this as I go through it. Same thing with a film.

That's not to say that you might not replay a very well written game again
(just as you might re-read a book or see a movie again), particularly if you
felt it was nuanced and you wanted to experience it again or if you felt
that later knowledge you gleaned from completing the game might make the
beginning parts more understandable in the light of your new knowledge.

Where text-based IF differs, of course, is that since it is more actively
interactive (as opposed, I guess, to passively interactive), there's
apparently this feeling that you can make the story more option-based and
thus more worthy of second, third, fourth (etc) trips to glean out all the
paths. (Granted, graphical games do this, as well. I played through the four
different endings of "Deus Ex" just to see them; but at the time, I felt I
was just doing it by rote. I wasn't actually having fun playing at that
point: I just wanted to see each of the endings because ... well, why not?
The same applies to "Blade Runner", which was really frustrating because the
branch points were all over the game.)

To me, this is somewhat like the "Choose Your Own Adventure" where you can
basically turn back to a page and try the other path. While I enjoyed those
books as a kid, I find I really don't as an adult because you're basically
just going through the motions to find the little snippets. I'm more
concentrating on the branch points (so I can remember to go back to them)
rather than the story. I'm not sure that's a good direction for text-based
IF but, of course, some players may very well want exactly that experience.

- Jeff


George Oliver

unread,
May 3, 2007, 11:34:01 AM5/3/07
to
Am I the only player of IF who does -not- explore every path in a
game? It's unusual for me even to play a game twice. It doesn't sound
reasonable to confuse a way of playing IF with what IF is. These are
like the arguments that conflate ways of roleplaying with what a RPG
'should be'.

fel...@yahoo.com

unread,
May 3, 2007, 1:29:02 PM5/3/07
to
On May 3, 11:34 am, George Oliver <georgeolive...@gmail.com> wrote:
> Am I the only player of IF who does -not- explore every path in a
> game? It's unusual for me even to play a game twice.

Same here. On the other hand, I do like to have choices, because
on that one playthrough I can actively seek the path/ending I like
better. I enjoyed Whom the Telling Changed a lot, precisely because
it allowed me to reach the ending *I* liked, and it made getting
there a challenge. Thus it was both a good story and a good game.

Felix

Jacek Pudlo

unread,
May 3, 2007, 1:51:21 PM5/3/07
to
George Oliver

> Am I the only player of IF who does -not- explore every path in a
> game? It's unusual for me even to play a game twice. It doesn't sound
> reasonable to confuse a way of playing IF with what IF is. These are
> like the arguments that conflate ways of roleplaying with what a RPG
> 'should be'.

Static fiction tends to revolve around a clash of forces, where "forces"
refer to characters, ideas, civilisations, fate, the individual contra
societal norms, etc. They are all diegetic entities -- they are all inside
the narrative. The funny thing about IF is that the central clash takes
place outside the narrative, between the player and the game. A reader
consumes a novel through the trivial act of reading. A player consumes a
game through the non-trivial act of playing. Reading entails attentiveness,
comprehension and erudition, but reading a novel is nowhere near as
fiendishly cruel as playing _All Things Devours_ or _Varicella_. These games
are not about what they purport to be; they are not about saving Boston or
killing rivals to the throne. It is neither narratively nor morally
interesting whether the next king will be the evil and clever protagonist or
the evil and stupid prince. These games are simply about beating the game.
While a reader asks himself, "How do the actions of this character reflect
on the human condition?," the player asks "How do I manipulate this
character in a way that will trigger the condition for a winning state?"

This brings me back to your question. Playing a game entails beating it,
which in turn entails exploring all the pre-plotted paths. If you have
failed to uncover one of the five paths available in _Slouching Toward
Bedlam_, then you have failed to beat the game. Note that this has nothing
to do with choice. The player is not excercising his agency -- he is simply
seeking to explore all the paths. The reward is the satisfaction of having
beaten the game, not the pleasure of a well-crafted narrative where complex
characters clash with eachother. This is why most IF games are replayable
only when multiple paths are provided. This is also the reason why it isn't
much fun to replay a game like _Anchorhead_. Once you've unlocked the game
mechanics, the conditions that trigger the winning state(s), there is little
point in returning because game mechanics are narratively trivial.

From an artistic point of view, this is a sad state of affairs. Since the
player is not clashing with NPCs through the proxy of the PC and thus
interacting with a narrative, but rather clashing with the concealed
mechanics of the program, plots tend to be crude and improbable and
characters tend to be shallow, empty vehicles the player can inhabit during
a session and discard without a second thought. Even the best games tend to
run out of narrative steam somewhere midway and devolve into banal weirdness
and nonsense. The otherwise excellent _Anchorhead_ is a good example of
this.

The most tragic example of replacing narrative with game mechanics is
_Galatea_. In no other game has so much code been devoted to a single
character, and in no other game is there a character quite as bloodless and
bland as Galatea. Only aesthetic strength can guarantee compelling
characters, which is why _Rameses_ is a far more competent character study
than _Galatea_.


Victor Gijsbers

unread,
May 3, 2007, 5:01:56 PM5/3/07
to
Jacek Pudlo wrote:

> Choice is meaningful in real life only if it is excluding and irrevocable.

It doesn't need to be excluding and irrevocable in a strong sense,
though. I can choose to eat the plum, and not eat the apple. I might at
any time reconsider this decision and decide to eat the apple as well,
but that doesn't make my choice any less real.

I agree with you that choice would not be interesting in a piece of IF
if all readers of IF always explored all possible paths. (I will qualify
this slightly below.) However, everyone who has chimed in on this thread
tells us that they _don't_ try out every possibility. It appears to me
that this gives the choices they make during the game the kind of
minimal irrevocability that is enough to make them real choices.

(I therefore don't understand why albtraum says he agrees with you, as
what he says seems to undermine your position.)


Now the qualification. An experience which I have had with some computer
games like "Baldur's Gate 2" and "Planescape: Torment" is that there is
a very real difference between the first playthrough and subsequent
replays. The first time, I care about the choices I make and what they
say about my character. I don't use the save/restore options in order to
find out the best deals I can make.

After I have finished the game, I may return to it with a very different
aim: seeing every nook and cranny. In that mode, I don't mind doing all
kinds of evil or otherwise aesthetically less pleasing things. I just
want to see everything and become the biggest badass in the game.

What this means for the topic under discussion is that I can easily
imagine, based on personal experience, that I would read a game first in
a kind of "narrative mode", where the choices matter to me; and
afterwards, if I am curious, I may play it again in a purely
"exploratory" mode in which I like to see all endings.


Regards,
Victor

Jacek Pudlo

unread,
May 3, 2007, 6:27:41 PM5/3/07
to
Rikard Peterson

> In article <1178206441.7...@h2g2000hsg.googlegroups.com>,


> George Oliver <georgeo...@gmail.com> wrote:
>
>> Am I the only player of IF who does -not- explore every path in a
>> game? It's unusual for me even to play a game twice. It doesn't sound
>> reasonable to confuse a way of playing IF with what IF is. These are
>> like the arguments that conflate ways of roleplaying with what a RPG
>> 'should be'.
>

> No, you're not. I'm the same way. I liked Slouching. It was my favourite
> game in that competition, but I still didn't feel the need to "beat it"
> by exploring every ending. I think I replayed it once to see it again in
> the light of my new knowledge and to see another ending, but I'm
> satisfied with having played it without having looked under every rock.

This is equivalent to skipping three fifths of the stories in a loosely
connected short story collection. While this approach may or may not be
satisfying, you are not entitled to rate or review the collection. Nor
should you review the stories you've read until you've finished the
remaining ones. So in a sense, it's a defeat. You will never see the whole
picture/unifying theme/whatever.

You can skim and pick through a novel, but you can play IF only in the
*exact* sequence intended by the author. So paradoxically interactive
fiction is less interactive than static fiction. A book can be read from
cover to cover more or less attentively, whereas a work of interactive
fiction is an obstacle run that demands the player's complete attention. A
work of interactive fiction may frustrate you, it may infuriate you, but one
thing it will never do is put you to sleep. In that sense, a work of IF is
an extra-diegetic competition between game and player with the narrative the
sham arena. You are not identifying themes, or bonding emotionally with the
characters, or being struck by the brilliance of the prose -- as you would
in a work of static fiction. You are instead directing a rat through a rat
race, an obstacle course devised with the sole intention of thwarting
progress. This is possible because an IF author possesses a level of
authorial control a novelist can only dream of. The belief that interactive
fiction is somehow more "user-malleable," less authorially controled than
static fiction is based on a profound misconception of the medium. The exact
opposite is the case -- the "interactive" in "interactive fiction" is a
misnomer, an illusion. This pernicious misconception is the Evil Seed which
has sprouted into the Simulationist Herecy.

"What is the cause of the Simulationist Herecy, Jacek?" The cause, Reader,
is the naive reification of the metaphor of spatial movement. When you type
WEST and the room description titled "Kitchen" is replaced by a room
description titled "Living Room," you assume -- quite naturally -- that was
has occurred is movement in space. What has really occurred is glossic
movement. Because the only thing that has changed, beside the room
description, is the set of available nouns. The nouns that were in scope in
"Kitchen" are no longer in scope in "Living Room." They have been replaced
by other nouns, hence the phrase "glossic movement," movement in words
rather than space. The simulationist heretics have simple-mindedly confused
metaphor with reality and assumed that the playing character is *really*
moving in space, or through a "world model," one of their favourite catch
phrases. They think IF is a kind of verbal 3D shooter where the player
directs the playing character through a "simulation" while "interacting"
with "a world model" and exercising "moral agency."


Adam Thornton

unread,
May 3, 2007, 7:38:07 PM5/3/07
to
In article <xlt_h.40034$E02....@newsb.telia.net>,

Jacek Pudlo <ja...@jacek.jacek> wrote:
>"What is the cause of the Simulationist Herecy, Jacek?" The cause, Reader,
>is the naive reification of the metaphor of spatial movement. When you type
>WEST and the room description titled "Kitchen" is replaced by a room
>description titled "Living Room," you assume -- quite naturally -- that was
>has occurred is movement in space. What has really occurred is glossic
>movement. Because the only thing that has changed, beside the room
>description, is the set of available nouns. The nouns that were in scope in
>"Kitchen" are no longer in scope in "Living Room." They have been replaced
>by other nouns, hence the phrase "glossic movement," movement in words
>rather than space.

But when I actually go from my living room into my kitchen in real life,
my motion through space (what is space?) isn't really the important
thing. The important thing is that the exits go to different places,
and there is a different set of nouns in scope, with applicability
towards solving different problems.

Adam


Jeff Nyman

unread,
May 3, 2007, 8:00:46 PM5/3/07
to
"Jacek Pudlo" <ja...@jacek.jacek> wrote in message
news:xlt_h.40034$E02....@newsb.telia.net...

> This is equivalent to skipping three fifths of the stories in a loosely
> connected short story collection. While this approach may or may not be
> satisfying, you are not entitled to rate or review the collection.

Well, no one is "entitled" with anything one way or the other. People rate
or review collections, regardless of how much they read of them. It happens
all the time. I'm not sure in what sense you're using "entitled" here. You
are perfectly within your rights to review or rate anything you want.
Hopefully the person who reviews in such a fashion, however, is honest about
this. (Much as a reviewer should admit if they have not read an entire book
or seen an entire film that they're claiming to rate or review. I always
love those reviews on Amazon that say something like: "I haven't read this
book, but I'm sure its crap.") It's up to the reader of the review to
determine how much to trust it.

> You can skim and pick through a novel, but you can play IF only in the
> *exact* sequence intended by the author.

True -- and a good distinction. It's obvious in a way, but it actually does
bear stating because it's easy to forget that.

> So paradoxically interactive fiction is less interactive than static
> fiction.

Well, I'm not so sure. In one sense, yes. In other sense, no. A novel
doesn't let me investigate things at my whim. If a given chapter takes place
in a room and mentions that the room has chairs and a table, then if the
novel writer didn't describe those, I'm out of luck. With text-based IF, I
at least have the chance to do so. (Whether the author implemented
descriptions for said items is another issue.) Characters I read in a novel
I can't "talk to" at all. I have to read the dialogue. With text-based IF, I
can "talk to" these characters. (Whether the author implemented various
conversation states with the characters is another issue.)

> A book can be read from cover to cover more or less attentively, whereas a
> work of interactive fiction is an obstacle run that demands the player's
> complete attention.

Again, I think this is a good point. It is hard to be too inattentive to a
work of text-based IF because you have to ultimately do something more than
just turn the pages. But this argument you make could also be said to goes
against the one you just said above: that interactive fiction is less
interactive than static fiction. An "obstacle run" that demands "complete
attention" could be said to be more interactive (even if frustratingly so)
than a novel.

> work of interactive fiction may frustrate you, it may infuriate you, but
> one thing it will never do is put you to sleep.

Hmmm -- I'm not sure about that. Some of the text-based IF I've played have
given rise to similar feelings of certain books that I was not too keen on
reading any further. Whether they "put me to sleep" is not really the issue;
my reaction to them, and the similarities of the reaction, are more the
issue for me.

> sham arena. You are not identifying themes, or bonding emotionally with
> the characters, or being struck by the brilliance of the prose -- as you
> would in a work of static fiction.

I agree that this is perhaps usually the case. But are you arguing that you
couldn't have a work of text-based IF crafted such that "bonding
emotionally" was made possible? Of course, the problem with that is what it
means to "bond emotionally" is very subjective and very different for
various people. I read about people who supposedly cried when Floyd in
"Planetfall" died. Fine. Didn't do that for me, however. In the film arena,
a woman had a heart attack while watching "The Passion of Jesus Christ." I
certainly didn't get that emotionally involved but clearly she did.

All of this, of course, only matters to those who are looking to "bond
emotionally" or who want to be "struck" by the "brilliance of the prose" or
to spend their time "identifying themes." I tend to like what you're
describing as well, and I do think it's missing from most text-based IF that
I've seen, but just because I like these things, it doesn't mean I need them
all the time. Sometimes I read certain books just because it's a good read.
("Dead City" by Joe McKinney comes to mind.) Sometimes I watch movies just
for the stupid action. ("xXx: State of the Union" comes to mind.)

So not all people do this with static fiction and more than they all do with
text-based IF. People's tastes differ.

>You are instead directing a rat through a rat race, an obstacle course
>devised with the sole intention of thwarting progress.

Yes -- or, at least, I think I agree. My focus on the narrative elements as
of late has been on to what extent the past history of text-based IF has led
to games where puzzles are the pacing mechanisms and where the puzzles are
largely designed to halt forward progress *just enough* to make things "fun"
but not enough to completely alienate the players.

> This is possible because an IF author possesses a level of authorial
> control a novelist can only dream of. The belief that interactive fiction
> is somehow more "user-malleable," less authorially controled than static
> fiction is based on a profound misconception of the medium.

Yes, I agree. But, again, this seems to go against what you said earlier:
that a work of text-based IF is "less interactive" than static fiction. That
level of authorial control is what I believe can and does make text-based IF
what I call actively interactive, as opposed to passively interactive.

> The exact opposite is the case -- the "interactive" in "interactive
> fiction" is a misnomer, an illusion. This pernicious misconception is the
> Evil Seed which has sprouted into the Simulationist Herecy.

Okay --- I think I get your point, then. You're saying that the
interactivity *from the player's point of view* is much less than with
static fiction *because* the authorial control is so much greater with
interactive fiction than with static fiction.

Is it accurate to say that your contention is that the more authorial
control that exists, the less player interactivity is allowed?

If so, that's an interesting assertion but how "interactive" someone finds
something can be somewhat subjective as well, unless you operationally
define what you mean by "interactivity." It's a malleable term and easy to
throw around.

> rather than space. The simulationist heretics have simple-mindedly
> confused metaphor with reality and assumed that the playing character is
> *really* moving in space, or through a "world model," one of their
> favourite catch phrases.

Or could it just be that the use of "world model" is an idea that helps
capture how people conceptualize what's happening? In other words, do you
assume that for all people the use of "world model" isn't itself the
metaphor? The use of a term doesn't necessarily mean people *really* think
something like spatial movement is *really* taking place. I use the term
"world model" and I can tell you that I don't think that.

By the same token, if I play a flight simulator I don't *really* think I'm
flying a plane, for example. (When I played "Doom 3", I didn't think I was
*really* a Space Marine who was *really* on Mars either.) But I accept the
metaphor of a "world model" through which I'm flying a plane or blasting
through the Legions of Hel. I suppose I could look at it more seriously and
pull out a ton of academic concepts from literary theory. Or I could just
have fun with what is quite clearly and understandably a game. Where I think
text-based IF is interesting is that it allows people to approach it with a
more literary intent than flight simulators or 3D shooters, and which leads
to these interesting differences of opinion among people and leads to the
different emphases people put on what their expectations are.

- Jeff


Jacek Pudlo

unread,
May 3, 2007, 9:44:14 PM5/3/07
to
Adam Thornton

> In article <xlt_h.40034$E02....@newsb.telia.net>,
> Jacek Pudlo <ja...@jacek.jacek> wrote:
>>"What is the cause of the Simulationist Herecy, Jacek?" The cause, Reader,
>>is the naive reification of the metaphor of spatial movement. When you
>>type
>>WEST and the room description titled "Kitchen" is replaced by a room
>>description titled "Living Room," you assume -- quite naturally -- that
>>was
>>has occurred is movement in space. What has really occurred is glossic
>>movement. Because the only thing that has changed, beside the room
>>description, is the set of available nouns. The nouns that were in scope
>>in
>>"Kitchen" are no longer in scope in "Living Room." They have been replaced
>>by other nouns, hence the phrase "glossic movement," movement in words
>>rather than space.

I happen to know that Adam Thornton is currently undertaking a sex safari in
the Caribbean, and has taken neither laptop nor cell phone, and has
instructed his butler/sex-gimp to do likewise. So whoever you are, you are a
mere Botkin to the Kinbote of his Shade.

Nonetheless, I shall indulge you.

> But when I actually go from my living room into my kitchen in real life,
> my motion through space (what is space?)

Matter in the right place.

> isn't really the important
> thing. The important thing is that the exits go to different places,
> and there is a different set of nouns in scope, with applicability
> towards solving different problems.

Exits are nouns, dear silly impostor. WEST is simply short for GO WEST.
Directional nouns happen to be highly deictic, meaning different things in
different "locations." A "location" is nothing but a glossary plus the
"room" description. From a purely glossic perspective, a description is a
textual presentation of one or more glosses. A gloss is a set of nouns and
adjectives that refer to the same "object." An "object" is an amalgam of
description, gloss and "behavior." Behavior is a tricky thing, a complex
illusion. Verbs are operative glosses available in all glossaries. (The
exception are games like _Adverbum_.) Several verbs can, and often do, refer
to the same operative gloss. By applying operative glosses (verbs) to
glosses (nouns and adjectives) behavior is triggered. The effect is an
illusion commonly known as "object." Some glosses share the same nouns,
hence disambiguation. The "playing character" and "inventory" and
"bodyparts" are metaphors for global glosses, which is a set of glosses
available in all glossaries. A dark room is an empty glossary. Darkness is a
metaphor for glossic emptiness, just as spatial movement is a metaphor for
change of glossary. What the heretics call "changing the state of the world
model" is in fact fiddling with the contents of glossaries. When the
operative gloss DROP is succesfully applied to an "inventory" gloss, that
gloss ceases to be a global gloss and becomes a local gloss. Next time the
glossary changes, the DROPped gloss will no longer be in scope.

Also sprach Jacek. Wovon Jacek sprechen kann, darüber sollte man am besten
schweigen.


(Note to future biographers: This post marks the onset of my
Nietzschean/Wittgensteinian madness. You may approach this phase from a
Freudian perspective, though it is highly doubtful whether a Freudian
reading of Pudlo will do justice to my complex and compelling personality. A
Pudlian reading of Freud might prove more productive, or at least more
amusing. Here are some Freudian slips you can play with. "The condom
denominator of cuntable and uncuntable nuns is that neither is a noun." Oh
well, I might as well throw you a bone. Yesterday I was dreaming that I was
putting a cup into a cupboard. The cupboard is clearly a symbol of my
mother's vagina, and the cup [Merciful snip of self-exegesis. The "note"
continues for some 80 pages, scaling new heights of bad taste and egomania.
M.L.])


Jacek Pudlo

unread,
May 3, 2007, 11:48:34 PM5/3/07
to
Emily Short

> I tend to play until I've got an ending I'm satisfied with

> I almost always do identify to myself a preference of endings, even if
> that ending is not the last one I explore.

> I am pretty

> I felt like I understood the parameters

> there are still things I haven't found out. This can be a little
> frustrating.

> there are a few things I never found out. I would like to know those
> things

> I also don't have specific ideas

> if I did, that might be a different matter.


You are so dull, I almost pity you. Almost.


Emily Boegheim

unread,
May 4, 2007, 3:35:00 AM5/4/07
to

Literally, that's true. However, if I willingly suspend disbelief,
that's not how I experience a work of IF. And it's the player's
experience that matteres, not the author's\IF system designer's
experience.

Emily

Victor Gijsbers

unread,
May 4, 2007, 6:03:59 AM5/4/07
to
Jacek Pudlo wrote:

> "What is the cause of the Simulationist Herecy, Jacek?" The cause, Reader,
> is the naive reification of the metaphor of spatial movement. When you type
> WEST and the room description titled "Kitchen" is replaced by a room
> description titled "Living Room," you assume -- quite naturally -- that was
> has occurred is movement in space. What has really occurred is glossic
> movement. Because the only thing that has changed, beside the room
> description, is the set of available nouns. The nouns that were in scope in
> "Kitchen" are no longer in scope in "Living Room." They have been replaced
> by other nouns, hence the phrase "glossic movement," movement in words
> rather than space. The simulationist heretics have simple-mindedly confused
> metaphor with reality and assumed that the playing character is *really*
> moving in space, or through a "world model," one of their favourite catch
> phrases. They think IF is a kind of verbal 3D shooter where the player
> directs the playing character through a "simulation" while "interacting"
> with "a world model" and exercising "moral agency."

Although there is clearly _some_ truth in what you are saying, you
nevertheless seem blinded to an important aspect of interactive fiction
that lends considerable comfort to simulationist ways of speaking.
Because the reader is an actor with purposes, reading interactive
fiction will always involve creating a predictive model of the text
under interactions: the reader needs to form and test expectations as to
how the text will respond to different commands. Now, this model can
take many different forms; but in a huge number of games, a 'model
world' in the strict sense of the term (that is: a model which assumes
that spatial, temporal and causal relationships much like those in the
real, physical world have been implemented) will be the most successful
model. In most games, the road to success consists of not just taking
the metaphors of space, time and so forth seriously, but of actually
basing your decisions to interact on what you would expect to work if
you were to interact with a world instead of a text.

This opens the door for all kinds of simulationist projects and
simulationist discourse, it seems to me. (Also, I can't really see how
3D shooters are supposed to be different from IF in this respect. Are
coloured pixels somehow less textual than words? Your textualist
inclinations should tell you otherwise!)

Regards,
Victor


PS. All this said, I don't give a damn about carefully constructed
simulations of physical properties in IF. I'm really not the tinkering type.

Jacek Pudlo

unread,
May 4, 2007, 7:23:32 AM5/4/07
to
Emily Boegheim

This is approximately the argument of the Flat Earth Society. The funny
thing about theorising experience instead of the reality underlying the
experience, is that in the long run the theories get unintuitive, they
become the very opposite of what they were supposed to be. How does a member
of the Flat Earth Society explain the behavior of the horizon? How about a
lunar eclipse? To account for these phenomena, he is forced to construct
increasingly extravagant, silly and unintuitive theories. It takes a leap of
faith to accept that Earth is spherical, but once you do, the other pieces
fall into place.

"Are you saying that the simulationists are the Flat Earth Society of IF,
Jacek?" Yes, Emily. From the point of view of a novice, a work of IF appears
to be first and foremost a simulation, a model of a physical world, a kind
of a turn-based, brainy, 3D shooter where the polygons have been replaced by
words. The novice types PUT PEN ON DESK and EXAMINE DESK reveals that the
pen is indeed on the desk. "Wow! This is like _Deus Ex_, only in words!"
exclaims the novice. The problem with this approach is that it doesn't
really explain much. It does perhaps explain games like Emily Short's
_Mystery House_, or Steve Breslin et al's _Battle of Walcott's Keep_,
"explain" in the sense of diagnosing the illness. Games that yield to
simulationist theories are very rare, and almost inavariably contain a
strong ideological component. They were written not with the intention to
entertain, to tell a story, but solely to verify a fallacious theory. Far
more interesting games, like _Photopia_ and _Rameses_ remain a mystery to
the simulationists, simply because they are not simulations, not in any
meaningful, theoretically productive sense. Even more perplexing to the
simulationists are are games like _Ad Verbum_ and _Letters From Home_. What
exactly are they "simulating"? Words?

Simulationism is a wasteland, a narrow barren desolate slice of land where
nothing will ever grow. What I am opening up with my glossicism is a wide
well-manured field, fecund enough to sprout joyous theoretical excess and
spacious enough to embrace *all* works of interactive fiction, not only
those written by Steve Breslin and Emily Short.


Jeff Nyman

unread,
May 4, 2007, 7:46:58 AM5/4/07
to
"Jacek Pudlo" <ja...@jacek.jacek> wrote in message
news:UIE_h.40060$E02....@newsb.telia.net...

> The funny thing about theorising experience instead of the reality
> underlying the experience, is that in the long run the theories get
> unintuitive, they become the very opposite of what they were supposed to
> be.

Except in physics and biology, where often the opposite is true. If you
theorize the "reality underlying" our experiences, you get to quantum theory
and then the theories get very unintuitive. If you theorize the "*really*
underlying reality" (according to some) you get superstrings and then you're
really getting into the unintuitive aspects. The same applies to some
aspects of self-organizing attributes of various types of biological systems
or the nature of viruses and prions. If you study consciousness research,
you'll probably come across the idea of consciousness being "generated" by
attributes of the cytoskeletons in microtubules. Here's another case where
trying to get at the (alleged) "reality underlying" our experience (of
consciousness) leads to quite unituitive notions of how such processes could
truly "generate" what we call conscious experience.

Bringing this back to text-based IF, a lot of people don't feel the need to
theorize about the "reality underlying the experience" of playing such
games. People are happy to theorize about the experience itself. Others want
to dig deeper. Personally, I think I fall somewhere between the two. Too
much focus on the "reality underlying the experience" seems to be to be
disconnecting from the vast majority of people who play such games. Yet too
much focus solely on the "experience" itself seems too limiting and has, in
my opinion, constrained how text-based IF is often written.

> Yes, Emily. From the point of view of a novice, a work of IF appears to be
> first and foremost a simulation, a model of a physical world, a kind of a
> turn-based, brainy, 3D shooter where the polygons have been replaced by
> words.

How does it appear to someone who is not a novice? Let's say you have
someone who has played numerous such text-based IF games. I presume then
they are not a novice. Are you saying that such people don't perceive a
simulation in any sense?

> "explain" in the sense of diagnosing the illness. Games that yield to
> simulationist theories are very rare, and almost inavariably contain a
> strong ideological component. They were written not with the intention to
> entertain, to tell a story, but solely to verify a fallacious theory. Far
> more interesting games, like _Photopia_ and _Rameses_ remain a mystery to
> the simulationists, simply because they are not simulations, not in any
> meaningful, theoretically productive sense.

But not everyone is looking for a "meaningful, theoreatically productive"
sense when they're playing games (or reading books, for that matter). I
notice you often put things in that context and that's fine, of course, but
not everyone is looking for what you apparently are in such games.

For my part, I do see "Rameses" as a simulation of sorts. A simulation is
basically an imitation of something. I suppose if you want to go with a
dictionary approach, it's a "representation of the behavior or
characteristics of one system through the use of another system." In the
case of "Rameses" the behavior or characteristics I was looking at were that
of a person whose actions are largely passive. I know the computer science
definition of a "simulation" is different but I'm not playing text-based IF
with the hope that it matches the computer science definition of what a
simulation is.

For me, you can simulate what it's like to be a certain *type* of person or
you can simulate a *type* of mentality, which I believe a game like
"Rameses" did and to good effect.

- Jeff


Jacek Pudlo

unread,
May 4, 2007, 9:53:42 AM5/4/07
to
Victor Gijsbers

Let's put IF under the microscope. Let's take a look at some pseudo I6
source code.

Object -> Chair "leather armchair"
with name "chair" "leather" "heavy" "furniture" "armchair" "old",
description [;
print "Stuffing shows through holes in the worn leather.
Despite the tattered state, there is a certain grandeur
in the armchair's ruin, an air of regality and aloofness.
Reclining in it is the privilege of the chosen few, moving
it would be tantamount to an insult.^";
],
before [;
DirPush: A number of puzzles are solved this way.
Take: "Too heavy and cumbersome to carry around.";
LookUnder: "You notice indentations in the rug, right next to where
the armchair stands.";
Touch, Search:
"You press your hand into the leather of the armchair,
making a vague imprint. It feels soft and supple, almost
like skin. As you lift your hand, the leather slowly
smooths itself regaining its cool form, forgetting you.";
Taste:
"The leather is smooth on the tongue, with a smoky aftertaste.";
Smell:
"The leather smells of cigarettes and rum, with a faint little whiff
of
expensive perfume.";
Enter:
"Given inspiration has turned a deaf ear to you for the moment,
you settle into the old piece, making the leather creak and rustle.
Resting your elbows on the desk, you steeple your fingers,
furrow your brow, and think deeply before heaving yourself out
of the armchair. Time is vital.";
],
has scenery enterable furniture;

What I see here is a gloss ("leather armchair") that can be referred to by
the player by a set of nouns and adjectives ("chair" "leather" "heavy"
"furniture" "armchair" "old") and operated upon with operative glosses
("Take" "LookUnder" "Touch" etc.) to trigger behavior. In this simplified
example the behavior consists exclusively of textual description (with the
exception of PushDir). The descriptions fulfill three purposes: aesthetic,
narrative and functional. The aesthetic purpose is to provide the
player/reader with good literary prose. Exactly what is implied by "good
literary prose" is external to my theory. Pudlian Glossicism aims to provide
an anatomy of interactive fiction, not an aesthetic theory of literary
prose. I can only say that the a modest aim is to avoid "noun scanning." IF
prose is often so drab and pedestrian that players merely scan the text for
interactable nouns. This can only be avoided if the prose is so pleasurable
that it *pays* to read it. The narrative and functional purpose is to
provide allusions (hints) to the solving of obstacles (puzzles). In the
above example the response to the operative gloss LookUnder makes the player
aware of the gloss "indentations," which hadn't been mentioned in the room
description. The description of the gloss "leather armchair" contains an
allusion to moving the chair, which is a means of overcoming several
obstacles. Ideally, overcoming obstacles should propel the story forward.
All this coagulates into the illusion commonly known as "object." What you
call "world model" is merely a superstructure composed of the above
components. From a theoretical point of view, anything above the level of
gloassaries is uninteresting. It would be like trying to understand physics
in terms of "objects."

What you call "creating a predictive model," I call "creating an
atmosphere." Note how the response to the operative gloss ENTER doesn't
change the game state. No flags are set because the playing character is
described as rising from the chair directly after having sat. The purpose is
not to provide a predictive model where the paying character can -- quite
predictably -- sit on chairs, because that is narratively trivial. The
purpose is to create an atmopshere, in this particular case a sense of
urgency thwarted by indecision.

You are correct in accusing me of willful blindness, Victor, but note that
it is the willful blindness of the scientist who sees not a "bug" under the
microscope, but cells and mitochondria. These components are theoretically
far more interesting than the superstructures they compose.

As to "testing expectations," I find the notion in its naked form trivial.
Testing whether a chair can be sat upon is certainly something players do --
and *should* do -- but merely responding "You sit down on the chair." to SIT
ON CHAIR is almost as bad as not repsonding at all. The primary goal of
typing SIT ON CHAIR is not to test the the coherence of the "world model,"
but to unlock a fragment of text, a fragment of text that, ideally, should
be aesthetically, narratively and functionally meaningful and interesting.

> This opens the door for all kinds of simulationist projects and
> simulationist discourse, it seems to me. (Also, I can't really see how 3D
> shooters are supposed to be different from IF in this respect.

Because 3D shooters aspire to mimesis. Interactive fiction aspires to
diegesis.

> Are coloured pixels somehow less textual than words?

Yes. Very much so.

> Your textualist inclinations should tell you otherwise!)

No.


fel...@yahoo.com

unread,
May 4, 2007, 1:40:30 PM5/4/07
to
On May 4, 9:53 am, "Jacek Pudlo" <j...@jacek.jacek> wrote:

> You are correct in accusing me of willful blindness, Victor, but note that
> it is the willful blindness of the scientist who sees not a "bug" under the
> microscope, but cells and mitochondria. These components are theoretically
> far more interesting than the superstructures they compose.

You know what they say about the scientist who tried to understand
clocks by dismantling them...

Felix

albtraum

unread,
May 5, 2007, 4:55:38 AM5/5/07
to

The way I heard it, he threw them out the window...

Stephen Bond

unread,
May 6, 2007, 6:32:25 PM5/6/07
to
On May 4, 1:46 pm, "Jeff Nyman" <jeffny...@gmail.com> wrote:
> For my part, I do see "Rameses" as a simulation of sorts. A simulation is
> basically an imitation of something. I suppose if you want to go with a
> dictionary approach, it's a "representation of the behavior or
> characteristics of one system through the use of another system." In the
> case of "Rameses" the behavior or characteristics I was looking at were that
> of a person whose actions are largely passive. I know the computer science
> definition of a "simulation" is different but I'm not playing text-based IF
> with the hope that it matches the computer science definition of what a
> simulation is.
>
> For me, you can simulate what it's like to be a certain *type* of person or
> you can simulate a *type* of mentality, which I believe a game like
> "Rameses" did and to good effect.

I don't see Rameses as a simulation. By "simulation" here we
seem to be talking about something like a flight simulator
-- an attempt to model or predict how some object reacts to
a various user inputs, or various changes in its environment.
A simulation keeps some representation of the state of the
simulated object; Rameses does no such thing. In a simulation,
the object's reactions to input are determined by a series of
encoded behavioural rules; in Rameses, the PC's reactions
to input are determined by me. Even the possible inputs are
determined by me.

If simulation goes on, it goes on outside the text. I "simulated"
a character in my head. I had a good idea how he'd react in
any given situation; perhaps players do too. But that's not the
game, that's not the work of IF.

In writing Rameses, it wasn't my intention to say "here -- see what
it's like to be a dork for a day!" (a blurb perhaps better suited
to A Moment of Hope, a game I loathe). Rameses is not Shyness
Simulator version 1.0, but an attempt to tell a story using the
medium of IF. It is not an attempt to simulate a character, but to
create one. There's a difference.

I think someone going in expecting a simulation will get a lot
less from the game. As someone on IF ratings said, it's no fun
being an ineffectual whiner. But if instead you see it as an
interactive story, there's a chance that the experience will
be better.

Stephen.

Jacek Pudlo

unread,
May 6, 2007, 8:47:10 PM5/6/07
to
Jeff Nyman

> "Jacek Pudlo" <ja...@jacek.jacek> wrote in message
> news:xlt_h.40034$E02....@newsb.telia.net...
>
>> This is equivalent to skipping three fifths of the stories in a loosely
>> connected short story collection. While this approach may or may not be
>> satisfying, you are not entitled to rate or review the collection.
>
> Well, no one is "entitled" with anything one way or the other. People rate
> or review collections, regardless of how much they read of them. It
> happens all the time.

So does rape. You are conflating what is typical with what is desirable.

> I'm not sure in what sense you're using "entitled" here.

In the sense of professional honesty. A reviewer should not review a book
unless he's read it from cover to cover.

> You are perfectly within your rights to review or rate anything you want.
> Hopefully the person who reviews in such a fashion, however, is honest
> about this.

Honest about his flippancy and ineptitude? Like the pilot who admits shortly
before landing that he's drunk? Yeah, I know, lives are not endangered by an
inept reviewer. But the same principle applies.

> (Much as a reviewer should admit if they have not read an entire book or
> seen an entire film that they're claiming to rate or review. I always love
> those reviews on Amazon that say something like: "I haven't read this
> book, but I'm sure its crap.") It's up to the reader of the review to
> determine how much to trust it.

What you've quoted above isn't a review.

>> You can skim and pick through a novel, but you can play IF only in the
>> *exact* sequence intended by the author.
>
> True -- and a good distinction. It's obvious in a way, but it actually
> does bear stating because it's easy to forget that.
>
>> So paradoxically interactive fiction is less interactive than static
>> fiction.
>
> Well, I'm not so sure. In one sense, yes. In other sense, no. A novel
> doesn't let me investigate things at my whim. If a given chapter takes
> place in a room and mentions that the room has chairs and a table, then if
> the novel writer didn't describe those, I'm out of luck. With text-based
> IF, I at least have the chance to do so. (Whether the author implemented
> descriptions for said items is another issue.) Characters I read in a
> novel I can't "talk to" at all. I have to read the dialogue. With
> text-based IF, I can "talk to" these characters. (Whether the author
> implemented various conversation states with the characters is another
> issue.)

The fundamental difference between static fiction and interactive fiction is
that the latter is opaque. The text of a novel is transparent. It is
immediately accessible, either as a relentless flow of words that stretches
from first page to last page, or as a smörgåsbord where the reader can at
leasure pick individual chapters, paragraphs and sentences. The text of an
IF game is never available in its entirety. The text of an IF game is
instead meted out in small fragments that contain allusions to the unlocking
of the remaining fragments. Even after having finished the game, the player
can never be sure of having read the entire text. Worse still, different
sessions will more likely than not result in different transcripts. Even if
the *exact* same sequence of commands is issued on two sessions, the
resulting transcripts may differ because of randomisation. In static fiction
symbols and themes may remain enigmatic and require re-reading, but the text
is available in its entirety. In interactive fiction the very text is an
enigma, perhaps *the* enigma. This could explain why most IF games do not
yield to literary analysis. Once you've unlocked the required number of
textual fragments and reached all (or most of) the winning states, there's
little to analyse.

>> A book can be read from cover to cover more or less attentively, whereas
>> a work of interactive fiction is an obstacle run that demands the
>> player's complete attention.
>
> Again, I think this is a good point. It is hard to be too inattentive to a
> work of text-based IF because you have to ultimately do something more
> than just turn the pages. But this argument you make could also be said to
> goes against the one you just said above: that interactive fiction is less
> interactive than static fiction. An "obstacle run" that demands "complete
> attention" could be said to be more interactive (even if frustratingly so)
> than a novel.
>
>> work of interactive fiction may frustrate you, it may infuriate you, but
>> one thing it will never do is put you to sleep.
>
> Hmmm -- I'm not sure about that. Some of the text-based IF I've played
> have given rise to similar feelings of certain books that I was not too
> keen on reading any further. Whether they "put me to sleep" is not really
> the issue; my reaction to them, and the similarities of the reaction, are
> more the issue for me.

In interactive fiction, focus and attention are necessary conditions of
progress, no matter how dull the game.

>> sham arena. You are not identifying themes, or bonding emotionally with
>> the characters, or being struck by the brilliance of the prose -- as you
>> would in a work of static fiction.
>
> I agree that this is perhaps usually the case. But are you arguing that
> you couldn't have a work of text-based IF crafted such that "bonding
> emotionally" was made possible?

Theoretically, I believe it is possible. This doesn't, however, seem to be a
goal that IF authors are aiming for. There are a few cases when
player/character empathy is achieved, like the previously mentioned
_Photopia_ and _Rameses_, but the means of achieving the effect are crudely
manipulative. Compare _Rameses_ to any short story by Cheever or Updike, and
you'll see what I'm talking about. Both _Photopia_ and _Rameses_ lack the
subtlety readers of mainstream fiction have learned to expect from their
authors. The fact that these games are among the very best IF has to offer
says more about the state of the medium than about anything else. The
abysmal failure of Adam Cadre's novel is an example in point.

> Of course, the problem with that is what it means to "bond emotionally" is
> very subjective and very different for various people. I read about people
> who supposedly cried when Floyd in "Planetfall" died. Fine. Didn't do that
> for me, however. In the film arena, a woman had a heart attack while
> watching "The Passion of Jesus Christ." I certainly didn't get that
> emotionally involved but clearly she did.

You're right in that emotional involvement is a crude measure of aesthetic
success. But what else is there to _Photopia_ and _Rameses_? If these games
fail to touch you emotionally, what else do they have to offer? _Savoir
Fair_ has brilliant puzzle design and _All Roads_ has some of the best prose
ever written by an IF author, but seen as works of *fiction* they are tame
and bloodless. Despite their crudeness, _Photopia_ and _Rameses_ are light
years ahead of what usually passes for "fiction" in interactive fiction.

> All of this, of course, only matters to those who are looking to "bond
> emotionally" or who want to be "struck" by the "brilliance of the prose"
> or to spend their time "identifying themes." I tend to like what you're
> describing as well, and I do think it's missing from most text-based IF
> that I've seen, but just because I like these things, it doesn't mean I
> need them all the time. Sometimes I read certain books just because it's a
> good read.

As opposed to "demanding books," like _Rabbit, Run_, which are not "good
reads"?

> ("Dead City" by Joe McKinney comes to mind.) Sometimes I watch movies just
> for the stupid action. ("xXx: State of the Union" comes to mind.)
>
> So not all people do this with static fiction and more than they all do
> with text-based IF. People's tastes differ.

Yes, and some people have better taste than others.

>>You are instead directing a rat through a rat race, an obstacle course
>>devised with the sole intention of thwarting progress.
>
> Yes -- or, at least, I think I agree. My focus on the narrative elements
> as of late has been on to what extent the past history of text-based IF
> has led to games where puzzles are the pacing mechanisms and where the
> puzzles are largely designed to halt forward progress *just enough* to
> make things "fun" but not enough to completely alienate the players.

Interesting.

>> This is possible because an IF author possesses a level of authorial
>> control a novelist can only dream of. The belief that interactive fiction
>> is somehow more "user-malleable," less authorially controled than static
>> fiction is based on a profound misconception of the medium.
>
> Yes, I agree. But, again, this seems to go against what you said earlier:
> that a work of text-based IF is "less interactive" than static fiction.
> That level of authorial control is what I believe can and does make
> text-based IF what I call actively interactive, as opposed to passively
> interactive.

Please elaborate.

>> The exact opposite is the case -- the "interactive" in "interactive
>> fiction" is a misnomer, an illusion. This pernicious misconception is the
>> Evil Seed which has sprouted into the Simulationist Herecy.
>
> Okay --- I think I get your point, then. You're saying that the
> interactivity *from the player's point of view* is much less than with
> static fiction *because* the authorial control is so much greater with
> interactive fiction than with static fiction.

What other point of view is there?

> Is it accurate to say that your contention is that the more authorial
> control that exists, the less player interactivity is allowed?
>
> If so, that's an interesting assertion but how "interactive" someone finds
> something can be somewhat subjective as well, unless you operationally
> define what you mean by "interactivity." It's a malleable term and easy to
> throw around.

Interactivity of consumption. More specifically, the exact sequence and
level of attention and comprehension in which a text is read. You can *sort
of* get through _Moby Dick_, even if you have trouble untying the knots of
Melliville's somewhat tiresome and convoluted prose. Even a half-assed
reading will get you through the book. You may miss the metaphysical
implications of a mad guy chasing a whale, but at least you'll know it's
about a mad guy chasing a whale. Interactive fiction gives the author
*total* control over how the work is consumed. If you miss a single hint,
you're stuck.

>> rather than space. The simulationist heretics have simple-mindedly
>> confused metaphor with reality and assumed that the playing character is
>> *really* moving in space, or through a "world model," one of their
>> favourite catch phrases.
>
> Or could it just be that the use of "world model" is an idea that helps
> capture how people conceptualize what's happening? In other words, do you
> assume that for all people the use of "world model" isn't itself the
> metaphor? The use of a term doesn't necessarily mean people *really* think
> something like spatial movement is *really* taking place. I use the term
> "world model" and I can tell you that I don't think that.
>
> By the same token, if I play a flight simulator I don't *really* think I'm
> flying a plane, for example.

A flight simulator is *supposed* to be mimetic. It's supposed to look, more
or less, like you're moving through space. An IF game is diegetic. What is
happening on the level of interaction is not movement in space, but changing
glossaries.

> (When I played "Doom 3", I didn't think I was *really* a Space Marine who
> was *really* on Mars either.) But I accept the metaphor of a "world model"
> through which I'm flying a plane or blasting through the Legions of Hel.

In the case of a 3D-shooter, "world model" is not a metaphor. A 3D-shooter
is very much a world model, i.e. a simulation of physical reality.

> I suppose I could look at it more seriously and pull out a ton of academic
> concepts from literary theory.

Glossicism has nothing to do with literary theory. It's a theory devised
entirely by me with the aim of providing an anatomy of IF.

> Or I could just have fun with what is quite clearly and understandably a
> game.

Glossicism does not attempt to refute the fact that an IF game is a game. It
attempts to explain the mechanics of the game in non-simulationist terms.

> Where I think text-based IF is interesting is that it allows people to
> approach it with a more literary intent than flight simulators or 3D
> shooters, and which leads to these interesting differences of opinion
> among people and leads to the different emphases people put on what their
> expectations are.

This is not a question of different emphases. It's a question of theoretical
productivity. What has simulationism explained?


Jacek Pudlo

unread,
May 7, 2007, 6:52:41 AM5/7/07
to
albtraum

Two brown-eyed people give birth to a blue-eyed child. You can't explain
that from appearance alone, because there's nothing in the appearance of a
brown-eyed parent that says anything about whether they have latent genes
for blue eye pigment. You must go beyond appearance and investigate the
underlying genetic reality, because what we inherit is not the colour of our
parents' eyes but the genes that determine the colour.

Pudlian Glossicism is just that -- an attempt to map and explain the "DNA"
of IF, not the metaphors that make up its appearance.


Victor Gijsbers

unread,
May 7, 2007, 3:32:33 PM5/7/07
to
Jacek Pudlo wrote:

> All this coagulates into the illusion commonly known as "object." What you
> call "world model" is merely a superstructure composed of the above
> components. From a theoretical point of view, anything above the level of
> gloassaries is uninteresting. It would be like trying to understand physics
> in terms of "objects."

I very much doubt it. I have no quarrel with the theoretical tools you
present--they seem sensible enough--but I am dubious about the claim
that they are the only set of tools we need and should use.

It might be like trying to understand the world in terms of fundamental
physics: sometimes very useful, sometimes far worse than impractical.


But I doubt it's something we can discuss _in abstracto_, without any
concrete theoretical analysis to judge.

Regards,
Victor

Neil Cerutti

unread,
May 8, 2007, 7:26:47 AM5/8/07
to
["Followup-To:" header set to rec.games.int-fiction.]

On 2007-05-03, Jacek Pudlo <ja...@jacek.jacek> wrote:
> You can skim and pick through a novel, but you can play IF only
> in the *exact* sequence intended by the author. So
> paradoxically interactive fiction is less interactive than
> static fiction. A book can be read from cover to cover more or
> less attentively, whereas a work of interactive fiction is an
> obstacle run that demands the player's complete attention. A
> work of interactive fiction may frustrate you, it may infuriate
> you, but one thing it will never do is put you to sleep. In
> that sense, a work of IF is an extra-diegetic competition
> between game and player with the narrative the sham arena.

You see a book.
> TAKE BOOK
Taken.

> i
You are carrying:
a book

What's extra-diegetic about that?

There are plenty of extra-diegetic conventions in IF, e.g., "You
don't need to refer to that in the course of this game; SAVE; You
have attained the rank of pretentious butt," but generally the
whole *point* of the "narrative sham arena" is to be the
opposite.

--
Neil Cerutti
The Pastor would appreciate it if the ladies of the congregation would lend
him their electric girdles for the pancake breakfast next Sunday morning.
--Church Bulletin Blooper

Autymn D. C.

unread,
May 8, 2007, 10:25:41 PM5/8/07
to
Marks that don't belong in quotes go outside the quotes.

Reply all
Reply to author
Forward
0 new messages