FAQ (Frequently Asked Questions)
David A Graves
1) What is the purpose of this news group?
2) What topics are appropriate here?
3) What languages are available for writing text adventures?
4) Where can I read more about Interactive Fiction theory?
5) Who is doing R&D in Interactive Fiction?
6) What happened to that great publisher of adventures, Infocom?
7) What is IF anyway? What *are* you people talking about?
You may want to save a copy of this file for future reference, rather than
post a frequently-asked question. This file is updated regularly.
1) What is the purpose of this news group?
Rec.arts.int-fiction is a news group for discussion of Interactive Fiction
(IF). We do talk about adventure games here, but the discussions are from the
viewpoint of "the advancement of Interactive Fiction". Discussions range from
Adventure games and Interactive Fantasy to Hypertext, with emphasis on the
development of IF as a new literary genre and/or a new form of computer-based
art/entertainment. Discussions of adventure implementation languages is
appropriate in this group, as well as ideas on applying popular technologies
(such as Object Oriented programming) to problems in Interactive Fiction
development (such as environment representation, parsing, and natural language
generation).
2) What topics are appropriate here?
Topics related to Interactive Fiction design, theory, and implementation are
appropriate. Please don't post questions about specific adventure game
puzzles to this news group, as it was set up only for discussions of the
*development* of Interactive Fiction. Please post your queries about specific
adventure game puzzles to rec.games.int-fiction. Discussions of MUDs
(multi-user dungeons) belong on rec.games.mud. Info on LARPs (live-action
role playing games) can be found in rec.games.frp.*.
Posting to more than one group is not a good idea. Select the appropriate
news group and post only to that one. Just as you would not post questions
about specific games in this group, please refrain from posting questions on
IF design and implementation in rec.games.int-fiction.
Controversial viewpoints are sometimes posted here; when you post a dissenting
view, remember to attack the idea, not the person. Let us debate, not battle.
3) What languages are available for writing text adventures?
Here are the most frequently mentioned ones:
ADL: Adventure Definition Language by Ross Cunniff (cun...@fc.hp.com) and
Tim Brengle. Latest version is 3.2, June 1987. No object-oriented
programming, but still a robust and flexible system. LISP syntax.
Excellent parsing capabilities. System includes a debugger. Source
code (in C) is available. Package includes sample adventures, and
standard definitions for verbs. Runs on DOS, Amiga, Atari ST, Unix,
and VAX; could be ported to any platform. ADL is available from your
comp.sources.games archive. For example: ftp wuarchive.wustl.edu,
and get files under /usenet/comp.sources.games/volume2/adl
AdvSys: Adventure Authoring System by David Betz. Latest is version 1.2, July
1986. Written up in BYTE magazine, May 1987. This was probably the
most well-known adventure system, but recently seems to be eclipsed by
ALAN and TADS. LISP Syntax, fully object-oriented language. Source
code is available for compiler and interpreter (written in C), so the
language highly portable. Unrestricted for non-commercial use. No
debugger is available. Available on comp.sources.games archives.
Example: ftp ftp.uu.net, in /usenet/comp.sources.games/volume2/advsys
A library of standard verb and class definitions is available from
ftp.gmd.de as /if-archive/programming/advsys/standard.adi.*
AGT: Adventure Game Toolkit, by David Malmberg (73435...@compuserve.com)
and Mark Welch. Current version is 1.35, November 1991. (Soon 2.0).
Shareware, available on many BBS archives. Registration fee is $20.
Not an object-oriented language. Syntax resembles natural English.
Documentation (150 pages) states that no programming experience is
required. The standard definitions cover the ordinary commands, and a
meta-language is provided to create your own commands. Many games
have been written in AGT (over 35 are described in the documentation),
with source code available. AGT runs on MS-DOS, Macintosh, Atari ST,
and Amiga. Available from Softworks, 43064 Via Moraga, Mission San
Jose, CA 94539. Download free from Adventureland BBS, (606) 271-0558.
ALAN: Adventure Language, by Thomas Nilsson and Gorfo (th...@softlab.se,
go...@ida.liu.se), Sweden. Latest version is 2.4, Nov 1992. Syntax
similar to English. The main difference between ALAN and the others
is the non-programming design of the language. Not an object-oriented
language, but verbs attached to an object can override the global
definition of a verb. Source code not available. Manual includes a
tutorial on adventure game design. System is distributed free by
email from the ALAN owners only, no BBS or ftp sites. Send email to
alan-r...@softlab.se with a one line message of the form: SEND
<request>, where <request> = AMIGA, SUN, VMS, or PC (for complete
distibutions including executables, documentation and demos), TEXT or
POSTSCRIPT (for documentation only), or INFO (for an overview). A
port of ALAN to the MAC is planned.
OASYS: Object-Oriented Adventure System by Russell Wallace
(RWAL...@vax1.tcd.ie). Current version is 1.0. Written in C++. Not
a formal object-oriented language; procedures are called methods and
types are called classes, but it does not support inheritance,
encapsulation, or messages. Has documentation and two sample
adventures. Control structure is user-defined so fuses, daemons, etc,
are not a problem. System is public domain (no restrictions of any
kind on use or distribution). Runs on MS-DOS, could be ported to
others. Posted to comp.binaries.ibm.pc, vol 20, issues 081-083 in
1992; available from ftp archives of that group, or ftp from
ftp.gmd.de under /if-archive/programming/oasys (Zoo archive plus "the
missing file").
TADS: Text Adventure Development System by Michael Roberts.
(7373...@compuserve.com). Latest version is 2.0, October 1992.
Shareware. Object-oriented language, including multiple inheritance.
Syntax similar to C. Source code for TADS is not available.
Excellent documentation (including tips on adventure game design) for
those who register (which is highly recommended). Includes a library
of standard definitions of verbs and object classes. Large example
adventures written in TADS are available. The Unnkulian games were
written in TADS (contact is Dave Baggett, d...@ai.mit.edu; the
Unnkulian source code is not distributed). TADS runs on MS-DOS,
Macintosh, and Atari ST. Available from many ftp sites, including:
msdos.archive.umich.edu: msdos/games/adventure/tads.zip
atari.archive.umich.edu: atari/Games/Tads/tads.lzh
mac.archive.umich.edu: mac/game/gameutil/tads#.#.cpt.hqx
mac.archive.umich.edu: mac.bin/game/gameutil/TADS (via Appletalk)
If you can't download a copy, you get it from:
High Energy Software, P.O. Box 50422, Palo Alto CA 94303.
The $40 shareware fee covers the code and documentation.
Each of these systems contain a compiler and an interpreter for the compiled
"virtual machine code". All are text-only systems.
For a more detailed description of IF languages, plus advice on writing your
own adventure, see Nathan Torkington's "Adventure Authoring Systems FAQ"
posting, which appears here in rec.arts.int-fiction.
If you can't find what you want on a local archive, try ftp.gmd.de in Bonn,
Germany. Under /if-archive you can find IF languages, completed games, and
archives of discussions on rec.arts.int-fiction and rec.games.int-fiction.
They accept contributions in /tmp/if-uploads; if you do, notify bla...@gmd.de
explaining what it is and what it runs on.
4) Where can I read more about Interactive Fiction theory?
Some recommended background reading on Interactive Fiction includes:
Buckles, Mary Ann. "Interactive Fiction: the Computer Storygame
'Adventure'" (University of California at San Diego, 1985). Focuses only
on the original adventure game. Limited in scope, but a fun read for any
hard-core fan of Interactive Fiction theory.
Laurel, Brenda. "Towards the Design of a Computer-based Interactive Fantasy
System" (Ohio State University 1986). See also her "Computers as Theatre"
(1991, Addison-Wesley Publishing Co, ISBN 0-201-51048-0). This book extends
the work that Laurel began in her PhD dissertation, which addresses the
problems in automating plot which integrate the player's actions.
Meehan, James. "The Metanovel: Writing Stories by Computer" (Yale 1976)
Meehan used the natural language technology of Roger Schank's Yale group
to construct some primitive Aesop's fables.
Sloane, Sarah. "Interactive Fiction, Virtual Realities, and the
Reading-Writing Relationship" (Ohio State University 1991). Sloane presents
her views on narrative theory, as well as analyizing IF R&D in progress at
CMU (led by Joe Bates) and at Interactive Fantasies (Brengle and Graves).
These four PhD dissertations are available for ~$50 each from University
Microfilms, 800-521-0600. (Half price for students).
The following are also recommended:
Lebowitz, Michael. "Creating Characters in a Story-Telling Universe"
Poetics, 13, 171-194. (1984). Poetics is a periodical; check your library.
Thurber, Macy & Pope. "The Book, the Computer and the Humanities"
(Aug '91 issue of T.H.E. Journal (Technology in Higher Education)
Discusses a project to foster critical thinking using a computer with
hypertext and interactive elements becomes a humanistic new medium.
References on "rules" for story generation. Most of these authors never
guessed computers would exist, but that shouldn't stop you from using their
work as the start for your "AI rule base" for plot unit integration.
Aristotle. "The Poetics". Translated by Ingram Bywater. In "Rhetoric and
Poetics of Aristotle". New York: The Modern Library, 1954. Aristotle
defines the basic elements used to construct stories and drama, although
his model is frequently attacked by modern narrative-theory researchers.
Campbell, Joseph. "Hero with a Thousand Faces" (1949). New Jersey:
Princeton University Press. Campbell defines the elements which are common
to heroic folktales in all cultures, forming a single template, which is
called "the monomyth".
Polti, Georges. "The Thirty-Six Dramatic Situations" (1916). Boston: The
Writer, Inc. (617) 423-3157. ~$10. This is really quite unique and
brilliant, identifying a truly interesting and challenging subset of
story-outlines useful in developing a plot knowledge base.
Propp, Vladimir. "Morphology of the Folk Tale" (1968). University of
Texas Press, Austin. Written about the same time as Polti's analysis.
ART COM (electronic) Magazine #43 and #44 (Nov & Dec '90) were devoted to
Interactive Fiction. These two back issues are available by e-mail. Send
requests to d...@hpsemc.cup.hp.com, with the one-word Subject: ArtCom.
Requests are processed by a program; append a personal message if you wish.
I have published a few papers on Interactive Fiction technology, which are
available by e-mail. They are: "Second Generation Adventure Games" (which
focuses on the physical world model, parsing, text generation, and simple
agent planning), "Bringing Characters to Life" (which summarizes the progress
in Artificial Personality over the last two decades), and "Plot Automation"
based on my presentation at the Computer Game Developer's Conference in spring
of '91. To receive all three papers, send mail to d...@hpsemc.cup.hp.com, with
the one-word Subject: Papers. Requests are processed by a program, but I do
review personal messages appended to requests.
The Journal of Computer Game Design occasionally prints articles related to
Interactive Fiction. Subscriptions are $36 per year (a non-profit publication)
for six issues. Write to 5251 Sierra Road, San Jose California, 95132.
The editor welcomes articles from the readership; contributors receive a free
one-year extension on their subscription.
5) Who is doing R&D in Interactive Fiction?
The Oz Project, directed by Joseph Bates at the Carnegie-Mellon School of
Computer Science, is developing technology for high quality interactive
fiction. Their goal is to provide users with the experience of living in
a dramatically interesting simulated world that includes simulated people.
Their focus is on the simulations behind the interface, which they call
the deep structure of virtual reality. A good first reference for their
work appears the journal PRESENCE:
Bates, Joseph, 1992. Virtual Reality, Art, and Entertainment.
PRESENCE: Teleoperators and Virtual Environments 1:1, 133-138.
MIT Press. Cambridge, MA.
The article's bibliography contains references to their more technical papers.
In 1989, the Oz folks conducted a search for those doing research and
development in Interactive Fiction. Many individuals responded, stating their
interest in the results, but the search revealed only the Oz project itself,
and the team of Brengle and Graves. (See references above). Many others are
developing IF using traditional methods, of course.
6) What happened to that great publisher of adventures, Infocom?
Around 1986/87 Infocom was having serious financial trouble. Their database
product, Cornerstone, was expensive to develop but was not selling.
Activision then bought Infocom. Infocom was relocated to Mountain View,
California, but most of the staff did not stay with the company. Infocom
became a label, and Activision went through a radical reorganization.
Finally, Mediagenic was formed. Mediagenic was (and still is) the parent
company of Activision and Infocom. You may see new products come out under
the Infocom label, but the original writers have moved on.
7) What is IF anyway? What *are* you people talking about, anyway? Here are
some definitions related to IF:
User Paced Sequence: A linear sequence of fixed content.
Progressive Disclosure: Content is fixed, but exposure to the content
varies.
Point of View Shift: The first-person viewpoint moves between characters.
Browse Around: The player wanders and explores an environment.
Plot Branching: A tree or network of fixed content.
Physical Modeling: The system simulates a physical universe with which
the player may interact.
Character Modeling: The system simulates characters with whom the player
may interact. Characters may generate goals, actions, and emotions.
Character modeling may be broken down into intelligence modeling and
emotion modeling.
Intelligence Modeling: Rules for simulating intelligent behavior by
characters. This may include setting goals and making plans to
achieve them. Since intelligence modeling is typically focused on
problems of logistics, it is tightly coupled with physical modeling.
Emotion Modeling: Representation of emotions as data, and rules for
processing that data to derive behaviors consistent with a character's
perceived personality. Emotion modeling is typically focused on the
feelings associated with interpersonal interaction.
Dramatic Modeling: The system has a representation for dramatic elements
(plot fragments) and a "plot calculus". This allows the system do
plot generation on the fly -- taking into account the actions of the
player character.
Computer Adapted Story Telling: A presentation of different but consistent
experiences of the same story. Could be achieved through point of view
shift, browse around, or plot branching.
Interactive Fiction: Narrative based experiences that tend to be either
puzzle solving or plot branching. This term is usually applied to
the "first generation" Adventure games -- those developed in the last
decade. The second generation of IF is moving towards the development
of Interactive Fantasy as a new genre.
Interactive Fantasy: A first person dramatic experience. Achieved through
a combination of physical modeling, character modeling, and dramatic
modeling. (Consider the Star Trek Holodeck as a vision of a future
Interactive Fantasy platform).
Interactive Fiction is difficult to define concretely since it is a new
artistic form, still in its infancy. The first work of computer-based IF was
a story-game called "Adventure". To this day, games of this type are called
Adventure games, named after the original instance.
There are many forms of Interactive Fiction, but the one thing they have in
common is that the reader is allowed some degree of interaction with the
story. When we talk about IF in this news group, we usually are talking
about computer-based works of fiction. A traditional book is not interactive
-- you just read it from front to back, and get the same experience every time.
Pick-a-path books, however, are interactive; this is probably the lowest form
of IF. One goal of IF developers is to take advantage of the flexibility of
the computer to facilitate the creation of new forms of entertainment.
Adventure games are an early form of computer-based IF. They are *subjective*
IF, that is, the player has an influence on the "plot" of the story. The
reader can influence events via his choices about what to do next, the
ordering of his actions, etc. In *objective* IF works, the reader has some
influence on the presentation of the story, but not the content. Consider a
hypertext-based story, where you can ask for more information on a given
person or plot event, but you cannot influence the flow of events. Infocomics
are an example of objective interactive fiction.
IF offers great potential, but since its appearance about a decade ago, its
growth has been plagued by two problems: how to develop the computer
technology required to support a work of Interactive Fiction, and how to
develop stories that exploit this new genre.
Interactive fiction differs from traditional fiction in that the author gives
up much of the control of the story flow. This is because the reader (or
player) is allowed to participate to some degree in the shaping of the plot
through his role as a character in the action. Since the player/protagonist
will be making decisions about what he will do next, the author must allow for
multiple paths through a set of plot potentials. The most primitive way of
doing this is through plot branching: presenting the reader with a small set
of fixed choices, each set corresponding to a branch in a fixed set of
potential plot paths. Unfortunately, this technique is intrinsicly limited
and has historically resulted in relatively uninteresting games.
A more interesting approach (in my opinion) is to create a rich set of plot
fragments and character behaviors which may be assembled by the computer to
allow the creation of new stories each time the program is used. In the
finished product, the individual elements of the story can combine in new and
wonderful ways not anticipated by the author or programmer.
In this news group, we discuss the technical and artistic aspects of the
interactive fiction genre. While we occasionally do mention "off-the-shelf"
IF products, it is typically in the context of comparing and contrasting their
structure or artistic merit.
----------------------------------------------------------------------------
This file of Frequently Asked Questions (FAQ) is maintained by David Graves,
d...@hpsemc.cup.hp.com. Corrections and suggestions are welcome.
David Whitten
may be a scary term for some people. (Of course, other people may not even
care....)
Anyway, the way I read this term is that a Plot Calculus is a term for a
formal method to calculate a plot, and provide some analysis of a plot
as a logical object, rather than making it implicit in all the interaction
between all the elements that make up a particular work of IF.
Presumably a Plot Algebra would be simpler than a Plot Calculus, since it
could just describe the static portion of an IF work, rather than including
the dynamic portion.
Has anyone written any formal work on Plot calculus?
Dave (whi...@fwva.saic.com) US:(619)535-7764 [I don't speak as a company rep.]
David A Graves
"plot algebra" might be more comfortable for some people. The way I see it, a
plot calculus this would be a set of rules for manipulation of symbols
representing plot fragments. In a simple implementation it could be a formal
grammar, but I shy away from getting too deeply into a yacc-like grammar
implementation because the stories generated would be so predictable.
(Propp's analysis of the structure of folktales is of this structure. You
could define a yacc grammer which would recognize hundreds of plots within
Propp's vision of story structure). I prefer methods which take into account
some of the semantics of the situation. This means that you can't just use a
formal grammar. Thus, the term "plot calculus" seems to apply to my view of
plot representation and processing. Perhaps a more modern term would be "rule
base".
There are a number of references that focus on various peoples vision of what
a plot calculus would be, although few of the authors use that term. I direct
you to Propp, Ploti, Meehan, and Laurel. These references are spelled out in
the FAQ (Frequently Asked Questions) file in this news group.
Ben Harrison
>[...] Topics related to Interactive Fiction design, theory, and
>implementation are appropriate. [...] Discussions of MUDs
>(multi-user dungeons) belong on rec.games.mud.
My question is simple, is it not appropriate for discussion of
programmable MUDs (ie LpMuds) to take place here? At one point, for
example, I know that someone was trying to imbed Zork into an LpMud.
How, I wonder, does the concept of multiple self-aware (ie, humans)
agents interacting with the same "story" change things? Does IF take
account of any actions, in fact, not directly induced by the "reader"?
Or if not multiple self-aware agents, what about atomated, but yet
self-controlled agents, like monsters? Or even non-agents, lile
random events, probablistic reactions to commands, etc.
Multiple "readers" seems to be an obvious extension of the IF concept,
unless we are restricting ourselves to personal computers, or to
other implicitly "single-user" media. Or has the "single-reader"
paradigm been simply "carried along" from the implicit single reader
in a normal story (discounting over-the-shoulder readers)?
I personally am fascinated by interactive fiction, especially in terms
of the application to games like the Infocom games, and the variations
like the simpler-grammar-but-more-powerful-graphics games like the
ones Sierra put out, but this fascination is dwarfed by that I feel
for "stories" in which not only do you have to worry about "automated"
entities with predictable actions (like monsters), but also other
entities controlled by "sentients" (like other human readers).
Of course, without substantial hacking (of code and philosophy), it
does not seem likely that a non-Unix (or equivalent) or non-internet
machine could usefully support suuch a "story"...
Or has this entire thread already been hacked to pieces?
--- Ben Harrison ---
Adam Justin Thornton
it recommends that only one be interactive this may not be a strict require-
ment. I don't know how TADS works internally or how much of a fix it would
take to have several BasicMe objects around.
It would obviously be stupid to put all these on the same console. But for
under $1000 you can have a 386sx running Linux with four old DEC VT220s hooked
to its serial ports; this would be more than adequate to run an interactive
adventure.
The other problem is that storylines--indeed, the very concept of a plot--would
need to be reworked if there are two players competing for the same goals.
One could really hose the other if puzzles are not carefully designed.
This would also seem to be a natural for a naive physics implementation. For
instance, how loud something the other player does will vary with how far
away the first is and what is between them.
The problem is, though, that without certain well-defined puzzles, what you
will end up with is a MUD like structure. With certain well-defined puzzles,
you have a situation where the players have managed to each put the other in
a no-win situation.
Adam
--
"And in the heartbreak years that lie ahead, |++| ad...@rice.edu |++| Cthulhu
Be true to yourself and the Grateful Dead." --Joan Baez | 64,928 | fthagn!
"Very often, a common stone, thrown away and despised, is worth more than
a cow." -- Paracelsus | If these were Rice's opinions I'd shoot myself.
JOSEPH OSWALD
|>
|> The other problem is that storylines--indeed, the very concept of a plot--would
|> need to be reworked if there are two players competing for the same goals.
|> One could really hose the other if puzzles are not carefully designed.
|>
....
|>
|> The problem is, though, that without certain well-defined puzzles, what you
|> will end up with is a MUD like structure. With certain well-defined puzzles,
|> you have a situation where the players have managed to each put the other in
|> a no-win situation.
|>
I think a far more interesting idea would be having puzzles that require
cooperation and communication. Using a Zork-like example, needing help with
ropes to scale a cliff to get to a ledge with a "prize." And that prize, a magic
chest, might need the character with a certain past history or ability to open it.
And so forth.
Come to think of it, there's a lot of opportunity. If you like _Suspended_, think
if all the robots were actually controlled by someone on another console, so that
one could avoid laboriously instructing each robot step by step, but be able
to evoke discussion: "Say, Auda, do you think you can do anything to delay those
guys who keep coming in to unplug me?"
Of course, I don't know much about the MUD thing, beyond "multi-user dungeon."
Has anyone written a MUD or multiple-actor IF piece that goes for this sort
of cooperation? Is it the usual approach?
--Joe
Jamieson Norrish
OSWALD) writes:
I think a far more interesting idea would be having puzzles that
require cooperation and communication. Using a Zork-like example,
needing help with ropes to scale a cliff to get to a ledge with a
"prize." And that prize, a magic chest, might need the character
with a certain past history or ability to open it. And so forth.
Of course, I don't know much about the MUD thing, beyond
"multi-user dungeon." Has anyone written a MUD or multiple-actor IF
piece that goes for this sort of cooperation? Is it the usual
approach?
In most "competitive" (as opposed to social) MU*s, cooperation is not
stressed, and puzzles typically take the form of individual "quests".
I think this could reasonably said to be the standard approach, since
MU*s must cater for those players who log on when no one else is
around.
Having said that, it is entirely possible and easy to design puzzles
in MU*s which require multiple characters to solve/complete. Most MU*s
that I have seen, unfortunately, only promote cooperation as a means
of survival against hordes of mindless numbers (otherwise known as
NPCs :-), rather than in puzzle situations. That is not to say that IF
fans must discard the MU* approach entirely; I myself am working on
a very interactive MUCK, which might enventually provide a medium for
player cooperation in puzzles (although I admit I am biased more to
the roleplaying side of it, rather than puzzle-solving).
Does this help at all?
Jamie
Jacob Solomon Weinstein
>
>The other problem is that storylines--indeed, the very concept of a plot--would
>need to be reworked if there are two players competing for the same goals.
>One could really hose the other if puzzles are not carefully designed.
>The problem is, though, that without certain well-defined puzzles, what you
>will end up with is a MUD like structure. With certain well-defined puzzles,
>you have a situation where the players have managed to each put the other in
>a no-win situation.
Well, as you noted, the solution is to rework certain notions in IF. I
would say plots could be similar, but the puzzles would need to be
changed. Ideally, the players would have to work together. Consider the
following example, where we have two players, a and b:
You are in a room. A single door stands east.
PLAYER A>east
The door slams moments before you pass through it, swinging open again
only after you take a few steps backward.
PLAYER B> hold the door open
You stand behind the door and hold it tightly.
PLAYER A>east
As you pass through, the door trembles, but player B holds it tightly
in place.
It's not a great example, but I'm sure you see the potential that two-
player puzzles have. I think that a good two-player IF game is quite
feasible. I would think, though, that including an unlimited number of
players would almost invariably tend towards a MUD-like situation,
unless one were willing to create a highly detailed world. If one were
willing to do that, the players might have to agree on an overall
strategy, then spread out to various regions to accomplish their
sub-goals, rendezvousing at, say, the castle of the evil warlord.
Michael J. Roberts, if you're reading this: have you considered adding
multi-player capacity to TADS? It would require a large amount of
effort, I imagine, but it would be an incredible advantage.
Rusty Wright
guess is that the two won't really mix (multi-player IF). The problem
is that when you get multiple players, there's a very strong
inclination to just sit around and shoot the breeze. There's
something seductive about sitting in front of your
terminal/workstation in a text-based virtual reality room with people
from all over the country and globe. Even if you're not much of a
talker, it can be fun just watching the other conversations. There's
also a different set of ethics of behavior; just as people in cars are
less inhibited about being rude, in multi-user virtual realities
people are even more less inhibited because you know the other person
is miles away, and on top of that, your real identity isn't known (you
get to make up your player's name and description). It's an area ripe
for research by the sociologists.
I think it boils down to the the problem that living, talking people
are invariably more interesting than some interactive fiction puzzle
or story. Until we get some really good AI in our IF software, the IF
will always be more static in it's behavior, and therefore less
interesting, than people.
But I don't want to be a wet blanket; I still think people should try
because I'm sure some interesting things will come about from it. And
99% of the text on MUDs (scenery, descriptions of things, etc.) is of
such a mediocre level of creativity and originality that it would be
nice to even inject some new blood, so to speak. And maybe I'm just
plain wrong; maybe the reason MUDs haven't had any good IF for them is
that the vast majority of their players are people who are basically
uncreative and unoriginal. I do remember some quite creative MUD
players, but for some reason their stuff never "caught on." Maybe it
it just got drowned out by all the drek produced by the other players.
Paul Christopher Workman
plot--would
>need to be reworked if there are two players competing for the same goals.
>One could really hose the other if puzzles are not carefully designed.
>The problem is, though, that without certain well-defined puzzles, what you
>will end up with is a MUD like structure. With certain well-defined puzzles,
>you have a situation where the players have managed to each put the other in
>a no-win situation.
Has there ever been a good analysis of what plot is under interactivity?
The standard for IF seems to be a string of puzzles, to be solved more
or less sequentially, with text distributed between the puzzles. This
is easily conceptually parsed: text-puzzle-text-puzzle.....puzzle-text-YouWin.
(Which makes me wonder if anyone has ever made a parser/compiler for
IF which takes a text file -- I mean actual literary text and not
program code -- and produces a game as output. But I digress.)
But this is a far cry from the plot used in actual fiction. Certainly the
quality of literary elements varies, generally, with the quality of
the literary work itself (although the plot to a Mickey Spillane novel
might be more complex than that of "The Great Gatsby," the nature of
Fitzgerald's plot aspires to more), but in general the plot, or any
literary element, from fiction should allow the author to communicate
an idea or effect that's greater than just the sum of its (the element's)
parts. Interactivity, one would think, abandons that since it abandons
the author's control over the plot; and the traditional IF-game
approach mentioned above, although it does allow tight control over
the progression of the plot (since the user can't, usually, go to
the next step until the puzzle preceeding it is completed), reduces
the plot *too* far, I think -- the plot is interactive only to the
extent that the user can solve puzzles. That's not really what
your average fiction writer thinks of when you say "plot." I've heard
reports of non-game interactive fiction (usually with respect to
StorySpace), but I've never had the opportunity to see it close up,
and anyway the reports suggest that the fiction produced is really
more like meta-fiction, which is fine for literary critics but not
much else.
Multi-user IF seems integral to the idea of IF, implicitly if not
explicitly. (If an IF plot consists of the interplay between A, B, and C,
it shouldn't be a big difference if only A is human or if both A and B are.)
I would expect that if the true-plot-in-IF problem is solved, then
the multi-user IF problem would be solved immediately, as a result.
--paul
David Whitten
>Has there ever been a good analysis of what plot is under interactivity?
>but in general the plot, or any
>literary element, from fiction should allow the author to communicate
>an idea or effect that's greater than just the sum of its (the element's)
>parts. Interactivity, one would think, abandons that since it abandons
>the author's control over the plot; and the traditional IF-game
>approach mentioned above, although it does allow tight control over
>the progression of the plot (since the user can't, usually, go to
>the next step until the puzzle preceeding it is completed), reduces
>the plot *too* far, I think -- the plot is interactive only to the
>extent that the user can solve puzzles. That's not really what
>your average fiction writer thinks of when you say "plot."
>Multi-user IF seems integral to the idea of IF, implicitly if not
>explicitly. (If an IF plot consists of the interplay between A, B, and C,
>it shouldn't be a big difference if only A is human or if both A and B are.)
>I would expect that if the true-plot-in-IF problem is solved, then
>the multi-user IF problem would be solved immediately, as a result.
>
>--paul
>
>
characters, but I think the problem is that we have no way of specifying the
limitations on character's actions. (except keeping the parser from
recognizing the words)
This is a pretty poor method of limitation.
IMAO, I feel we should have ways of specifying what kinds of actions are
'in character' and which ones aren't.
This means we can allow more than one story character to be played by a real
person. I think this means that when we have people in the stories, we need
to specifically consider what roles the people play (and probably have a
library of roles to make writing IF easier)
What I mean is that if I write a story with a taxi, I have a person in the
story who drives the taxi. Now, the taxi may only be an insignificant part
of my story, suitable primarily to transport my primary character from one
part of the city to another.
It is concievable that a real person could play the part of the taxicab
driver. This would be interesting from a development point of view since
it is easier to get the text a program-emulated-person in a role says by
having a real person play the role for a while, and just to record what they
said. Then go back and incorporate the kinds of things they say into the
game...
Anyway (I digressed), If I let a real person play the taxicab driver, I want
to put severe limits on their actions, since I need a taxicab several times in
the story. I might let them talk, and to drive forward, but I probably
wouldn't want them to leave the taxi, nor would I want to let them take the
taxi on a scenic tour of the nation, since I as Author, need the taxi to be
available for my primary character when they need it.
In the same way, if my primary character is supposed to fall in love or deal
with the relationship with some secondary character, I as Author have limited
what kinds of relations are reasonable. The player can't just try to deal
with anyone who just happened to walk by.
I know it is a big project to talk about all the roles a person may play,
but that is the essense of writing fiction rather than just providing an
interactive chat board.
I think that some of the most interesting fiction is that which is written
with some pretty severe restrictions, and still manages to come to a satisfying
conclusion.
(I guess that is the difference between the generic plot:
1) "boy meets girl"
2) "boy loses girl"
3) "boy regains girl"
and a real work of fiction based on that plot.
David (whi...@fwva.saic.com) US:(619)535-7764 [I don't speak as a company rep.]
JOSEPH OSWALD
(Rusty Wright) writes:
|> As one who has played on MUDs for several years, and who loves IF, my
|> guess is that the two won't really mix (multi-player IF). The problem
|> is that when you get multiple players, there's a very strong
|> inclination to just sit around and shoot the breeze.
Well, I guess I can understand this, but it would help if someone could e-mail
me some info on exactly what a MUD is, and how it works, either in a theoretical
sense or a real example.
|>
|> I think it boils down to the the problem that living, talking people
|> are invariably more interesting than some interactive fiction puzzle
|> or story. Until we get some really good AI in our IF software, the IF
|> will always be more static in it's behavior, and therefore less
|> interesting, than people.
|>
....
This indicates to me that the problem might be that an essentially un-regulated
number of people can get on board (like xtrek), so that roles are essentially
all equal and also very general--you can't count on having a certain number of
people around, so you can't have a certain number of roles that need to be
filled. It's a lot easier to write a multi-player game that way. 1000 looters-
and-pillagers in a game have very similar effects to 100 looters-and-pillagers,
just more so. This may be my misunderstanding of MUDs, but it's one possibility.
My thinking was of a fixed number of players (perhaps not all needed at the same
time, but just periodically logging in to catch up on their thread) each with an
assigned persona that is part of the game. To use a _Suspended_ analogy, one
person is Auda, the listening robot, and one person is Whiz, the computer-
interfacing robot, etc. Perhaps the character one plays in the one-player
version, the central directing intelligence, would NOT be one of the players,
and the organization and direction would have to arise as cooperation between
the players, none of whom alone could solve the puzzle.
Another problem could be the means of communication, and how aware one is of the
other characters being around. If you are constantly aware of where everyone is,
or can talk to anybody on demand, it might degenerate more easily than if you're
not as in touch.
Of course, this issue of communication would be pretty central to the idea of
multi-player IF. In _Suspended_ (I sure hope most people understand this game
at least in concept, if not in play), the central director interacts solely
by a "radio" connection that lets him assume the usual IF role of one robot.
That means that one can always tell, on demand, where Auda is, what he hears,
and what journey you sent him on, if you did so.
If multi-human player IF works like this, with essentially unlimited contact,
then it might degenerate. BUT, if you could limit it in some way, like
"person is in the room to talk, or you have picked up the communicator and he
has the other one, or you have sent a message to his computer terminal
(fictional, not physical), and he has checked his (fictional) e-mail account"
perhaps it would be more interesting.
Note, that any limitation could be potentially overcome. If four people are all
wired in to one PC box on 25' serial lines, then there's going to be a lot of
table talk. In xtrek (a X-window shoot-em-up with ships logged on from possibly
all over the country), two ships on consoles right next to each other in a lab
are on a different relationship than "Joe from Duke" and "Max from MIT", although
if the game were slow-paced, e-mail would let Joe and Max communicate equally
well.
This would be rather difficult, then--if one has limited communication to help
the plot, then this extra-game communication would overcome the limitations,
allowing "cheating:" Joe on Planet X can talk to Max, on Planet Y, just by
calling across the computer lab (which xtrek players do, to much annoyance to
people trying to do Comp.Sci. homework), or sending him e-mail on the net,
even if "communications" between planets X and Y are disrupted by an "ion
storm"
On the other hand, if GOOD communication is needed in the game, as in a battle
situation, or a run-away robot that needs to be stopped before he throws himself
in an acid bath, then how do you do that, if it is potentially being played
over the net, or if Max may have logged off to do real work?
A difficult balance--technical communication issues, uncertainty of whether
players may be more able to talk than you think, and how to write a plot that
is fun (or at least frustrating in the entertaining sense) in whatever mode
of communications is REALLY in effect.
I've raised a lot of points here, (and used a lot of net.bandwidth)
respond/flame as desired.
*please, remember to fix my ignorance of what multi-player stuff has been
already accomplished.
--Joe
JOSEPH OSWALD
|> I think Paul is right. IF seems to be tied to the idea of interacting
|> characters, but I think the problem is that we have no way of specifying the
|> limitations on character's actions. (except keeping the parser from
|> recognizing the words)
|>
|> This is a pretty poor method of limitation.
|>
|> IMAO, I feel we should have ways of specifying what kinds of actions are
|> 'in character' and which ones aren't.
|> What I mean is that if I write a story with a taxi, I have a person in the
|> story who drives the taxi. Now, the taxi may only be an insignificant part
...
|> Anyway (I digressed), If I let a real person play the taxicab driver, I want
|> to put severe limits on their actions, since I need a taxicab several times in
|> the story. I might let them talk, and to drive forward, but I probably
|> wouldn't want them to leave the taxi, nor would I want to let them take the
...
Not to criticize, but to get a better idea of what would be involved in a
real work of multi-player IF
A taxi-driver who was necessary to provide transportation at several points
in the game might not be a good player--his role would have to be only to
respond to his dispatcher, go to that place, pick someone up, and deliver him
to the requested destination. Not too much fun, which is why most people don't
drive cabs without being paid (although some cabbies have more fun than others
:-). Also, think of how many times one uses a part of the game when still
in the haven't-really-figured-out-what-I'm-doing stage that arises whenever
you're stumped by a new development. Imaging Joe taking you repeatedly from the
bank at 5th and 32nd to his apartment workshop at 3rd and 103rd (and
back) because he doesn't anticipate what tools he needs to break into the
bank safe.
Just thought of this--in an Infocom-style game, I often save the game at
a certain point so that I can try several options, restoring each time one of
them doesn't work out. How does that sort of saving affect the status of other
people in the game, and what implications does that have for the kind of
trial-and-error play that is often a part of Infocom puzzles?
I think part of this is your implicit (and explicit) classification of characters
into "my primary character" and service characters "my taxicab driver." I think
equality of necessity is required among these players, though not necessarily
identical roles. Think of _It's a Mad, Mad, Mad World_ (or any ensemble picture,
for those of you who haven't seen the film)...several sets of characters, all
pursuing their own way in one film. None of those actors had a particularly
better role than the others.
Of course, that gets back to the original problem--in the movie, the screen-
writer can FORCE one character to drive up just as another character finds
a lost treasure, or whatever....That's how farce works--forcing what would
be otherwise unlikely scenarios. But if the one character is left behind because
he couldn't figure out how to open his (game) bedroom door this morning, because
the landlord character forgot to have it fixed....then the whole thing collapses.
Again, it's very hard for me to think of an example of a plot with essentially
1) independently thinking
2) independently acting
characters going through "plots" with
3) independent results
but still remain within a plot structure thought up in advance by one person
and still require enough
4) co-dependent participation/cooperation/competition (I like the first 2 better)
that it makes sense to have multiple people playing!
Of course, being hard is no reason not to think about the problem--if I
or anyone else can possibly come up with a good one, I'd like to hear it, and
more importantly, PLAY it!!!
--Joe
Mike Roberts
In article <1993Jan20.1...@Princeton.EDU> jac...@phoenix.Princeton.EDU
writes:
> Michael J. Roberts, if you're reading this: have you considered adding
> multi-player capacity to TADS? It would require a large amount of
> effort, I imagine, but it would be an incredible advantage.
We'd like to do it at some point, but I don't anticipate having time
to think much about it in the near future.
--
Mike Roberts mrob...@hinrg.starconn.com
High Energy Software 415 493 2430 (Voice)
PO Box 50422, Palo Alto, CA 94303 415 493 2420 (BBS)
"Time to ring in the new year, and watch as it transforms itself
from a beautiful promise of tomorrow into the ugly reality of today."
--- Lars Fusco
Alejandro Rivero
->Not to criticize, but to get a better idea of what would be involved in a
->real work of multi-player IF
->
->
->Just thought of this--in an Infocom-style game, I often save the game at
->a certain point so that I can try several options, restoring each time one of
->them doesn't work out. How does that sort of saving affect the status of other
->people in the game, and what implications does that have for the kind of
->trial-and-error play that is often a part of Infocom puzzles?
This is not really a problem, each player saves his status and the server machine
stores the world status.
->Of course, that gets back to the original problem--in the movie, the screen-
->writer can FORCE one character to drive up just as another character finds
->a lost treasure, or whatever....That's how farce works--forcing what would
->be otherwise unlikely scenarios. But if the one character is left behind because
->he couldn't figure out how to open his (game) bedroom door this morning, because
->the landlord character forgot to have it fixed....then the whole thing collapses.
->
I think this FORCE is typical in all classic plots. It is called FATE. Think of
Edypus (edipo), by example, or Macbeth...
Hmm this point could be worked a little.
->Again, it's very hard for me to think of an example of a plot with essentially
->1) independently thinking
->2) independently acting
->characters going through "plots" with
->3) independent results
->
->but still remain within a plot structure thought up in advance by one person
->and still require enough
->
->4) co-dependent participation/cooperation/competition (I like the first 2 better)
->
->that it makes sense to have multiple people playing!
I think of an *really classical* example: Odisea, from Homerus.
(which I suposse actually not copyrighted, so if someone want to try
a implementation, feel free... :-)
Main players are Ulysses and Telemacus.
One stars from egan sea, perhaps traped in Cyclops cave; the other starts from
Itaca, in Ulysses home. So (1) and (2) are fullfilled
going for (3), minimun requeriments are different for each player.
Ullysses goal is return to Ithaca. Telemacus goal is finding Ulysses, dead or alive.
as for (4) we can observe that the happy ending -which could be not
necessary, thought on examples above- condition, or *aditional bonus*, implies
the cooperation of both character. They both want to clean the pretenders
out of the house, eventually kill them. but here comes the puzzle:
Telemacus is young, have not authority, and not the power to fight alone.
Ulysses can not enter in his own house, it coulde be atacked by the pretenders
before he were unable to get his bow (and sword, etc) which are in house,
waiting for him.
I refer you to Homerus to see as the puzzle is solved.
-Alejandro Rivero
Theoretical Physics
Zaragoza Univ, Spain
Disc: ! I have not relation with Beyond Juslibol authors!
Paul Christopher Workman
>1) independently thinking
>2) independently acting
>characters going through "plots" with
>3) independent results
>
>but still remain within a plot structure thought up in advance by one person
>and still require enough
>
>4) co-dependent participation/cooperation/competition (I like the first
2 bette
r)
>that it makes sense to have multiple people playing!
One could design a puzzle-oriented game, in which (1) several different
methods can be used to get to a particular point in the game (eg, you
could get to the secret laboratory either from the secret passageway
in the bookcase or by asking the imprisoned evil scientist for the keys
after bribing the guard), and (2) some or perhaps most puzzles require
the cooperation of two or more players (eg, one player twists the
candleholder, causing the bookcase secret door across the room to swing
open, and another player shoves a chair into the secret doorway to
stop the bookcase from closing so you can get through). You'd need some
sort of common goal to provide competition. You'd probably also need
inter-player conflict (eg, one player could kill another. Then you
might get situations where one group of players spends their money
on spelunking gear to retrieve the chalice at the bottom of the cave,
but another group spends their money on guns to kill the first group
when they come out of the cave!). I think that this sort of game would
end up being closer to a MUD set in a predefined, unalterable universe,
than to IF as built from the available authoring systems (from what I've
heard. I haven't tried building one yet.).
This sounds like it could be a lot of fun, but it is just ducking the
fundamental problem of interactivity vs plot. The phrase "plot structure"
above helps show the conflict: the interactivity destroys any pre-designed
plot structure.
So now it is necessary to come up with a new definition of plot, one
which doesn't require a specific plot structure to get the required
effect that the author wants. The key is the author's desired effect;
a book with two different effects due to plot changes, is really two
different books. The new definition of plot should be intuitively
equivalent to the current one; that is, they should "feel" the same,
I think -- at least this might give us a clue as to how to proceed.
--paul