[general] Totally Unbelievable Game -- design criteria
Jim Aikin
Someone has recently released a Totally Unbelievable Game (text-based,
of course). Word on the street has reached fever pitch, and the servers
are chugging trying to keep up with the download requests.
My question is: What are the features of this game?
Of course it would have a great story, that's essential. But what other
features would it have? I can think of five or six things, but here's
one to start the discussion:
The parser would have to be much more flexible and "intelligent" than
current parsers. Current IF parsers embody some fairly stringent (and
hidden) assumptions about the input. People who know IF understand these
assumptions; others don't.
We could make a long, long list of grammatical constructions novice
players might try: “is david smiling?” “[tell david] grandma fell down
the stairs and broke her ankle.” “I want the ice cream cone.” “I want
david to give sheila the ice cream cone.” “ask sheila if david is
kidding about grandma falling down the stairs.” In a game that achieves
runaway success, I believe the parser would need to be able to handle
all of these structures (and hundreds of others) gracefully.
The full list of needed grammatical input structures would have to be
arrived at through extensive testing. That's a separate issue, and
doesn't concern me at the moment. The question is, what _else_ would
this Totally Unbelievable Game have that currently available games don't?
--Jim Aikin
Andrew Plotkin
>
> The full list of needed grammatical input structures would have to be
> arrived at through extensive testing. That's a separate issue, and
> doesn't concern me at the moment. The question is, what _else_ would
> this Totally Unbelievable Game have that currently available games don't?
Are you interested in a list of wishes, or a list of goals for current
designers to take into account? Because your first entry sounds more
like the former than the latter.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
"Bush has kept America safe from terrorism since 9/11." Too bad his
job was to keep America safe *on* 9/11.
Jim Aikin
>
> Are you interested in a list of wishes, or a list of goals for current
> designers to take into account? Because your first entry sounds more
> like the former than the latter.
The former, I think. What I didn't quite make clear (because this post
was an outgrowth of another thread) was that I'm interested in what
might be needed for an IF game to reach a MUCH larger audience.
Goals for current designers to take into account are of course
interesting too. But what I'm envisioning is a fantasy in which a
hundred thousand people who have never tried IF before are downloading
this game. What would need to be in that download for them to get
excited and recommend it to their friends in breathless terms?
I mentioned a more flexible parser. Here's another item on my wish list
(which would clearly not apply well to folks who are using the current
tools):
The distinction between the interpreter and the game software would have
to go away. Those menu headers at the top of the window? They would have
to refer, only and always, to the game, NOT to some separate layer of
software that doesn't know about the game. A hundred thousand people are
not going to download and install a separate interpreter; nor are they
interested in learning that the Help command at the top of the window
(or, if you're a Mac person, at the top of the screen) applies to
something other than the game itself.
Here's a more radical idea: It's impossible to get stuck. The player who
is a complete dunce and does nothing but hit the Enter/Return key over
and over should be able to read and enjoy the entire story. And yet the
story should reward interaction. Those may seem to be disparate criteria
... but I don't think they're mutually exclusive. I think it's a design
problem, but not impossible to achieve.
--JA
ChicagoDave
> Goals for current designers to take into account are of course
> interesting too. But what I'm envisioning is a fantasy in which a
> hundred thousand people who have never tried IF before are downloading
> this game. What would need to be in that download for them to get
> excited and recommend it to their friends in breathless terms?
Based on my own assumptions, 100,000+ downloads with a high viral
aspect would probably mean the story itself was based on an existing
and extremely popular story or set of characters. I doubt anything
else, including; parser, the level of writing, the user experience,
even the publisher would have that sort of impact. Well, spending a
hundred thousand dollars or more for public relations and marketing
would go a long way too.
David C.
Jim Aikin
>
> Based on my own assumptions, 100,000+ downloads with a high viral
> aspect would probably mean the story itself was based on an existing
> and extremely popular story or set of characters. I doubt anything
> else, including; parser, the level of writing, the user experience,
> even the publisher would have that sort of impact. Well, spending a
> hundred thousand dollars or more for public relations and marketing
> would go a long way too.
As I indicated, this is a fantasy -- but it's a fantasy with an
underlying agenda, which is to attempt to identify the features that IF
software would need to have in order to appeal to a large market of
people who are entirely unfamiliar with the conventions of the medium.
You can make whatever assumptions you like, and please feel free to pick
a number out of a hat. I could have said a million downloads. The point,
I hope, is not to shoot down any discussion of possible features by
saying, "That's unrealistic," but rather to envision some of the things
that would be needed in order for it to _become_ realistic.
You'll note that I didn't rule out any of the other criteria you
suggested. I didn't say the game wouldn't be a spinoff from a movie or
TV series. I didn't say it wouldn't be published by a multinational
conglomerate that was prepared to invest major bucks in a promotional
blitz. Those things are unlikely, but not impossible.
What I _am_ suggesting is that in order to interest either a major
publisher or the copyright holder of an existing franchise, the software
itself would need to be something that could achieve mass market
momentum. A limited sampling of anecdotal evidence suggests that quite a
lot of people who try IF are baffled by or simply impatient with the
existing paradigm, and quickly lose interest in trying to play the game.
Redesigning the UI in ways that would change that reaction is arguably
achievable. At least, I think it's worth discussing.
--JA
Poster
Jim Aikin <midig...@sbcglobal.net> wrote:
> Here's a blue-sky question that might be worth kicking around.
>
> Someone has recently released a Totally Unbelievable Game (text-based,
> of course). Word on the street has reached fever pitch, and the servers
> are chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
>
> Of course it would have a great story, that's essential. But what other
> features would it have? I can think of five or six things, but here's
> one to start the discussion:
>
> The parser would have to be much more flexible and "intelligent" than
> current parsers. Current IF parsers embody some fairly stringent (and
> hidden) assumptions about the input. People who know IF understand these
> assumptions; others don't.
Future Boy! and 1893 didn't have fantastically complex parsers, and they
are two successful recent IF ventures. Based on that evidence, I'm not
at all sure that parser gymnastics are required for success.
-- Poster
www.intaligo.com Building, INFORM, Seasons (10,000 lines and counting)
Ron Newcomb
Juicy question, Jim!
It would have to be something that -only- I-F could do, something that
couldn't be done with books or video games. Maybe... the player/
reader learns something about his/her own personal self, like a new
perspective on their own life or attitudes toward the real world, as a
result of the actions and situations they've put themselves through;
situations with the psychological depth of the novel but also the
personal accountability of the game/sim.
The work "learns" the player, and then uses that knowledge to teach
somehow.
-R
Emily Short
> Here's a blue-sky question that might be worth kicking around.
>
> Someone has recently released a Totally Unbelievable Game (text-based,
> of course). Word on the street has reached fever pitch, and the servers
> are chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
>
> Of course it would have a great story, that's essential. But what other
> features would it have? I can think of five or six things, but here's
> one to start the discussion:
>
> The parser would have to be much more flexible and "intelligent" than
> current parsers. Current IF parsers embody some fairly stringent (and
> hidden) assumptions about the input. People who know IF understand these
> assumptions; others don't.
>
> We could make a long, long list of grammatical constructions novice
> players might try: "is david smiling?" "[tell david] grandma fell down
> the stairs and broke her ankle." "I want the ice cream cone." "I want
> david to give sheila the ice cream cone." "ask sheila if david is
> kidding about grandma falling down the stairs." In a game that achieves
> runaway success, I believe the parser would need to be able to handle
> all of these structures (and hundreds of others) gracefully.
I agree that the parser would ideally be more polished; I disagree
that it would need to be better *in this way*. (In fact, I think if it
*were* better in this way, it would no longer be IF as we know it; it
would be something else, probably something with diffuse and unfocused
interaction, more like a chat with a mediocre AI than like a game.)
Yes, it's worth doing some work to catch novice input that is patently
not of the verb-noun form, and it's also worth offering a bit of help.
But in my experience people seem to catch on pretty quickly about what
sort of grammar the game is going to be able to understand, and don't
keep typing random questions, etc., unless they are bored and
deliberately fooling around.
What I do think is that players get frustrated when they type a lot of
things that don't produce any interesting output, so that the effort-
to-fun ratio drops.
There are several ways to address this.
(a) Use typo detection to weed out failure through user typing error.
(b) Implement rigorously to avoid "you don't see that here" sorts of
responses.
(c) Make sure that it's always, always clear in the game what the
player's goals are and what kinds of interaction will produce
interesting outcomes.
(d) See that all command words are hinted, and failed uses of valid
commands always make it clear why the command failed.
(e) Permit multi-step actions to be expressed through macro commands
(like GO TO KITCHEN instead of N. N. W., but perhaps extending to
other aspects of the game as well).
(f) Eliminate default messages as much as humanly possible, replacing
them with extensive *game-appropriate* feedback. This one is
unfortunately very laborious and impossible to pass from game to game
by library, but my observation is that it makes a huge difference to
the way relatively-novice players experience a game. If they're stuck
and trying to think of something to do, they may well run around
trying to eat everything, taste everything, smell everything, etc.;
and having the game *continue to entertain them* while they do this is
what keeps them around long enough to think of a more productive
action.
This is of course a tricky line to tread because you don't want to
reward them *so* much for unproductive actions that they spend the
whole time licking scenery instead of having a gripping time with your
plot and puzzles.
Jim Aikin
>
> Future Boy! and 1893 didn't have fantastically complex parsers, and they
> are two successful recent IF ventures. Based on that evidence, I'm not
> at all sure that parser gymnastics are required for success.
Fair enough. But let me respond with two questions:
First, might they have been _more_ successful if they had had more
flexible parsers? I don't think we can rule it out.
Second, what features _would_ you see as essential for a fantastically
successful text game? I'm talking about hundreds of thousands of
downloads. And I'm not really thinking about external factors, such as
huge promotional budgets. I'm talking about features that are intrinsic
to the game experience itself.
One possible answer is, "Nothing needs to change. The current game
engines are fine just the way they are." But that answer may be too
complacent. At the very least, I think contemplating some refinements
might have real value.
Just to clarify, I'm not out to criticize the talented folks who have
developed the current implementations! If you look at either Inform or
TADS, you'll find an implementation that is miles beyond Zork. There are
lots of slick features in today's IF that work extremely well. I'm just
wondering if a redesigned UI might have the potential to win over a
bunch of converts.
Part of the reason I'm asking is because a few days ago someone on this
group mentioned Roguelike games. I was all but oblivious to the genre,
so I looked it up. My reaction to reading about how NetHack is played
was in two parts: (1) Wow, this is amazing! I'm sure it would be fun.
(2) It's much, much too complicated for me to bother trying to learn it.
And that's a game where all of the commands are explicitly listed in the
rules! In text-based IF, it's quite normal for some commands to be
hidden from the player. That may not be a good thing. Maybe one of the
features of mass-market IF in the 21st Century ought to be, "All of the
possible commands are listed." Either in a help file, or perhaps in a
separate window that changes dynamically depending on what objects are
in scope. I'm not saying this is a good idea, necessarily. It's just an
example of the kind of thing I'm mulling over.
--JA
Nikos Chantziaras
> Poster wrote:
>>
>> Future Boy! and 1893 didn't have fantastically complex parsers, and
>> they are two successful recent IF ventures. Based on that evidence,
>> I'm not at all sure that parser gymnastics are required for success.
>
> Fair enough. But let me respond with two questions:
>
> First, might they have been _more_ successful if they had had more
> flexible parsers? I don't think we can rule it out.
Well, considering that "more flexible" probably means more "guess the
verb," I'd say no; they wouldn't have been more successful. Unless the
parser is really that good to the extend that it appears to be human.
> Second, what features _would_ you see as essential for a fantastically
> successful text game? I'm talking about hundreds of thousands of
> downloads. And I'm not really thinking about external factors, such as
> huge promotional budgets. I'm talking about features that are intrinsic
> to the game experience itself.
One thing I really, really, really liked back in the days was that the
games were full screen. I can't play Quake 4 in a window really. Same
goes for text adventures. They also came with cute graphics and music
(see "Eric The Unready"), and some with character portraits during
conversations and even with 3D videos (see "Frederik Pohl's Gateway" and
"Gateway 2"). The parser was an evolved version of Infocom's one;
nothing too fancy. Note that these were *not* "point&click" games.
Playing IF while the OS or interpreter interface is visible makes me
feel like I'm reading a PDF rather than playing a *game*.
Jim Aikin
>
> I agree that the parser would ideally be more polished; I disagree
> that it would need to be better *in this way*. (In fact, I think if it
> *were* better in this way, it would no longer be IF as we know it; it
> would be something else, probably something with diffuse and unfocused
> interaction, more like a chat with a mediocre AI than like a game.)
Fair enough. Perhaps if the newcomer starts by typing questions that the
parser isn't equipped to handle, it could at least diagnose that that's
the nature of the problem and respond, "It looks as if you're trying to
move through the story by giving complex instructions. Most of the time,
a simple command of the form 'verb the noun' (that is, 'action -> thing
acted on') will do the job nicely." That's ever so much friendlier than,
"The word 'intimidate' is not needed in this story."
> (a) Use typo detection to weed out failure through user typing error.
And the first time this happens, the software can say, "It looks as if
you typed 'tkae' when you meant 'take'. Would you like the story to
automatically correct typing mistakes for you? (You can answer yes or no.)"
> (b) Implement rigorously to avoid "you don't see that here" sorts of
> responses.
> (c) Make sure that it's always, always clear in the game what the
> player's goals are and what kinds of interaction will produce
> interesting outcomes.
> (d) See that all command words are hinted, and failed uses of valid
> commands always make it clear why the command failed.
> (e) Permit multi-step actions to be expressed through macro commands
> (like GO TO KITCHEN instead of N. N. W., but perhaps extending to
> other aspects of the game as well).
> (f) Eliminate default messages as much as humanly possible, replacing
> them with extensive *game-appropriate* feedback.
Yes on all points. That's a very good list! And these are things that
can easily be done with existing development systems. Are there things
you'd like to see, or think novices could benefit from, that are not as
easily accomplished with I7 or T3?
For instance, auto-completion of the command line? For instance, a
built-in auto-mapper that would show the player exactly where she has
been and what she has discovered there?
A random idea from last week: a 'solve' command. Whenever you're faced
with a puzzle and are getting irritated by it, just type 'solve' and the
puzzle will go away. Like hints, this is something the experienced gamer
should be able to turn off....
--JA
Jim Aikin
>>
>> First, might they have been _more_ successful if they had had more
>> flexible parsers? I don't think we can rule it out.
>
> Well, considering that "more flexible" probably means more "guess the
> verb," I'd say no; they wouldn't have been more successful. Unless the
> parser is really that good to the extend that it appears to be human.
I'm not sure why it would mean "guess the verb". To me, "guess the verb"
is what happens when the parser is poorly implemented and lacks basic
synonyms. If the parser understands both more synonyms and more variety
of syntax, how is that a bad thing?
For instance, how about allowing 'I want the sword' as a synonym for
'take the sword'? Easily implemented, and it will help someone somewhere
along the line.
> One thing I really, really, really liked back in the days was that the
> games were full screen. I can't play Quake 4 in a window really. Same
> goes for text adventures. They also came with cute graphics and music
> (see "Eric The Unready"), and some with character portraits during
> conversations and even with 3D videos (see "Frederik Pohl's Gateway" and
> "Gateway 2"). The parser was an evolved version of Infocom's one;
> nothing too fancy. Note that these were *not* "point&click" games.
>
> Playing IF while the OS or interpreter interface is visible makes me
> feel like I'm reading a PDF rather than playing a *game*.
This is an interesting point. I had never considered it. I think you're
probably right that it would contribute to a sense of immersion.
Of course, these days many of us have large monitors. Normal-sized text
on a large monitor would be difficult to read, because the lines would
be too long. (Some research has been done on this, I believe. The point
is, if your eye has to jump back too far at the end of each line, it's
hard to find the correct spot to jump to. There's a maximum tolerable
character-height-to-line-length ratio, though I don't know what it is.)
And large text might look dumb. But if the game takes over the whole
monitor and arranges the text in a nice-sized central pane that's
surrounded by other useful panes (an inventory list, a clickable compass
rose, illustrations of NPCs' faces or whatever), I think it could work
very well.
And speaking of music, I would love to be able to do music within a
game. My problem with audio in T3 is that it can't be faded out smoothly
when the player does something that would render the music superfluous
or weird, such as leaving the beautiful garden and entering the grisly
slaughterhouse. Unless HTML TADS has been updated and I missed it, the
audio can only be chopped off abruptly in that type of situation. For
me, that's a deal-breaker.
So yeah, I think the ability to display graphics (moderately well
supported at present ... sort of) and play music (not well supported)
would be very strong additions to a consumer-friendly IF delivery system.
--JA
S. John Ross
> One thing I really, really, really liked back in the days was that the
> games were full screen.
While the choices certainly vary by operating system and game
language, z-code games running in Frotz for Windows go full-screen
with a simple ALT-E ... the OS vanishes, the interpreter vanishes.
Just you and the game, like in days of old (but with vector fonts and
color choices) :)
S. John Ross
(ALT-ENTER), I meant. D'oh!
Nikos Chantziaras
> Nikos Chantziaras wrote:
>> Well, considering that "more flexible" probably means more "guess the
>> verb," I'd say no; they wouldn't have been more successful. Unless
>> the parser is really that good to the extend that it appears to be human.
>
> I'm not sure why it would mean "guess the verb". To me, "guess the verb"
> is what happens when the parser is poorly implemented and lacks basic
> synonyms. If the parser understands both more synonyms and more variety
> of syntax, how is that a bad thing?
Allowing more variety is different then implementing more variety :P
Unless you mean just synonyms.
>> One thing I really, really, really liked back in the days was that the
>> games were full screen. I can't play Quake 4 in a window really.
>> Same goes for text adventures. They also came with cute graphics and
>> music (see "Eric The Unready"), and some with character portraits
>> during conversations and even with 3D videos (see "Frederik Pohl's
>> Gateway" and "Gateway 2"). The parser was an evolved version of
>> Infocom's one; nothing too fancy. Note that these were *not*
>> "point&click" games.
>>
>> Playing IF while the OS or interpreter interface is visible makes me
>> feel like I'm reading a PDF rather than playing a *game*.
>
> This is an interesting point. I had never considered it. I think you're
> probably right that it would contribute to a sense of immersion.
>
> Of course, these days many of us have large monitors. Normal-sized text
> on a large monitor would be difficult to read, because the lines would
> be too long. (Some research has been done on this, I believe. The point
> is, if your eye has to jump back too far at the end of each line, it's
> hard to find the correct spot to jump to. There's a maximum tolerable
> character-height-to-line-length ratio, though I don't know what it is.)
> And large text might look dumb. But if the game takes over the whole
> monitor and arranges the text in a nice-sized central pane that's
> surrounded by other useful panes (an inventory list, a clickable compass
> rose, illustrations of NPCs' faces or whatever), I think it could work
> very well.
I should have linked to this page in my previous post, but what you just
described is indeed exactly how it was done:
http://www.mobygames.com/game/dos/gateway-2-homeworld/screenshots
These are screenshots from Gateway 2. The interface was configurable
(verb list and inventory were optional). Also note that it implemented
both "ask/tell about" conversation as well as menu-based conversation
(along with portraits).
I think it's obvious by now that this is my favorite text adventure ;)
Aaron A. Reed
An axiom from traditional writing which I don't think is carried over
to IF as often as it should be is the old notion that you have to hook
the reader in the first paragraph to get their attention.
Unfortunately, the first paragraph of IF is clicking to a directory
listing of obscure filenames that, even if you can figure out you need
the one called "Flaxo" with "precompiled Windows binary" in the
description, downloads at 1K per second on a good day.
However, the fact that there are now several Flash-based IF players on
the web indicates this is probably not the real problem.
Issues of writing quality aside, at the end of the first paragraph you
are given a blank prompt with no instructions as to what to do with
it.
If you and the author are both astute, you may have noticed a message
for new players to type "about." That may give you some ideas of what
to type next.
Past evidence suggests, however, that even in this case, within the
next two or three moves the player will type something that the game
does not understand. With the Internet's attention span, this is
roughly equivalent to getting a message about a missing video codec or
a 404 error, and it means "No soup for you. Move along." And the vast
majority will.
So perhaps authors should spend their time desperately attempting to
get their games to recognize as many possible inputs in the first few
moves. Implement the hell out of every scenery object; add every typo-
correction and input-comprehension extension you can find; go through
the thesaurus and pile on the verb synonyms; add a tutorial mode. If a
player can get past the first few moves, long enough to start seeing
how their actions are having an impact on or are somehow important to
the story, then they get interested.
I think more ways to recognize broader ranges of input is a key place
where IF is lacking here. I remember one comment I read from a college
student briefly exposed to IF that went along the lines of: "I saw the
first prompt and typed 'I don't know what to do.' The game said back
'I only understood you as far as wanting to inventory.' So I gave up."
And I don't really blame him.
The collegiate hackers of the '70s and aging geeks who like to fiddle
with things will put up with a response as clueless as that; the
millions necessary to make something the massive hit that Jim wishes
for will not.
--Aaron
Nikos Chantziaras
I usually run them in a console with a large font and the monitor a bit
further away than usual.
Though for me, the "perfect game" *has* to offer graphics and sound.
Don't take me wrong; I enjoy pure text-based games. But we're talking
about "perfect" here ;)
S. John Ross
> Though for me, the "perfect game" *has* to offer graphics and sound.
> Don't take me wrong; I enjoy pure text-based games. But we're talking
> about "perfect" here ;)
I heard that. One of the challenges, I guess, is how to daydream
sexifying a text adventure while keeping it (I presume) a text
adventure. Tricky to suss out the boundaries (reminds me of when I was
working for Steve Jackson, and he'd give a company pep-talk that
amounted to "let's think outside the box ... well, not actually
OUTSIDE the BOX ... but ... inside the box ... at the edges ... but
RIGHT up to the edge, without actually leaving the box ... and if you
come up with something outside the box, we can all brainstorm and
maybe think of a way to cram it back inside the box ...) :)
I mean, if James Earl Jones or Ian McKellan is narrating the adventure
with audio and I'm responding to a voice-recognition "parser" without
typing anything [so I can play while I wash the dishes or fold clothes
or something], would that qualify as a dream feature appropriate to
the thread? 'Cause those guys have groovy voices, is all I'm saying :)
Heck, I'd download a hundred thousand copies all by myself if I could
simulate having a conversation/storytelling jam while folding
clothes ...
This is freeware, right? As long as I'm wishing, I wish it is :)
Anyway, fun thread, even if I don't really understand the question.
Damien Neil
> Someone has recently released a Totally Unbelievable Game (text-based,
> of course). Word on the street has reached fever pitch, and the servers
> are chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
It has one hell of a hook. The first sentence of the game demands that
you read the first paragraph. The first paragraph makes it nearly
impossible to not type in your first command. The result of your first
command leaves you burning to find out what happens if you try THIS
next...
It's easy to get into--easy enough that people who haven't played a text
adventure in decades can start playing with little pain.
It's deep--deep enough that people who write articles on IF theory for
fun find it interesting.
It has network effects. There are many forms that this could take:
Maybe everyone is talking about the plot, and their interpretation of
it. Maybe everyone is trying to figure out how to get the Babel fish,
or wondering what will happen in the next episode.
Everything else is negotiable.
For example, from the graphical side of things, Portal. The opening
scene hooks you right off the bat, the game is very deliberately
designed to train people to develop the skills needed to solve each
puzzle as they reach it, the puzzles are nonetheless complex enough to
be interesting, and everyone who finishes just HAS to find someone and
do GLaDOS impersonations at them.
Or Pokemon, which is childishly simple on the surface and filled with
unexpected depths for anyone who really does want to "catch 'em all".
Or outside of gaming, Harry Potter, where the increasingly complex story
fed fuel to an inferno of speculation about each next book, which then
pulled yet more people into the series.
- Damien
Victor Gijsbers
The player should never become stuck. No, I mean something even
stronger... Let me try to explain.
_The_ psychological barrier that interactive fiction has and which
almost no other type of game has, is that in all those other games,
there is always something you can do which you know will probably move
the game forwards to a good conclusion, and at least will move the game
forwards.
In Monopoly, all you have to do is pick up the dice and roll.
In Quake, all you need to do is walk to that part of the level where you
haven't and other first-person shooters yet been and shoot everything.
(There used to be some FPS-games with "puzzles"; where you had to press
the right combination of buttons, or find the hidden keys, in order to
proceed. I don't think that's popular anymore.)
In Baldur's Gate and other CRPGs, you can either go to some location you
haven't visited yet, slay some monsters in a location you have visited,
or read your Quest Log to see where you should be going.
In Sudoku, you can always just apply the standard methods of solving and
(unless the Sudoku is fiendishly difficult) this will allow you to make
steady progress.
In a Jigsaw puzzle, you can always start picking up the pieces one by
one a _trying_ if they fit. Progress will be slow, but it will be
assured. The same with the moves in a chess puzzle. Progress might be
incredibly slow, but it is assured.
... and so on.
In all these games, there is always this utterly clear way of
progressing--which doesn't mean they are linear, or stupid, or anything
like that. It just means that there always are clear paths of progress.
(In a sense, the same holds for reading a book--all you have to do is
read on.)
This is not the case in IF. In IF, there are many, many times when you
just sit staring at your screen and thinking: "I don't know what to do
right now. I'm out of ideas."
People aren't used to that, and they don't like it. That's not because
they are lazy--these might be the same people who solve hard chess
puzzles and 2000-piece Jigsaws, beat their friends in 7-hour games of
Axis & Allies and complete Baldur's Gate 2 with a single character
instead of the usual party of six. It's not that they want to be
spoon-fed all the answers. (In fact, they may even be too stubborn to
check the hints or the walkthrough!) It's just that they don't like the
feeling that there is currently no way for them to proceed, nothing to
do, nothing to try. That feeling of "stuckness" (which is more
ubiquitous than our usual "I'm totally stuck") is something people don't
like and aren't willing to put up with. And thus they quit the game.
I think this thing is the biggest barrier for interactive fiction today.
People don't want the feeling for stuckness, and yet, almost all pieces
of IF will at one point or another make people feel stuck.
This is the problem that the Totally Unbelievable Game would have solved.
Regards,
Victor
Jason Stokes
"Nikos Chantziaras" <rea...@arcor.de> wrote in message news:ftpmi6
> I should have linked to this page in my previous post, but what you just
> described is indeed exactly how it was done:
> http://www.mobygames.com/game/dos/gateway-2-homeworld/screenshots
>
> These are screenshots from Gateway 2. The interface was configurable
> (verb list and inventory were optional). Also note that it implemented
> both "ask/tell about" conversation as well as menu-based conversation
> (along with portraits).
>
> I think it's obvious by now that this is my favorite text adventure ;)
Yeah, all those legend entertainment games were fun. I really liked
"Timequest", and "Eric the Unready" was pretty funny (although hopeless
unstructured and disorganised as a game, and not all that challenging.) I'm
working through "Gateway" at the moment.
Something like the legend interface would have to be a given in this day and
age, for a wider audience, who demand their pretty pictures... and even then
I doubt it would be enough. I feel that a new media or platform might
rekindle some interest, like IF on cellphones or ebook readers.
Doug Egan
My first thought, after reflecting on your description of the "ideal
parser", is the story of Automon Chess Player (also called the
"Mechanical Turk"), a chess-playing machine exhibited from 1770-1854.
Although the inside of the machine was decorated with elaborate
clockworks to fool observers, the machine was actually operated by a
human chess master, hidden inside.
http://en.wikipedia.org/wiki/The_Turk
Interactive Fiction, played on-line and facilitated by a human story
teller (or multiple story tellers) is a goal easily achieved within
the limitations of existing software. Examples and variations of this
form of story telling are undoubtably already being used, though the
fact that I can't cite any of them without additional research tells
me something about their popularity.
How far from the current model of IF are you willing to stray to make
it more popular? The following are examples of interactive fiction
(defined more broadly) which have reached viral popularity:
Gigapets, and Tomagatchi
Sim City, Baldur's Gate, Grand Theft Auto.
What do these games have, that conventional IF games do not? They
don't have any more elaborate parser. I think the common feature of
these games is graphics and sound.
Dave
Jimmy Maher
> The parser would have to be much more flexible and "intelligent" than
> current parsers. Current IF parsers embody some fairly stringent (and
> hidden) assumptions about the input. People who know IF understand these
> assumptions; others don't.
>
> We could make a long, long list of grammatical constructions novice
> players might try: “is david smiling?” “[tell david] grandma fell down
> the stairs and broke her ankle.” “I want the ice cream cone.” “I want
> david to give sheila the ice cream cone.” “ask sheila if david is
> kidding about grandma falling down the stairs.” In a game that achieves
> runaway success, I believe the parser would need to be able to handle
> all of these structures (and hundreds of others) gracefully.
I don't really accept this at all. Some anecdotal evidence: a few years
ago I had to sit down with my girlfriend (now fiancee) and have the
dreaded "IF conversation," where I explain just what this crazy, geeky
hobby I devote so much time to is really all about. We fired up a game
-- Paul O'Brian's first Earth and Sky, I believe -- and I showed her the
ropes -- navigation, the handful of meta-verbs, other common verbs like
examine and inventory, etc. -- and within an hour or so she could play
quite comfortably on her own.
Now, my fiancee is of a course a very smart person :), but she's not
remotely geeky, at least computer geeky. She's a medical student, has
no particular interest in computers as anything more than tools for
email and so on, has never used a command line in her life, etc. Yet
she picked up on IF quite quickly and easily.
I think most reasonably intelligent people can be perfectly competent
IFers with just an hour or so of "training." Compare this to other
sorts of computer entertainment. How long does it take to really
understand the Warcraft interface, or to get comfortable with all the
shortcuts and options in the average FPS? (Let's not even go near
NetHack.) I just don't accept that the IF learning curve is all that
steep. Instead of twisting ourselves into knots trying to accept
constructions like "I want david to give sheila the ice cream cone," I
think we would be better off producing materials -- online videos,
interactive Flash demonstrations, perhaps even online person-to-person
training -- to teach the basics of interaction with IF as it currently
works. This would allow our hypothetical newbie to not only play our
new super-awesome game, but everything that has come before as well.
I'm not saying for a moment that IF doesn't still bring occasional
guess-the-verb problems and other parser frustrations, just that those
are really authorial oversights rather than indicative of a more
systemic problem with the form itself.
Even the oft-cited "problem" of having to download a separate
interpreter is hardly mind-boggling to most people in the modern world.
Everyone understands that they need a separate player to, say, play an
MP3 file, and judging by the amount of legal and illegal music
downloading going on I think we can assume this hasn't been much of a
barrier to anyone. People also understand that the same player cannot
always play MP3 and Ogg Vorbis files. We simply need to explain how IF
works in these terms. Going with a stand-alone executable might be a
good option for our hypothetical unbelievable game since it would give
us more interface options, but it seems to me hardly a requirement for
IF as a whole.
If you target the game market, I think a breakout IF game would look a
bit like the Legend games, with music, graphics, menus for those who
don't feel like typing, auto-mapping, and even possibly occasional
cut-scenes. I think it most of all, though, would need to be something
very epic and high-concept, with many hours of play-time and a sweeping
story to offer. Even with all this I'm not sure you could create more
than a modest success fueled at least partially by nostalgia ("Remember
text adventures?? Aikin Productions updates them for the new millenium!").
A better choice would be to go after a different market entirely:
readers who are left flat by conventional computer games. I haven't
figured out how to reach this market, though. Even with all the great
outreach that has been done by Emily and others, their efforts have
largely been through gaming websites -- Jay is Games, Play This Thing,
etc. I think only a small minority of these folks are ever going to
appreciate IF, not because they're stupid or somehow less than us but
just because they aren't looking for the same thing we are. The trick
is to reach readers rather than videogame players.
In short, I don't think the problem is the learning curve. The problem
is making people aware of IF and willing to give it a shot, willing to
invest that hour or so to learn how it works. That's where we fail to
some extent. Paradoxically, I think the fact that virtually all modern
IF is free actually hurts us in a way, because it causes the mainstream
to dismiss it and not value it. 1893 and Future Boy! both got reviews
on adventure gaming sites because they were commercial. It's one reason
I would like to see more commercial IF, which I would more than happily
pay for.
--
Jimmy Maher
Editor, SPAG Magazine -- http://www.sparkynet.com/spag
Thank you for helping to keep text adventures alive!
Conrad
> this Totally Unbelievable Game have that currently available games don't?
Well... since you're asking for _totally unbelievable_...
How about a game which has a sufficiently advance simulator and
narrative tracker built-in that there is no set scenario, no set
characters, no set plot line; you tell it what kind of genre you want
to play in (sci-fi, wild west, contemporary) and what kind of
character you'd like to play, and it generates a game on the fly.
IOW, how about something that passes the Turing Test for being a
passably good DM?
Conrad.
Andrew Plotkin
> If they're stuck
> and trying to think of something to do, they may well run around
> trying to eat everything, taste everything, smell everything, etc.;
> and having the game *continue to entertain them* while they do this is
> what keeps them around long enough to think of a more productive
> action.
>
> This is of course a tricky line to tread because you don't want to
> reward them *so* much for unproductive actions that they spend the
> whole time licking scenery instead of having a gripping time with your
> plot and puzzles.
Or maybe that's not a big concern.
I have long thought that _Myst_ became a multiple-zillion-seller
mostly because it was great at entertaining people who were failing to
get anywhere in it. If you had no grasp of the interaction model, no
notion of how the initial island was laid out, no spatial sense, no
idea of how to solve the first puzzles... you could still click
randomly on things and feel sucked into a gorgeous and varied world.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
If the Bush administration hasn't shipped you to Syria for interrogation, it's
for one reason: they don't feel like it. Not because you're innocent.
Jimmy Maher
> Past evidence suggests, however, that even in this case, within the
> next two or three moves the player will type something that the game
> does not understand. With the Internet's attention span, this is
> roughly equivalent to getting a message about a missing video codec or
> a 404 error, and it means "No soup for you. Move along." And the vast
> majority will.
I think the thing to realize here, though, is that probably somebody who
quits at that point was not all that interested anyway. If we manage to
find a way to keep them from quitting in the first minute, they will
just quit in the tenth, when they run into the first puzzle. At some
point we have to be willing to respect our audience's intelligence a
little more. Nothing rewarding comes completely without a learning curve.
Maybe we should make more of an effort to tell people WHY IF is worth
taking an hour to figure out.
> So perhaps authors should spend their time desperately attempting to
> get their games to recognize as many possible inputs in the first few
> moves. Implement the hell out of every scenery object; add every typo-
> correction and input-comprehension extension you can find; go through
> the thesaurus and pile on the verb synonyms; add a tutorial mode. If a
> player can get past the first few moves, long enough to start seeing
> how their actions are having an impact on or are somehow important to
> the story, then they get interested.
>
> I think more ways to recognize broader ranges of input is a key place
> where IF is lacking here. I remember one comment I read from a college
> student briefly exposed to IF that went along the lines of: "I saw the
> first prompt and typed 'I don't know what to do.' The game said back
> 'I only understood you as far as wanting to inventory.' So I gave up."
> And I don't really blame him.
Sorry, but I do. :) Honestly, this just sounds to me like someone who
wasn't really all that interested to begin with. What do you do with
any other computer program when you first encounter it? You fiddle
around until you figure out how it works, or you don't. If IF was
really that difficult, I would be more sympathetic, but it's just not.
There are steps we could take to help people out, of course. Maybe if
the player types something nonsensical a couple of times in a row, we
say, "If you haven't played IF before, type HELP for some instructions
on how to interact." But if we start trying to understand complicated
sentence like those Jim suggested, I can tell what will happen already:
we will just start to MISunderstand said sentences, and that will lead
to even more player confusion.
Aaron A. Reed
> Aaron A. Reed wrote:
> > Past evidence suggests, however, that even in this case, within the
> > next two or three moves the player will type something that the game
> > does not understand. With the Internet's attention span, this is
> > roughly equivalent to getting a message about a missing video codec or
> > a 404 error, and it means "No soup for you. Move along." And the vast
> > majority will.
>
> I think the thing to realize here, though, is that probably somebody who
> quits at that point was not all that interested anyway. If we manage to
> find a way to keep them from quitting in the first minute, they will
> just quit in the tenth, when they run into the first puzzle.
Right, but I expect that more people, possibly an order of magnitude
more, are willing to try something for 1 minute than for 10 minutes.
Once they start to get interested and like it, they'll play for
longer.
Look at everything in the last few years that attains success on the
level that Jim wants. Youtube videos, for example. It takes 0 effort
or intelligence and about 2 seconds of time to start playing a Youtube
video. But if the first 30 seconds interest you, you might watch the
whole thing, even if it's half an hour long. I think IF needs to be
more front-loaded towards making those first few moments more
enjoyable.
--Aaron
David Fisher
news:ftp2hg$c6t$1...@aioe.org...
>
> Someone has recently released a Totally Unbelievable Game (text-based, of
> course). Word on the street has reached fever pitch, and the servers are
> chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
As well as a parser that responds to a large range
of inputs (even if the response is to guide the player
towards a more limited range of inputs) -- an underlying
world model that lets the player do everything that
"makes sense" to them (within reason). Combined with
a wonderful plot which is reasonably resistant to
corruption by the player's actions, at the same
time allowing the player to influence the plot in a
logical way.
(Phew! Pauses to take a breath ...)
Well, I'd play it. :-)
I think "simulation" has a bit of a bad name, maybe
because it is seen as an alternative to story based IF
rather than supporting / enhancing it. But a big problem
new players have with IF is getting used to the idea that
there is a very limited range of solutions to most
problems; if you need a small object to throw at
something, no way are you going to be able to pick up
a pebble from the road or take off your shoe and use
that -- you need to use whatever object(s) the author
wants you to.
If the world model includes a detailed simulation of
the world, then a lot more things that the player tries
will work; the "rules" of the world could even allow
the player to come up with solutions the author hadn't
originally intended (see the last Magnetic Scrolls thread).
The main problems with this level of simulation seem
to be:
* Text generation. Not only are some descriptions difficult
to generate in themselves (eg. overlapping tracks through
the snow), but default messages are boring. The ideal would
be for every message to be context sensitive and to include
things like characterisation of the PC, reactions from NPCs
who are observing, subtle guidance about whether the player
is on the right track or not, etc. "Authorial burden" indeed ...
- definitely a problem.
* A detailed simulation of some aspect of the game world
can act as a red herring. If there are no puzzles
involving mixing liquids, but the player discovers that
the game responds impressively when they try to do this,
it gives an unintended message about its importance in
the rest of the game.
- players would need to understand the type of game they
are playing, and not rely some much on implementation as
a clue to something being the "correct" thing to do.
More helpfully, the game might need to give more guidance
than usual about where the plot is going and what kinds
of things would be a useful for the player to focus on
at the moment.
There are some other objections to deep simulation which
I do not think must always apply:
* "If you model reality too closely, it will be boring
(and you'll have to go to the toilet all the time)."
- but only if that's what you want to include in your
model. A detailed simulation does not necessarily have
to follow all of the normal rules of reality (I would
be happy with an unlimited inventory, for example).
Playability trumps realism!
* "Simulation is opposed to good story telling. If there
is lots of simulation in a game, then the game will
inevitably be all about fiddling with objects and the
mechanics of the simulation."
- the aim of all this would be to allow many more of
the things players try to work properly. This the usual
aim of Beta testing -- to catch things players might try,
and add them to the game. A deep simulation has the same
aim. The game could still be just as story oriented as
any other game that contains multiple puzzle solutions.
* "If the player has freedom to do lots of simulationy things,
especially things the author hasn't explicitly thought of,
then it would be too easy for them to wreck the plot
(bypassing puzzles, etc.)"
- I think the best solution would be to have a tool which helps
authors track down alternate solutions. They can then either
forbid those solutions with reasonable excuses to the player,
or modify the plot as necessary to take them into account
(but actions that modify the plot dramatically would still
make a lot more work for the author).
Disclaimer: not saying that all IF should head in this kind
of direction, or that this is actually feasible (though I
have some ideas on an approach), just that I think it would
be a lot of fun to play something like this.
People love it when stuff they try in a game actually works,
and their lateral (but logical) thinking pays off; I think
this kind of approach would make that happen a lot more.
David Fisher
Aaron A. Reed
To tie in Victor's post with what I was saying, I think the command
prompt can be terribly intimidating for people who have never seen one
in DOS or Unix before. Especially with people past 40 or so who didn't
grow up with computers, I've frequently witnessed almost a panic when
that command prompt appears. I think this is mostly psychological:
residual fear of "breaking" the computer, or of typing something
incorrect and looking foolish, maybe.
Which gives me an idea: Maybe the "prologue" part of the game (before
the first few command prompts) should contain a few prompts and
responses, to get new players used to what it is and what it's for
before they're suddenly expected to use one. (Maybe this isn't an idea
so much as something I saw somewhere before; has some game done this
already?)
--Aaron
Aaron A. Reed
> Even the oft-cited "problem" of having to download a separate
> interpreter is hardly mind-boggling to most people in the modern world.
> Everyone understands that they need a separate player to, say, play an
> always play MP3 and Ogg Vorbis files.
Yikes, Jimmy. Based on some of the people I've done tech support for
in the past year, I think these are extraordinarily generous
assumptions. :)
They hold if we're limiting our audience to "computer people who know
the difference between MP3 and OGG." And the popularity of things like
HomeStarRunner suggests that's an audience large enough to garner
popularity on the scale of millions. But in my experience that would
be less than half, possibly far less, of adults with computers.
--Aaron
Jim Aikin
> Aaron A. Reed wrote:
>> So perhaps authors should spend their time desperately attempting to
>> get their games to recognize as many possible inputs in the first few
>> moves. Implement the hell out of every scenery object; add every typo-
>> correction and input-comprehension extension you can find; go through
>> the thesaurus and pile on the verb synonyms; add a tutorial mode. If a
>> player can get past the first few moves, long enough to start seeing
>> how their actions are having an impact on or are somehow important to
>> the story, then they get interested.
>>
>> I think more ways to recognize broader ranges of input is a key place
>> where IF is lacking here. I remember one comment I read from a college
>> student briefly exposed to IF that went along the lines of: "I saw the
>> first prompt and typed 'I don't know what to do.' The game said back
>> 'I only understood you as far as wanting to inventory.' So I gave up."
>> And I don't really blame him.
>
> Sorry, but I do. :) Honestly, this just sounds to me like someone who
> wasn't really all that interested to begin with. What do you do with
> any other computer program when you first encounter it? You fiddle
> around until you figure out how it works, or you don't. If IF was
> really that difficult, I would be more sympathetic, but it's just not.
Sorry, Jimmy, but I think I agree with Aaron here. It seems to me (and I
could be wrong -- feel free to correct me) that what you're saying boils
down to, "We don't need to change anything. If people don't like it,
it's their fault."
IF is not difficult for you or me, because we understand the conventions
of the interface -- but it's NOT a natural interface, and people are not
born with an instinct that tells them how to use it.
People won't read the help file, so you can't put instructions there.
What needs to happen is that during the first few turns the game will
steer the player gracefully in the direction of the type of input the
software can understand. Typing "I don't know what to do" is perfectly
rational, and the game response quoted above would be baffling to anyone
who didn't know what to do.
An easy fix (easy in TADS, at least -- I don't know about Inform) would
be to set up a widget that would look for not-understood inputs during
the first few turns and give one-line instructions in response rather
than spitting out the parser's usual not-understood message.
--JA
Jim Aikin
Built-in hints will do the job -- at least if people know to use the
hints and if the hints are well-written.
In one of the Spring Thing games, the hint system gives you a couple of
vague nudges and then punts without ever telling you what you need to
do. This is infuriating!
In-story clues are even better than hints, but not everyone will notice
them. Eric Eve very cleverly constructed a system (in "Mrs. Pepper's
Nasty Secret", source code available) that can look at the player's
situation, count a few turns, and then say, "You seem to be stuck here.
Would you like some help?"
--JA
Aaron A. Reed
> An easy fix (easy in TADS, at least -- I don't know about Inform) would
> be to set up a widget that would look for not-understood inputs during
> the first few turns and give one-line instructions in response rather
> than spitting out the parser's usual not-understood message.
A game with such functionality exists.
--Aaron
Jimmy Maher
Okay, fair enough. But I have to ask, even though it's going to make me
sound like even more of an elitist than this thread already has: can you
really picture those people playing and enjoying IF, even IF with the
various newbie-friendly features you and others have suggested?
IF is always going to be a niche, just because it requires reading and
thinking. It's immensely rewarding, but the rewards are cerebral rather
than immediate, and it's just not going to be a mass-market phenomenon.
For the most part people are looking for effortless, fluffy
entertainment, and IF doesn't do that. It's never going to be World of
Warcraft, nor YouTube. (And every time I read the comments after a
YouTube video, I thank God for that.) By the time we change it enough
to appeal to the average YouTube user, it's no longer going to be IF at
all.
I just think it might be possible to make it a much, much larger niche.
Perhaps, though, I am talking at cross-purposes with Jim's idea for
this thread. :) When I think of IF "success," I think of independent
films and music that appeal to niche audiences but nevertheless earn
enough to give their creators a modest living. I don't think of
Hollywood blockbusters, the Super Bowl, or World of Warcraft.
Jimmy Maher
> Look at everything in the last few years that attains success on the
> level that Jim wants. Youtube videos, for example. It takes 0 effort
> or intelligence and about 2 seconds of time to start playing a Youtube
> video. But if the first 30 seconds interest you, you might watch the
> whole thing, even if it's half an hour long. I think IF needs to be
> more front-loaded towards making those first few moments more
> enjoyable.
Not to beat this thing to death, but just a quick counter-example: World
of Warcraft is enormously successful, and yet getting online with it is
not effortless at all. They've been successful because they've done a
good job of promoting their brand, letting people know WHY it's worth
the effort to purchase and install a big piece of software, set up an
account, etc. And then of course you reach a certain critical mass
where "everyone is playing," and then the thing pretty much promotes
itself.
Maybe we could do a better job of marketing what we already have to offer...
Andrew Plotkin
> Aaron A. Reed wrote:
> > On Apr 12, 8:48 am, Jimmy Maher <mahe...@SPAMgrandecom.net> wrote:
> >
> > They hold if we're limiting our audience to "computer people who know
> > the difference between MP3 and OGG." And the popularity of things like
> > HomeStarRunner suggests that's an audience large enough to garner
> > popularity on the scale of millions. But in my experience that would
> > be less than half, possibly far less, of adults with computers.
>
> Okay, fair enough. But I have to ask, even though it's going to make me
> sound like even more of an elitist than this thread already has: can you
> really picture those people playing and enjoying IF, even IF with the
> various newbie-friendly features you and others have suggested?
Yes.
> IF is always going to be a niche, just because it requires reading and
> thinking. It's immensely rewarding, but the rewards are cerebral rather
> than immediate, and it's just not going to be a mass-market
> phenomenon.
*A* niche is not the same as *the* niche. There are plenty of brainy,
literate people out there whose computer competence (and comfort zone)
is no better than average. I can picture those people enjoying IF.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
If the Bush administration hasn't subjected you to searches without a warrant,
it's for one reason: they don't feel like it. Not because you're an American.
Jimmy Maher
> Jimmy Maher wrote:
>> Aaron A. Reed wrote:
>
>>> I think more ways to recognize broader ranges of input is a key place
>>> where IF is lacking here. I remember one comment I read from a college
>>> student briefly exposed to IF that went along the lines of: "I saw the
>>> first prompt and typed 'I don't know what to do.' The game said back
>>> 'I only understood you as far as wanting to inventory.' So I gave up."
>>> And I don't really blame him.
>>
>> Sorry, but I do. :) Honestly, this just sounds to me like someone who
>> wasn't really all that interested to begin with. What do you do with
>> any other computer program when you first encounter it? You fiddle
>> around until you figure out how it works, or you don't. If IF was
>> really that difficult, I would be more sympathetic, but it's just not.
>
> Sorry, Jimmy, but I think I agree with Aaron here. It seems to me (and I
> could be wrong -- feel free to correct me) that what you're saying boils
> down to, "We don't need to change anything. If people don't like it,
> it's their fault."
Well, that's a little more harshly than I would phrase it. :) I
wouldn't say it's anyone's "fault" if they don't like IF. There are
many worthwhile things I don't like, after all. I just think we should
be realistic and understand that just the necessity for reading is going
to make it a niche.
> IF is not difficult for you or me, because we understand the conventions
> of the interface -- but it's NOT a natural interface, and people are not
> born with an instinct that tells them how to use it.
Well, at some point nothing is natural. Reading a novel seems natural
to you only because the form has been so ingrained in you through
cultural exposure. Film is the same. I just think our emphasis should
be on teaching the conventions of IF, which work pretty well, rather
than distort those conventions in the hope of attracting folks who
probably still won't be that interested anyway.
> People won't read the help file, so you can't put instructions there.
> What needs to happen is that during the first few turns the game will
> steer the player gracefully in the direction of the type of input the
> software can understand. Typing "I don't know what to do" is perfectly
> rational, and the game response quoted above would be baffling to anyone
> who didn't know what to do.
>
> An easy fix (easy in TADS, at least -- I don't know about Inform) would
> be to set up a widget that would look for not-understood inputs during
> the first few turns and give one-line instructions in response rather
> than spitting out the parser's usual not-understood message.
Yes, I agree, and actually suggested this somewhere else. I'm not quite
as much of a hard-ass as all that. :) This is good because it works
with the player to teach her to play IF, rather than tries to guess what
she means by "ask sheila if david is kidding about grandma falling down
the stairs."
--
Aaron A. Reed
> When I think of IF "success," I think of independent
> films and music that appeal to niche audiences but nevertheless earn
> enough to give their creators a modest living. I don't think of
> Hollywood blockbusters, the Super Bowl, or World of Warcraft.
What about Harry Potter?
I may be wrong, but in my mind the potential audience for IF is as
large as the audience for fiction novels. Which is dwindling as of
late, but still pretty potent.
--Aaron
Jimmy Maher
> Here, Jimmy Maher <mah...@spamgrandecom.net> wrote:
>> Aaron A. Reed wrote:
>>> On Apr 12, 8:48 am, Jimmy Maher <mahe...@SPAMgrandecom.net> wrote:
>>>
>>> They hold if we're limiting our audience to "computer people who know
>>> the difference between MP3 and OGG." And the popularity of things like
>>> HomeStarRunner suggests that's an audience large enough to garner
>>> popularity on the scale of millions. But in my experience that would
>>> be less than half, possibly far less, of adults with computers.
>> Okay, fair enough. But I have to ask, even though it's going to make me
>> sound like even more of an elitist than this thread already has: can you
>> really picture those people playing and enjoying IF, even IF with the
>> various newbie-friendly features you and others have suggested?
>
> Yes.
You know what? I can too, now that I think about it. I'll concede this
point gladly. :)
Jimmy Maher
> On Apr 12, 10:22 am, Jimmy Maher <mahe...@SPAMgrandecom.net> wrote:
>> When I think of IF "success," I think of independent
>> films and music that appeal to niche audiences but nevertheless earn
>> enough to give their creators a modest living. I don't think of
>> Hollywood blockbusters, the Super Bowl, or World of Warcraft.
>
> What about Harry Potter?
Well, Harry Potter is not something anyone can really plan for.
Cultural phenomena like that just kind of happen on their own, just the
occasional indie flick breaks into the mainstream in a big way. It'd be
great if something like that would happen with an IF game, but I'm not
holding my breath...
> I may be wrong, but in my mind the potential audience for IF is as
> large as the audience for fiction novels. Which is dwindling as of
> late, but still pretty potent.
Yes, it's pretty potent. I'm not at all convinced that everyone or even
most people who read would enjoy IF, but I think some percentage would
if given the opportunity to experience it. I think this is the best
chance for really growing IF, in fact...
Aaron A. Reed
> Aaron A. Reed wrote:
> > Look at everything in the last few years that attains success on the
> > level that Jim wants. Youtube videos, for example. It takes 0 effort
> > or intelligence and about 2 seconds of time to start playing a Youtube
> > video. But if the first 30 seconds interest you, you might watch the
> > whole thing, even if it's half an hour long. I think IF needs to be
> > more front-loaded towards making those first few moments more
> > enjoyable.
>
> Not to beat this thing to death, but just a quick counter-example: World
> of Warcraft is enormously successful, and yet getting online with it is
> not effortless at all. They've been successful because they've done a
> good job of promoting their brand, letting people know WHY it's worth
> the effort to purchase and install a big piece of software, set up an
> account, etc. And then of course you reach a certain critical mass
> where "everyone is playing," and then the thing pretty much promotes
> itself.
Very good point.
> Maybe we could do a better job of marketing what we already have to offer...
I agree. I think things like Emily Short's Cover Art Drive are good
steps in that direction.
There was a thread a week or so ago about better promotion within the
community, too. I very much liked the idea of authors releasing new
games along with issues of SPAG.
I also think it would be cool if competitions had better "program
notes" that give people a reason to get interested in the games, more
like what film festivals do: see http://tinyurl.com/5x2hjl for an
example from Slamdance. I'd be willing to help set up such a page for
any future competition that wanted one.
--Aaron
Emily Short
> A better choice would be to go after a different market entirely:
> readers who are left flat by conventional computer games. I haven't
> figured out how to reach this market, though. Even with all the great
> outreach that has been done by Emily and others, their efforts have
> largely been through gaming websites -- Jay is Games, Play This Thing,
> etc. I think only a small minority of these folks are ever going to
> appreciate IF, not because they're stupid or somehow less than us but
> just because they aren't looking for the same thing we are. The trick
> is to reach readers rather than videogame players.
I think this doesn't have to be an either/or proposition, and in fact
I think it *shouldn't* be. The smart but not IF-aware audience might
be found in any of several places. It's clear from the feedback and
from the blog posts that crop up in other places saying things like "I
just heard about this game Photopia on Play This Thing! ..." that some
video game players are interested.
Other places where there's already successful if modest outreach going
on include new media studies, secondary education, and people
interested in Chicago history. (Though it's not clear how many of that
last group play on beyond 1893, I think it's pretty cool that they try
it.) There's also moderate room for outreach to video game *authors*
-- that is, those who work on professional games -- through articles
on Gamasutra and presentations at conferences. Historically, there's
also been a little outreach to people interested in puzzles and pen-
and-paper games; I haven't seen anything like this very recently, but
a few years back there were some bits about IF in Games magazine, for
instance. I wonder whether the sorts of people who like to solve one-
minute mystery puzzles might take an interest in An Act of Murder?
Each of these is a niche market, but my suspicion is that we'll get
farther by advertising in a series of niche places than by trying to
find some really wide-circulation venue through which to reach a lot
of people at once. That's only partly because of the competition for
attention; the other part is that even when there has been (say) an
article in a newspaper, it hasn't really led to a flood of new IF
players, and I think that's partly because the investment that readers
have in following up on newspaper articles is pretty low.
In all these cases, the trick about these things seems to be to find
the right conferences, forums, and magazines (or in Peter's case, the
right gift shops); convince a few specific people that IF will be of
interest to their readers/buyers; and work with them to find ways to
reach their audience.
So where do we go to reach conventional readers? In fact, as soon as I
pose the question, I see a problem with it: not all conventional
readers are in the same place! An interactive fantasy novel will be of
interest -- and reviewable -- in a different place from a short
interactive murder mystery. From *our* point of view, those are all
forms of IF, and fundamentally related, but for the reading population
they're different genres with different if overlapping fan bases.
Moreover, commercial publishing already involves spending large
amounts of money to get things into bookstores and keep them there in
locations and displays that will attract attention. So I think without
a substantial marketing budget, the in-store placement approach is not
likely to work unless we have a very specific, highly targeted piece
of IF that is appropriate to a specific specialty store. No big
display rack of the top 50 IF games in Barnes & Noble, is what I'm
saying. (I know Textfyre is likely to take a more aggressive approach
about getting things physically in front of people, and that's great
-- but for promoting freeware IF, it's not such a plausible approach.)
So here's a question: those of you who read reviews of standard
fiction, or magazines of short stories, are there any of these venues
where you think the editors might be possible to interest in reviewing
or mentioning an IF game? And which specific IF game(s) would be most
appropriate there?
I'll go first: I've thought for a while that it would be interesting
if one of us approached the online but paying SF zine Strange Horizons
( http://www.strangehorizons.com/Guidelines.shtml ) either with an IF
review or, more daringly, with an IF story as a submission to the
magazine. The latter would be a much harder sell, of course, and they
might flatly say that they weren't interested; I imagine it would need
to be short, browser-runnable, story- rather than puzzle-oriented, and
of a literary standard they found appealing to their readership. But
they *do* say that they're interested in hypertext stories, which
suggests an openness to the possibility of debuting alternative story-
telling technology, and the online presence means they're unusually
suited to running this kind of thing. (F&SF can hardly publish an IF
work.)
I've meant for some time to work on outreach through Strange Horizons
myself, but the pile of things I mean to work on is higher than I can
see over. Does anyone but me think it would be interesting? Anyone
want to try?
George Oliver
>
> So here's a question: those of you who read reviews of standard
> fiction, or magazines of short stories, are there any of these venues
> where you think the editors might be possible to interest in reviewing
> or mentioning an IF game? And which specific IF game(s) would be most
> appropriate there?
>
> I'll go first: I've thought for a while that it would be interesting
> if one of us approached the online but paying SF zine Strange Horizons
> review or, more daringly, with an IF story as a submission to the
> magazine.
Yeah, I've had the same thought. I'm pretty sure that SH has never
published a hypertext story, but from what I've heard they're still
open to it. Does anyone know the technical requirements for sending a
work of IF to a website, say, bundled with a Flash interpreter? Jay Is
Games just dealt with this for Lost Pig, right?
djmeister
Automapping! If you are really out for the casual player or the non IF
player, you should think about implementing that as an extension or
into the i7 code somehow. During beta testing I realized that all my
players had an instant enormous loss of motivation when confronted
with the necessity of needing pen and paper and drawing complicated
maps offline.
This is maybe not a suggestion for getting millions, but it could be a
way to getting at least a few dozens.
"An easy fix (easy in TADS, at least -- I don't know about Inform)
would
be to set up a widget that would look for not-understood inputs
during
the first few turns and give one-line instructions in response rather
than spitting out the parser's usual not-understood message."
This suggestion of Mr. Aikin I would like to see realized maybe as an
extension too. This could be very helpful for getting players who just
'stumbled' upon IF maybe during searching abandon- or freeware game
sites.
Lg, Stefan
James Jolley
Aaron A. Reed wrote:
Maybe the "prologue" part of the game (before
> the first few command prompts) should contain a few prompts and
> responses, to get new players used to what it is and what it's for
> before they're suddenly expected to use one. (Maybe this isn't an idea
> so much as something I saw somewhere before; has some game done this
> already?)
>
> --Aaron
Hi,
Seastalker by Infocom did something like this and moonmist I think.
Seastalker did it so new kids could get into typing things into the
parser. It worked relatively well as I remember.
-James-
Emily Short
> For instance, auto-completion of the command line? For instance, a
> built-in auto-mapper that would show the player exactly where she has
> been and what she has discovered there?
Both of those are offered to some extent by the ADRIFT runner, and I
am not crazy about them; I'm also not the hypothetical novice player
either, though.
The reason I steered clear of offering grand-gesture suggestions is
that I really think a lot can be achieve with IF through better
craftsmanship rather than through more fancy interface devices, and I
think focusing on the interface devices can be a distraction from the
game and story design. ADRIFT games do both of the things you name, if
you play them in the Windows runner, which I pretty much never do.
Interpreters for PDAs offer compass roses and point-to-paste-to-
command-line features, as I understand it. But beyond a certain point,
you can't conceal the fact that this is a parser-based game without
making it, well, *not* a parser-based game.
We've had a number of IF pieces in the semi-recent past that did
ambitious things with the interface, either in folding in multimedia
features or in offering the player new, less demanding ways to
interact: Moments Out of Time 2, Ferrous Ring, Deadline Enchanter
(sort of), Jealousy Duel X, Little Falls, Ekphrasis, Dead Cities,
Ecdysis. I'd be interested in hearing other people's impressions. My
own have ranged from luke-warm to positive on the multimedia stuff
(sound is used quite effectively a few times in Little Falls, as I
discuss in my review, and Ekphrasis is so much about the visual arts
that it wouldn't make much sense without the pictures).
Attempts to let the player pick between several kinds of interaction,
or to make puzzles semi-optional, though, I generally dislike and find
weak. Dead Cities with hyperlink options I found significantly *less*
gripping than Dead Cities without. Ferrous Ring was an even more
unfortunate hybrid, because I found the puzzles too hard or underclued
to work out in parser form, but when I switched to menu or walkthrough
mode, I didn't always follow the story very well -- admittedly it's a
surreal one -- and I felt much less investment in the game as a whole.
(I'd say Ferrous Ring comes closest to exemplifying your hypothetical
'solve' command.)
I think what this boils down to is that it's really hard to design a
good work that allows the player multiple options about how to
interact. A good game will be designed to support its preferred
interaction format really well; if you then glue on other kinds of
interaction, they're likely to suffer by comparison; and if you try to
provide for both at once, you get a horse-designed-by-committee effect.
James Jolley
Jim Aikin wrote:
> Jimmy Maher wrote:
> Sorry, but I do. :) Honestly, this just sounds to me like someone who
> > wasn't really all that interested to begin with. What do you do with
> > any other computer program when you first encounter it? You fiddle
> > around until you figure out how it works, or you don't. If IF was
> > really that difficult, I would be more sympathetic, but it's just not.
>
> Sorry, Jimmy, but I think I agree with Aaron here. It seems to me (and I
> could be wrong -- feel free to correct me) that what you're saying boils
> down to, "We don't need to change anything. If people don't like it,
> it's their fault."
>
I have to say that all this talk of making IF easier seems rather a
waist of time to me. As a blind player who started out with a copy of
Zork I and had absolutely no idea what to do when I was in school, all
I can say is that it's easy to learn. If a description says to the
east is a place, type east and see what happens. Really, parsers are
now good enough to make any game instantly easy to pick up and play.
We make IF complex because we can. Kids play video games don't they?
They learn how to use an XBox 360 don't they? Give people credit,
there not all stupid.
Best
-James-
Emily Short
It did, though JIG wrote their own Flash interpreter for the purpose.
(Jay is pretty interested in reviewing more IF in the future, as that
amount of effort probably suggests.) I don't know whether he plans to
let other people use it in the long run, but it's currently their own
in-house thing.
On the other hand, last I saw Flaxo was coming along nicely, so if I
*were* doing this, I might check with Peter Mattson about what state
it's currently in. (Saving, restoring, and scrollback would be nice to
have.) I'd then set up the game I wanted to submit on my own website
(at a special URL with no other links incoming, so it was not
'published' yet), and email the editors of SH inviting them to try it.
If they liked it, we could talk about how to set it up on their
servers.
That's pretty much what I did with my IF submission to Up Right Down,
and they had no complaints, though in the end they opted just to
provide a link to my webpage rather than set up hosting for it
themselves. But it worked fine. Admittedly that piece was really tiny
and simple, so the lack of save/restore wasn't a problem.
Emily Short
> > This is of course a tricky line to tread because you don't want to
> > reward them *so* much for unproductive actions that they spend the
> > whole time licking scenery instead of having a gripping time with your
> > plot and puzzles.
>
> Or maybe that's not a big concern.
>
> I have long thought that _Myst_ became a multiple-zillion-seller
> mostly because it was great at entertaining people who were failing to
> get anywhere in it. If you had no grasp of the interaction model, no
> notion of how the initial island was laid out, no spatial sense, no
> idea of how to solve the first puzzles... you could still click
> randomly on things and feel sucked into a gorgeous and varied world.
Fair point.
Jimmy Maher
> Other places where there's already successful if modest outreach going
> on include new media studies, secondary education, and people
> interested in Chicago history. (Though it's not clear how many of that
> last group play on beyond 1893, I think it's pretty cool that they try
> it.) There's also moderate room for outreach to video game *authors*
> -- that is, those who work on professional games -- through articles
> on Gamasutra and presentations at conferences. Historically, there's
> also been a little outreach to people interested in puzzles and pen-
> and-paper games; I haven't seen anything like this very recently, but
> a few years back there were some bits about IF in Games magazine, for
> instance. I wonder whether the sorts of people who like to solve one-
> minute mystery puzzles might take an interest in An Act of Murder?
Tabletop RPG players are another natural outreach target: not so much
Dungeons and Dragons players perhaps, but players of some of the more
niche-oriented, literary games such as Call of Cthulhu. The game I'm
working on is based on a COC adventure, and I will definitely be trying
to promote it through website like yog-sothoth.com.
Of course, the tabletop RPG market is itself not exactly thriving, so
the potential is somewhat limited there.
> So where do we go to reach conventional readers?
That's what I haven't really figured out. :) My own reading has been
almost exclusively classics for the last few years, and I don't really
know where readers of the genre literatures -- which I think are
definitely the best target for IF -- tend to hang out these days.
> So here's a question: those of you who read reviews of standard
> fiction, or magazines of short stories, are there any of these venues
> where you think the editors might be possible to interest in reviewing
> or mentioning an IF game? And which specific IF game(s) would be most
> appropriate there?
This might be one of those places where being commercial would be a real
advantage, because you could run advertisements for your games in print
SF magazines and so on. (Maybe David is already planning on doing this.)
Victor Gijsbers
> Thank you for saying this, Victor. I've felt this way for years.
> Built-in hints will do the job -- at least if people know to use the
> hints and if the hints are well-written.
Yes, there are basically two ways to go. Either write games where there
is automatically a clear way of progress--which means that you have to
write quite an atypical IF game, though there precedents*. Or write a
more typical IF game, but use really good in-game hints.
I think the first option should be explored more, and the second should
become more standard--though we have already moved into the right
direction there, I think. Many recent games have excellent built-in
hints, and this _really_ helps make them more accessible.
Now, some IF purists might not like this trend, and their reaction is
not unreasonable. You're definitely moving away from what traditional IF
play is like when built-in and even pro-active hints become something
you are basically _supposed_ to consult when frustration rears its head.
That frustration used to be the _motor_ of IF play, and that which gave
the moment of breakthrough its power.
I think there will always remain a market for that kind of 'less
accessible' IF; but it's not going to be the greater market.
(* Precedents for the first option are most of Photopia; Whom the
Telling Changed; most of The Baron; Deadline Enchanter; but also, in an
entirely different sense, most of All Things Devours. And in again
entirely different sense, both of my WIPs, which rely on exploration and
tactical combat rather than puzzles, and are thus more akin to CRPGs as
far as making progress is concerned.)
> In one of the Spring Thing games, the hint system gives you a couple of
> vague nudges and then punts without ever telling you what you need to
> do. This is infuriating!
It is. Very.
Regards,
Victor
Greg Boettcher
> Someone has recently released a Totally Unbelievable Game (text-based,
> of course). Word on the street has reached fever pitch, and the servers
> are chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
Being well marketed. Press releases, free copies sent to reviewers,
free demo, promotional web site, etc. That's way more important than
actual quality, for what you're talking about.
Indeed, if sales are your main goal, it may be a waste of time to
focus on quality. I mean, Howard Sherman keeps on making games; he
must be getting enough sales to justify that.
Greg
Greg Boettcher
wrote:
I guess I need to qualify this. You weren't really talking about
sales, but just getting a large audience. But marketing and promotion
is still the biggest factor for getting big audiences, even if your
game is freeware. If I mentioned sales, it's because people tend to be
unmotivated to do promotional grunt work if there's no monetary
reward.
Greg
Mike Roberts
> Here's a blue-sky question that might be worth kicking around.
>
> course). Word on the street has reached fever pitch, and the servers are
> chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
Interesting question! I dusted off my copy of PROGNO, the famous 1965
TOS/360 program by Don Knuth (you know, the one that predicted everything
from the 1966 Nigerian coup to the sub-prime mortgage meltdown just this
year) to see what it thinks. Of course, I only have the unclassified
civilian version (private ownership of the full version would be highly
illegal, so if a copy had ever accidentally fallen into my hands, I would
have been obliged to delete it at once, wink, wink :), but even so it's been
a great investment tool (ask your stock broker about it!) and has saved me
countless hours by warning about serialized sci fi TV shows subject to early
cancellation (Jericho, anyone?).
So, I interpreted your query into a PROGNO virtual punched card input deck,
fired up my TOS/360 emulator, and let the batch job run overnight. Here are
the official, inerrant results... (Note that any prices are in 1965 US
dollars, so you may need to adjust for inflation.)
1. It will have a mildly controversial theme, on the level of The Da Vinci
Code or Zumanity - just enough to exercise a fringe group or two and make
mainstream readers feel "edgy".
2. Many people in the raif community will argue that it's not really IF
because it has a simplistic parser.
3. It will have an on-line social component external to the game.
4. It will have a compelling central mystery that is only 90% solved in the
course of play.
5. It will have puzzles.
6. Most people will run it on handheld devices.
7. Business managers will complain that people are playing during meetings.
Symantec will sell a $3.01 add-on to Norton AntiVirus that blocks play
during business hours.
--Mike Roberts
George Oliver
> 7. Business managers will complain that people are playing during meetings.
> Symantec will sell a $3.01 add-on to Norton AntiVirus that blocks play
> during business hours.
Which, incredibly, is still below the commonly accepted 'buy now'
price point of $19.95 once you adjust for inflation, proving that
Symantec is indeed an offshoot of the Masonic Templars.
Mike Roberts
news:ftp2hg$c6t$1...@aioe.org...
> My question is: What are the features of this [totally unbelievably
> great] game?
>
> The parser would have to be much more flexible and "intelligent" than
> current parsers. Current IF parsers embody some fairly stringent (and
> hidden) assumptions about the input. People who know IF understand these
> assumptions; others don't.
>
> We could make a long, long list of grammatical constructions novice
> players might try: “is david smiling?” “[tell david] grandma fell down the
> stairs and broke her ankle.” “I want the ice cream cone.” “I want david to
> give sheila the ice cream cone.” “ask sheila if david is kidding about
> grandma falling down the stairs.” In a game that achieves runaway success,
> I believe the parser would need to be able to handle all of these
> structures (and hundreds of others) gracefully.
Hundreds? More like thousands, hundreds of thousands, millions, billions...
The problem with real natural language is that it's a combinatoric system
with a really big set of things to combine and a really elaborate set of
rules for combining them. The idea of a big list of syntax templates has
been tried before - that's how Eliza worked - and it just doesn't work very
well.
>What's the rope made of?
>Is the rope strong enough to hold my weight?
>Can the rope hold me?
>If I climb down the rope, will it break?
>How strong does the rope look?
>Does the rope look strong enough to hold me?
>I want to climb down the rope, but only if it's strong enough.
>Do I think the rope is strong enough to climb down?
You're right that current IF parsers are hard for newbies to come to grips
with. Unfortunately, I just don't think it's feasible to solve that by
making them more elaborate. (Short of honest-to-Turing AI, anyway - *that*
would be great, but I can tell you one sure way *not* to get there: enlist a
huge team of beta testers to think up syntax templates.)
Fortunately, there *is* one truly intelligent actor in the equation - the
user. I think the more fruitful approach is to take advantage of the user's
smarts. It's relatively easy for humans to learn to deal with computers on
their terms; so far the opposite has proven impossible.
A great way to train the user is to plop down a 50-page manual and tell them
to master it before starting the game. That's basically what we do now.
But now that you point it out, that might not be the best way to reach a
wider audience.
One thing to look at might be real-world computer voice-response systems.
Try calling up United Airlines Reservations some time and booking a flight
through their IVR - they use entirely voice inputs in their current system,
not touch-tones, and it actually works pretty well. They achieve this feat
not by having an awesomely studly AI with a million syntax templates; quite
the opposite. They achieve it by very tightly limiting the inputs at each
stage, and letting you know exactly what you're expected to say next.
I'm not saying that a commercial IVR ought to be the template per se. The
key thing I'm getting at is that the IVRs succeed by basically do on-the-job
training to guide the user through the system. They don't ever just stop
and say "Give me a command!" They always let you know what your options
are. The thing that makes IF hard for newbies isn't that the input language
is so difficult (because it's not), it's that no one bothers to explain the
input language to them - at least, not in a way that harmonizes with their
impatience to start playing.
--Mike Roberts
Lazarys
something to interactive fiction, BUT STILL KEEPING IT INTERACTIVE
FICTION, is that for the most part we live in too much of an instant
gratification society. In general, people want lots of noise and
pretty pictures, animations, etc. and no level of complex parser and
high quality prose is going to break it out of an essentially niche
market of those Luddites who actually like to read and solve puzzles.
There are however some things that could be done to widen the
audience a little. There is a concept out on the web called
alternate reality gaming (ARG) that is based on the premise of web
and real life puzzles, multimedia and text woven together to tell a
multiplayer story and work towards solveingthe story together. (For
more specific information on this genre, see www.unfiction.com. They
are often well done viral ads (the movie AI hosted the first (known
as the BEAST now) and Audi did a very good one a while ago).
I've played in a couple of the stories over the years, and to me the
biggest frustration with them is that you need to get into them at
the start for the most fun, but proper ettiquette on these games
hides the entry point, rather than making straightforward
announcements.
What makes them pertinent to where I am going is that while they do
combine multimedia components, they are essentially a story. What
they offer that IF doesn't usually, are multimedia puzzles and
graphics supporting the text. For example, if a story was a Celtic
Fantasy, It might start with "pages" with celtic knotwork and
appropriate fonts decorating the prose. As the story advanced into the
realms of fairy,the stylesheet would lend itself to fonts and
decoration setting flavor. Plunging into a dungeon becomes dark and
oppressive. Graphics can supply some of the clues, and music as well.
Combine this with the well behaved game thoughts above (package
theinterpreter and game together; help pertains first and foremost to
the package, not just the player., etc.) and they could reach a
much wider audience without have to make huge parser advances.
Just my thoughts.
Mike Roberts
> _The_ psychological barrier that interactive fiction has and which almost
> no other type of game has, is that in all those other games, there is
> always something you can do which you know will probably move the game
> forwards to a good conclusion, and at least will move the game forwards.
This is something I've been trying to get a handle on for a long time. I
don't think it's precisely a matter of moving the game forwards, but it's
related.
Some game designers talk about "process" - the little mechanical chores that
you do in the course of play; the micromanagement aspects of play. The
really successful single-player computer games all seem to have deeply
involving processes that induce that "flow" state in the player, where you
go into a time vortex that makes hours pass like seconds. That doesn't seem
to happen in IF.
> In Monopoly, all you have to do is pick up the dice and roll.
Yeah, but in IF all you have to do is type NORTH, SOUTH, EAST, WEST... You
can almost always do *something* - being stuck doesn't cut off access to the
process of play. But somehow the process elements in IF don't seem
conducive to flow state, so when you do get stuck in the broader sense, it
does tend to feel like the micromanagement processes are pointless. In
other kinds of single-player computer games you can still get sucked into
the process even when you're not making any forward progress on the
big-picture goals.
I don't have any great ideas about how to "fix" this, if indeed it's
something in need of fixing, but I have some vague intuitions about it. My
feeling is that many IFs would benefit from a sort of intermediate layer of
interactivity, between the overarching goal of progressing through the story
and the micromanagement level of NORTH, TAKE, DROP. Something that involves
mini-goals that are always open and achievable, open-ended so you can never
really finish them, but at the same time measured so that you can always
feel like you're making progress, and having some kind of effect in the
broader game. I think character-development activities (in the RPG sense)
are a good place to look for this kind of thing - earning money, improving
your character's firearms accuracy through practice, etc.
--Mike Roberts
Mike Roberts
> Interactive Fiction, played on-line and facilitated by a human story
> teller (or multiple story tellers) is a goal easily achieved within
> the limitations of existing software. Examples and variations of this
> form of story telling are undoubtably already being used, though the
> fact that I can't cite any of them without additional research tells
> me something about their popularity.
Wow - I've been thinking about exactly this sort of project for the past
year or so. Here I've been smugly going around thinking it was my great
out-of-the-box idea to solve the parser AI problem by substituting a real
human for the parser.
I haven't had time to develop the idea at all yet, but I did do a little
looking around for existing systems. I couldn't find anything quite like
it, much to my surprise, so either (a) it's not interesting to anyone, or
(b) it just hasn't been tried yet.
Here's the outline of the project I've had in mind:
1. Server software would facilitate play. Users would ideally connect
through web browsers, but maybe special clients would be needed.
2. A game session is a closed-ended multi-player affair: a group of players
and one GM arrange to meet up on-line at a particular time to play for a
particular interval. (It's not a MMORPG - it's just an on-line version of
meeting at a friend's house for an RPG session.)
3. Authors would prepare games a la table-top roleplaying scenarios ("D&D
modules"). The degree of completeness and automation in the scenario is up
to the author; some scenarios might be purely descriptive text, others might
be almost like complete IF programs, with lots of programmed responses.
4. The server software gives the GM access to objects, object properties,
etc. The GM can update object states to keep track of game conditions. The
scenario can pre-program certain actions and conditions to help the GM by
providing suitable state changes and descriptive text.
5. All narrated text is under the GM's control. The GM can copy-and-paste
from prepared text written by the author and from generated text produced by
the author's program, but is free to improvise.
Now for the big blue-sky goal. The hope is that this scales up to the point
where, at any given time, there are enough people on-line that you can
always just log in and relatively quickly find a GM and a group of players
ready to go with a game. That is, the "appointment" aspect in #2 goes away
at some point - you can just log in and expect to find a quorum. There'd be
a reputation system for GMs, players, and games, to make it easier to find a
group and game that interests you.
--Mike Roberts
ChicagoDave
> Redesigning the UI in ways that would change that reaction is arguably
> achievable. At least, I think it's worth discussing.
The user experience requires a state of the art upgrade from what we
in the community are used to in playing games. The standard cross-
platform interpreter, take the best one we have, and it still is far
from what I would consider "state of the art". Big buttons, colors,
professional help, hints, mapping, note-taking, and stuff can all make
a user experience more interesting. But in my mind it's not
necessarily to enhance, but to hide the fact that our user is reading
and typing. We need to get them to _forget_ what they are doing and if
you show them a big screen of text, well, they're not going to be
fooled.
In emily's response she cites a laundry list of very important items.
A detailed help system. I would add that hints for the entire game
need to be implemented and in an entertaining manner, similar to the
snarky hints that were common in InvisiClues. A map of the game. One
of the things we can do with today's tools (PhotoShop) is create an
appealing view of our setting as a map. In the tools we're using in
Textfyre, we can even make this interactive, animated, and a part of
the experience.
But more importantly, emily mentions the thoroughness in which
responses are implemented within the story itself. I believe this is
the most important thing we can do. It's also the hardest. In order to
raise the level of detail responses, you need solid testing and
relentless updating of every single potential response within the
game. Each response has to move the story forward, deflect the user,
inform the user, but at all times it must _entertain_ the user. This
is a tall order for the hobbyist author. I have thought that we could
develop tools to track this sort of thing. That would be enormously
helpful. Within Textfyre we have not had time to develop such tools.
We're doing it the same way anyone else would have to do it.
I think original artwork throughout a game can add enormous value and
Zarf's Myst comment is dead-on. If we can strive for something like
Myst, that would bring in a ton of new users.
David C.
Jim Aikin
>
> Well, that's a little more harshly than I would phrase it. :) I
> wouldn't say it's anyone's "fault" if they don't like IF. There are
> many worthwhile things I don't like, after all. I just think we should
> be realistic and understand that just the necessity for reading is going
> to make it a niche.
Fair enough. I don't think reading is the sticking point; I think the
command line is the sticking point.
> Well, at some point nothing is natural. Reading a novel seems natural
> to you only because the form has been so ingrained in you through
> cultural exposure.
Unclear. Storytelling is a basic human instinct. However, _nonlinear_
storytelling (as practiced in the modern novel, film, and IF) does
require cultural exposure.
--JA
JDC
I wonder if some of this could be done using Guncho. One player could
be designated as GM, and have access to a set of additional meta-
commands (along the lines of standard debugging commands, like
purloin, etc.). The parser would attempt to handle things like in
normal IF to catch most actions. Instead of the game giving default
responses or parser errors in response to unanticipated inputs,
though, the player's command would be sent to the GM, who would then
be able to either translate it into a standard action (if it's just a
matter of non-recognized input), or manually alter the game-world and
issue text responses to the players to handle actions not anticipated
by the "module" and so forth. I haven't played around with Guncho
much, yet, but this certainly seems doable. It would maybe not scale
as well as you would like, but it would give a central location for
groups to meet up.
-JDC
Jim Aikin
>
> I'm not saying that a commercial IVR ought to be the template per se. The
> key thing I'm getting at is that the IVRs succeed by basically do on-the-job
> training to guide the user through the system. They don't ever just stop
> and say "Give me a command!" They always let you know what your options
> are. The thing that makes IF hard for newbies isn't that the input language
> is so difficult (because it's not), it's that no one bothers to explain the
> input language to them - at least, not in a way that harmonizes with their
> impatience to start playing.
Excellent point.
Here's a fairly bad idea that would sort of address what you're talking
about: I was thinking about introducing a sidekick character in the
opening scene. This character could say things like, "Hey, I wonder
what's up at the other end of the meadow. Why don't we go north?" Or,
"Wow, look at the portrait of the duchess!" As the player's input
becomes more appropriate, the sidekick might taper off and eventually
say, "I have to go do my homework now. You're on your own."
--JA
Lazarys
> Jimmy Maher wrote:
> Fair enough. I don't think reading is the sticking point; I think the
> command line is the sticking point.
>
The command line really isn't that much of a sticking point. While
most people seem to prefer high end graphics, most use the command
line interface in chat panes and text messages. The problem comes in
getting people into that mindset, and forming the game where they fall
into that conversational mode (There seems to be a "conversational
mode of thinking, because I have noticed over the years that even if
someone is texting "proper English", when deeply engrossed in
chatting, they (and I for that matter) will make homophone errors
they would never think of making when typing a paper. In that mode the
lack of graphics isn't such a big deal because they are "talking and
listening" not reading and writing.
Existing Parsers can accomplish it, with work on the authors parts,
though a Short, Nelson or Plotkin type story gets closer to that state
of mind than one framed after early infocom games (nothing wrong with
those style games; I'm just referencing the whole wider audience
angle).
Making as many synonyms (and homophones) for actions and objects can
truly make a world of difference. I recently downloaded a free game
trial from "the only commercial IF company" and was more than a little
frustrated with the old game of "find-the-verb-or-noun". I knew what
I wanted to do, but the environment wasn't letting me,even though
the words I was trying were used in descriptions of the objects I
wanted to manipulate.
Emily Boegheim
> On Apr 12, 10:03 am, Jim Aikin <midigur...@sbcglobal.net> wrote:
>
> > An easy fix (easy in TADS, at least -- I don't know about Inform) would
> > be to set up a widget that would look for not-understood inputs during
> > the first few turns and give one-line instructions in response rather
> > than spitting out the parser's usual not-understood message.
>
>
> --Aaron
Since Aaron is obviously itching to mention this and can't, I had
better.
Blue Lacuna is designed to ease new players into playing it, with a
tutorial mode, highlighted keywords, and verbless commands. Typing
"door", for instance, will mean either "examine the door", "go through
the door", or "talk about the door", depending on what highlight style
the word "door" uses in the game text. (If you try a non-highlighted
noun, it will default to "examine", unless you're in the middle of a
conversation, when it will default to "talk about".)
Emily
Jim Aikin
>
> I should have linked to this page in my previous post, but what you just
> described is indeed exactly how it was done:
> http://www.mobygames.com/game/dos/gateway-2-homeworld/screenshots
>
> These are screenshots from Gateway 2. The interface was configurable
> (verb list and inventory were optional). Also note that it implemented
> both "ask/tell about" conversation as well as menu-based conversation
> (along with portraits).
Thanks for the link! That's exactly what I had in mind. (I'm just 15
years too late....)
--JA
Jim Aikin
>
> However, the fact that there are now several Flash-based IF players on
> the web indicates this is probably not the real problem.
I'd be curious to learn more about this. I found JACL by googling. Its
syntax looks ... rather primitive.
I'm wondering whether one could build ("one" meaning somebody -- I'm not
enough of a code wrangler to try it) a Flash-based IF interpreter that
could load z, glulx, and t3 game files. That would be interesting.
--JA
Jim Aikin
>
> Since Aaron is obviously itching to mention this and can't, I had
> better.
>
> Blue Lacuna is designed to ease new players into playing it, with a
> tutorial mode, highlighted keywords, and verbless commands. Typing
> "door", for instance, will mean either "examine the door", "go through
> the door", or "talk about the door", depending on what highlight style
> the word "door" uses in the game text.
At first I found this navigation system quite nice, but as I got deeper
into the game I found that I was actually having a harder time
visualizing the model world because the author avoided describing things
as being to the north, south, east, or west. Drawing a map turns into a
sort of creative/random enterprise....
--JA
Jim Aikin
>
> So here's a question: those of you who read reviews of standard
> fiction, or magazines of short stories, are there any of these venues
> where you think the editors might be possible to interest in reviewing
> or mentioning an IF game? And which specific IF game(s) would be most
> appropriate there?
Since you asked, I've been mulling over the idea of approaching Gordon
Van Gelder at F&SF about doing something IF-related on his website. I
haven't done it yet, because I'm not sure what to suggest.
The plusses: It's a real magazine, and he has just bought two of my
stories, so I think he'd at least consider it if I proposed it.
The downside: It's not much of a website (http://www.sfsite.com/fsf/). I
don't know how much traffic it gets.
More to the point, I don't know exactly what I would offer or suggest to
him. A nice little anthology of fantasy games, bundled with terp
software and a PDF how-to-play -- he might go for it, if someone gave
him an html page and all he had to do was upload it. And if the games
are pro-quality fantasy stories, of course.
If anyone is interested in this idea, perhaps we should explore it via
email. I would request (though of course I can't enforce it) that I be
the one to contact Gordon, since I'd like to manage my contacts with him
rather than have people he doesn't know coming at him from five
directions and dropping my name.
> I'll go first: I've thought for a while that it would be interesting
> if one of us approached the online but paying SF zine Strange Horizons
> ( http://www.strangehorizons.com/Guidelines.shtml ) either with an IF
> review or, more daringly, with an IF story as a submission to the
> magazine.
There's also helixsf.com. I know very little about them, except that
they're an online zine and they pay authors.
--JA
Dino
>>> I think more ways to recognize broader ranges of input is a key place
>>> where IF is lacking here. I remember one comment I read from a college
>>> student briefly exposed to IF that went along the lines of: "I saw the
>>> first prompt and typed 'I don't know what to do.' The game said back
>>> 'I only understood you as far as wanting to inventory.' So I gave up."
>>> And I don't really blame him.
>>
>> Sorry, but I do. :) Honestly, this just sounds to me like someone who
>> wasn't really all that interested to begin with. What do you do with
>> any other computer program when you first encounter it? You fiddle
>> around until you figure out how it works, or you don't. If IF was
>> really that difficult, I would be more sympathetic, but it's just not.
>
> Sorry, Jimmy, but I think I agree with Aaron here. It seems to me (and I
> could be wrong -- feel free to correct me) that what you're saying boils
> down to, "We don't need to change anything. If people don't like it,
> it's their fault."
>
> IF is not difficult for you or me, because we understand the conventions
> of the interface -- but it's NOT a natural interface, and people are not
> born with an instinct that tells them how to use it.
>
> People won't read the help file, so you can't put instructions there.
Just some points that sprang to mind regarding reading the text file,
and about fiddling with the parser for people not familiar with IF. You
may find the connection between the points is vague, but it should give
you the gist of my opinion.
* People unfamiliar with IF should be presented, via text file
accompanying the game and/or the first command in-game (being told to
type HELP or ABOUT for instructions if you are a first time player),
with a small sample transcript which shows how the parser operates, and
how the conversation system in the game works.
* I think the basic verb-noun parser, and n s w e for movement, etc, is
are conventions that are intrinsic for efficient interaction with IF
worlds. (Here I raved on justifying my statement, but then deleted it
because I was getting bogged down and straying too far.)
* If you can't pick up the required convention for successful
interaction in IF after reading a small transcript (or you are not
motivated to read an example of play before you even begin), then you
are not the type of person that is going to be able to get to the point
where they are comfortable with the parser conventions to be able to
appreciate what IF is.
* There are better analogies, but for example: in chess, you must learn
how each piece moves just in order to play, but in order to appreciate
the game of chess, you must further learn how to use the basic moves in
order to manipulate the battle with your opponent. The appreciation
comes from the struggle occuring within the confines of the game
structure - such as it being turn based, only certain moves recognised,
etc.
* IF can minimally be described as a turn based, text input system (if
it wasn't, I'd hesitate to call it IF). Within that system, a range of
input conventions exist. These conventions must be learned in order to
appreciate what IF actually is. The onus is on the new player to learn
these conventions if they wish to be able to appreciate IF.
Andrew Plotkin
>
> Now, some IF purists might not like this trend, and their reaction is
> not unreasonable. You're definitely moving away from what traditional IF
> play is like when built-in and even pro-active hints become something
> you are basically _supposed_ to consult when frustration rears its head.
> That frustration used to be the _motor_ of IF play, and that which gave
> the moment of breakthrough its power.
Primus, if that's the case, then the IFComp has long since moved away
from what traditional IF play is like. (Since most entries have either
hints or walkthroughs, and players seem to feel no shame about using
them when the two-hour limit looms.) And I hear the IFComp is this
massive weight which shapes all of modern IF around it, right?
Secundus, I call a moratorium on declaring what "IF purists" don't
like. Unless you're the IF purist in question. It's fair to say "Some
people dislike..." if history supports that, but don't go boxing up
their reasons as a straw ideology.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the borogoves..."
*
It's a nice distinction to tell American soldiers (and Iraqis) to die in
Iraq for the sake of democracy (ignoring the question of whether it's
*working*) and then whine that "The Constitution is not a suicide pact."
Damien Neil
> I have long thought that _Myst_ became a multiple-zillion-seller
> mostly because it was great at entertaining people who were failing to
> get anywhere in it. If you had no grasp of the interaction model, no
> notion of how the initial island was laid out, no spatial sense, no
> idea of how to solve the first puzzles... you could still click
> randomly on things and feel sucked into a gorgeous and varied world.
I've known many people who did exactly that. They never really got
anywhere in the game, but they enjoyed playing tourist in the world.
I think there's a point to be made involving the old Sierra adventures
as well--brutally unforgiving, with incomprehensible puzzles, and yet
very popular in their day. What all of them gave you was a lively and
interesting world that was fun to play around in.
- Damien
Cubist
[much is snipped]
> Here's a blue-sky question that might be worth kicking around.
>
> Someone has recently released a Totally Unbelievable Game (text-based,
> of course). Word on the street has reached fever pitch, and the servers
> are chugging trying to keep up with the download requests.
>
> My question is: What are the features of this game?
My 2 cents' worth: Such a game should *absolutely ERADICATE* that
damned guess-the-word problem which haunts so many existing instances
of IF. The game should, in realtime, keep/build/update a list of *all*
possible words which are valid input at any specific moment. This list
will of course include all one-word commands which are valid at
whichever moment -- but in addition, this list will *also* include all
words which are the *first* words of any *multi*-word command that
happens to be valid at any given moment. When the user is responding
to the game, he types his first word normally, one character at a
time. With the first character, the game puts up a scrollable list of
valid words, 'centered on' whatever that first typed character is. The
second typed character, together with the first typed character,
either will or will not be the first two characters of any valid word.
If the two characters *don't* begin a valid word, the game beeps and
doesn;'t let that second typed character show up in the commandline;
otherwise, the scrollable word-list filters out all words which
*don;t* begin with the two typed characters, and the remainder of the
word-list re-"centers" on that valid two-starting-character combo.
Alternately, the player can use the arrowkeys to scroll thru the
'popup' list of valid words.
This process continues until the player either (a) hits Return
(meaning, he's finished his input), or else (b) hits the spacebar
(meaning, the player's finished typing that word). The player *will
not* be able to type a space *unless* the existing string of typed
characters is one of the words that's valid in the current situation;
if he types a space anyway, the game beeps and refuses to let him do
it. Similarly, the game won't allow the player to type a Return
*unless* the existing string of typed characters is a valid command.
In all other cases, the game just provides a 'popup menu' of all the
valid words which begin with whatever the current string of typed
characters.
I believe that this scheme would completely eliminate the situation
where the user has NO BLEEDIN' *IDEA* what words the game will even
bother to acknowledge, while at the same time preserving the user's
freedom to try *anything* the game will allow.
mikea...@hotmail.com
news:d355b3ac-3845-421b...@a9g2000prl.googlegroups.com...
> I believe that this scheme would completely eliminate the situation
> where the user has NO BLEEDIN' *IDEA* what words the game will even
> bother to acknowledge, while at the same time preserving the user's
> freedom to try *anything* the game will allow.
This strikes me as the most maddening game scheme I could ever imagine.
Having no idea of what a game will accept is part of the process of playing
the game and learning. This whole 'guess the verb' thing that people keep
bitching about speaks more to the writer of the game (who didn't accept
enough input variations) rather than as a problem with the input medium
itself.
Victor Gijsbers
S
P
O
I
L
E
R
Did you find the compass in the hut? You can use it to change the
directions to compass directions.
Regards,
Victor
Doug Egan
Mike,
I think you have a brilliant idea, and the skills to implement it.
The one case I was already aware of live author-mediated IF is
neither quite so ambitious, nor particularly satisfying. "Choose Her
Fate" www.chooseherfate.com was a serialized on-line novel written
last year by ??? (I can't find the author's name on the website).
After each episode, readers were offered a limited number of branching
options to choose from, and then polled. The most popular choice
determined the direction of the story in the following installment.
Dave
Victor Gijsbers
> Here, Victor Gijsbers <vic...@lilith.gotdns.org> wrote:
>> Now, some IF purists might not like this trend, and their reaction is
>> not unreasonable. You're definitely moving away from what traditional IF
>> play is like when built-in and even pro-active hints become something
>> you are basically _supposed_ to consult when frustration rears its head.
>> That frustration used to be the _motor_ of IF play, and that which gave
>> the moment of breakthrough its power.
>
> Primus, if that's the case, then the IFComp has long since moved away
> from what traditional IF play is like. (Since most entries have either
> hints or walkthroughs, and players seem to feel no shame about using
> them when the two-hour limit looms.) And I hear the IFComp is this
> massive weight which shapes all of modern IF around it, right?
All of that is true.
> Secundus, I call a moratorium on declaring what "IF purists" don't
> like. Unless you're the IF purist in question. It's fair to say "Some
> people dislike..." if history supports that, but don't go boxing up
> their reasons as a straw ideology.
Okay--I probably misjudged the precise meaning of "purist". So please
read "people who prefer IF as it was written and played in the first
decades of its existence" for "some IF purists".
Regards,
Victor
David Fisher
news:12080151...@kangaroo.ozonline.com.au...
>> course). Word on the street has reached fever pitch, and the servers are
>> chugging trying to keep up with the download requests.
>>
>> My question is: What are the features of this game?
>
> of inputs (even if the response is to guide the player
> towards a more limited range of inputs) -- an underlying
> world model that lets the player do everything that
> "makes sense" to them (within reason). Combined with
> a wonderful plot which is reasonably resistant to
> corruption by the player's actions, at the same
> time allowing the player to influence the plot in a
> logical way.
...
[other stuff about simulation]
I'm OK with posting things without any particular response (no
obligations!), but I am curious about anyone's reactions to these ideas ...
Here's another thought:
I haven't tracked it down yet, but somewhere I read a great quote about the
interface of IF implying a lot more than it provides -- the invitation to
"type something" is a very broad promise to a new player. Non-IF games with
much more restricted interfaces tend not to have boundaries that are as
potentially frustrating as IF (to new players, at least), because they
promise less.
David Fisher
George Oliver
<davidfis...@australiaonline.net.au> wrote:
> Here's another thought:
>
> I haven't tracked it down yet, but somewhere I read a great quote about the
> interface of IF implying a lot more than it provides -- the invitation to
> "type something" is a very broad promise to a new player. Non-IF games with
> much more restricted interfaces tend not to have boundaries that are as
> potentially frustrating as IF (to new players, at least), because they
> promise less.
>
This sounds like one of the main points of Jeremy Douglass' recently
released dissertation, _Command Lines_.
I think what it comes down to is that the TUG would be released for
mobile devices (i.e. cell phones) and so the interface follows from
that. Compare with, for example, the new Japanese mobile phone novels:
http://en.wikipedia.org/wiki/Mobile_phone_novel.
Jim Aikin
line somewhere in this thread; your scheme would take this somewhat further.
On the whole, the pop-up list thing strikes me as more intrusive than
helpful. I agree with Mike that a good author will minimize the problem
by implementing as many of the sensible synonyms as he or she can think of.
But I'm tending to disagree with Mike that hidden commands are a good
idea. To the extent that one would like to attract a wider audience, I
suspect that having an explicit list of ALL of the available commands
(and perhaps of all the commands that can be applied to a given object)
would be a friendly thing to do. I would put it in a separate window
(or, in a simple interface, make it available via a "commands" verb), as
I feel this is less intrusive than messing with the command-line.
This relates to the puzzle/story dichotomy, I think. If you're
envisioning IF as a game full of puzzles, then "What could I possibly do
with this huge wrinkly lump of plastic?" is part of the game. Having the
software tell you that "inflate it with the pump" is a valid command
really does spoil the fun, because part of the puzzle is figuring out
that the wrinkly lump of plastic is a rubber raft. On the other hand, if
you're envisioning IF as a story that requires reader participation,
then the interesting thing is what will happen when the player inflates
the raft and floats down the river.
Two different philosophies of IF design. Neither is right or wrong.
--JA
Taylor Vaughan
> But I'm tending to disagree with Mike that hidden commands are a good
> idea. To the extent that one would like to attract a wider audience, I
> suspect that having an explicit list of ALL of the available commands
> (and perhaps of all the commands that can be applied to a given object)
> would be a friendly thing to do. I would put it in a separate window
> (or, in a simple interface, make it available via a "commands" verb), as
> I feel this is less intrusive than messing with the command-line.
I don't know about listing the commands available on any given object.
I think that might make things a bit too easy. With the inner tube
example you gave, if the player knows that they can "inflate" the
plastic then they don't even have to figure out that it's an inner
tube. Now, using something like that as a sort of hint system might
work though.
But making a list of all the verbs needed in the GAME could, I think,
be a good idea. It might also make the game too easy but it would
DEFINITELY cut down on the guess-the-verb problem.
JDC
<davidfis...@australiaonline.net.au> wrote:
> Here's another thought:
>
> I haven't tracked it down yet, but somewhere I read a great quote about the
> interface of IF implying a lot more than it provides -- the invitation to
> "type something" is a very broad promise to a new player. Non-IF games with
> much more restricted interfaces tend not to have boundaries that are as
> potentially frustrating as IF (to new players, at least), because they
> promise less.
GTA recently linked to some remarks of Chris Crawford, which can be
found at:
http://www.watercoolergames.org/archives/000916.shtml
While I don't agree with many of his ideas, the following does seem
worth considering (and is apropos your comment):
"(4) A Toy Language for a Toy World
You can't do language on a computer, not easily. The Sapir-Whorf
hypothesis: language is closely tied to the perceptual reality of the
speaker. Language mirrors reality.
Natural language on a computer would require reality as well. And
reality is hard to get inside a computer.
When you create software, you are creating a model universe -- a toy
reality. If a toy reality fits, a toy language can fit."
I really don't like the idea of explicitly telling the player what
commands are available, though. Some of my favorite IF moments have
been those flashes of insight (think Spider and Web, Rematch, etc.)
where figuring out what you want to do is the challenge (and would be
totally spoiled if you were given a list of choices). I think this is
one of the strengths of IF which really distinguishes it. Even though
a game is necessarily only going to respond to a limited set of
commands, the player doesn't _know_ the limitations, and pushing them
is part of the fun.
-JDC
Daryl McCullough
>Here's the outline of the project I've had in mind:
>
>1. Server software would facilitate play. Users would ideally connect
>through web browsers, but maybe special clients would be needed.
>
>2. A game session is a closed-ended multi-player affair: a group of players
>and one GM arrange to meet up on-line at a particular time to play for a
>particular interval. (It's not a MMORPG - it's just an on-line version of
>meeting at a friend's house for an RPG session.)
>
>3. Authors would prepare games a la table-top roleplaying scenarios ("D&D
>modules"). The degree of completeness and automation in the scenario is up
>to the author; some scenarios might be purely descriptive text, others might
>be almost like complete IF programs, with lots of programmed responses.
>
>4. The server software gives the GM access to objects, object properties,
>etc. The GM can update object states to keep track of game conditions. The
>scenario can pre-program certain actions and conditions to help the GM by
>providing suitable state changes and descriptive text.
>
>5. All narrated text is under the GM's control. The GM can copy-and-paste
>from prepared text written by the author and from generated text produced by
>the author's program, but is free to improvise.
I love this idea. There is a sense in which an online version of a RPG
might be better in some ways than a traditional one in which all the
players gather around a table to play. One little advantage is that
it becomes possible for a PC to disguise himself as an NPC (or vice-versa)
without all the other players knowing about it. Also, if there is some
weird circumstance in which communication between player characters is
cut off (a player character is unconscious, or is kidnapped, or something),
it's a lot easier to enforce these communication limitations online.
--
Daryl McCullough
Ithaca, NY
Adam Thornton
Doug Egan <dgen...@hotmail.com> wrote:
>My first thought, after reflecting on your description of the "ideal
>parser", is the story of Automon Chess Player (also called the
>"Mechanical Turk"), a chess-playing machine exhibited from 1770-1854.
>Although the inside of the machine was decorated with elaborate
>clockworks to fool observers, the machine was actually operated by a
>human chess master, hidden inside.
>http://en.wikipedia.org/wiki/The_Turk
>
>Interactive Fiction, played on-line and facilitated by a human story
>teller (or multiple story tellers) is a goal easily achieved within
>the limitations of existing software. Examples and variations of this
>form of story telling are undoubtably already being used, though the
>fact that I can't cite any of them without additional research tells
>me something about their popularity.
Yeah, but there's something to be said for the theory that IF exists
largely because people wanted to be able to play D&D without getting
four or five people over at their house for several hours.
I mean, in a very basic sense, what IF is about is computer-driven
storytelling. If you're willing to take the computer out of it, then
why not get together for a) an evening's storytelling, or, at the more
structured end, b) a good old-fashioned pen-and-paper RPG?
Adam
Adam Thornton
Jim Aikin <midig...@sbcglobal.net> wrote:
>Of course, these days many of us have large monitors. Normal-sized text
>on a large monitor would be difficult to read, because the lines would
>be too long. (Some research has been done on this, I believe. The point
>is, if your eye has to jump back too far at the end of each line, it's
>hard to find the correct spot to jump to. There's a maximum tolerable
>character-height-to-line-length ratio, though I don't know what it is.)
There's a reason, I think, that traditional terminals are 24x80. That
generally (with line wrap at 72 characters) works just about right for
me.
Adam
Adam Thornton
Jimmy Maher <mah...@SPAMgrandecom.net> wrote:
>Tabletop RPG players are another natural outreach target: not so much
>Dungeons and Dragons players perhaps, but players of some of the more
>niche-oriented, literary games such as Call of Cthulhu. The game I'm
>working on is based on a COC adventure, and I will definitely be trying
>to promote it through website like yog-sothoth.com.
How about over at indie-rpgs.com? I mean, it doesn't get more niche and
literary-oriented than that.
>Of course, the tabletop RPG market is itself not exactly thriving, so
>the potential is somewhat limited there.
Well, S. John Ross would disagree.
Me, I'm not sure. I think that the same thing is likely to happen in
tabletop RPGs that happened in IF: the paying market takes a substantial
downturn, and there's enough free excellent stuff out there that
everyone gets habituated to a zero price tag.
Adam
Adam Thornton
Mike Roberts <mj...@hotmail.com> wrote:
>I'm not saying that a commercial IVR ought to be the template per se. The
>key thing I'm getting at is that the IVRs succeed by basically do on-the-job
>training to guide the user through the system. They don't ever just stop
>and say "Give me a command!" They always let you know what your options
>are. The thing that makes IF hard for newbies isn't that the input language
>is so difficult (because it's not), it's that no one bothers to explain the
>input language to them - at least, not in a way that harmonizes with their
>impatience to start playing.
You might want to look at ZOIP. Zork-over-IP, but it's Zork hooked up
to a (freeware) IVR. The hard part is teaching the IVR system to
recognize the various meaningful sounds that refer to objects in the
game, and I don't think it was ever playable past getting to the Troll
Room.
Which is pretty much what you're saying. Which suggests that CYOAs, if
cleverly disguised, are the wave of the future. Which makes me want to
stick my head in the oven (or a fork in the toaster, etc).
Adam
Adam Thornton
Mike Roberts <mj...@hotmail.com> wrote:
>Wow - I've been thinking about exactly this sort of project for the past
>year or so. Here I've been smugly going around thinking it was my great
>out-of-the-box idea to solve the parser AI problem by substituting a real
>human for the parser.
>
>I haven't had time to develop the idea at all yet, but I did do a little
>looking around for existing systems. I couldn't find anything quite like
>it, much to my surprise, so either (a) it's not interesting to anyone, or
>(b) it just hasn't been tried yet.
c) you're not looking in the right place.
Seriously, what you describe sounds like a large part of the online
component of D&D v4. And although I am deeply dubious about the system
overall, and about the ability to have a fun online D&D game (since for
us a lot of the fun is sitting around the table smack-talking), I have
to admit some of the collaborative online tools look pretty neat.
Adam
Lazarys
> Interesting idea, Cubist. I had suggested auto-completion of the command
> line somewhere in this thread; your scheme would take this somewhat further.
>
> On the whole, the pop-up list thing strikes me as more intrusive than
> helpful. I agree with Mike that a good author will minimize the problem
> by implementing as many of the sensible synonyms as he or she can think of.
>
> But I'm tending to disagree with Mike that hidden commands are a good
> idea. To the extent that one would like to attract a wider audience, I
> suspect that having an explicit list of ALL of the available commands
> (and perhaps of all the commands that can be applied to a given object)
> would be a friendly thing to do. I would put it in a separate window
> (or, in a simple interface, make it available via a "commands" verb), as
> I feel this is less intrusive than messing with the command-line.
>
> This relates to the puzzle/story dichotomy, I think. If you're
> envisioning IF as a game full of puzzles, then "What could I possibly do
> with this huge wrinkly lump of plastic?" is part of the game. Having the
> software tell you that "inflate it with the pump" is a valid command
> really does spoil the fun, because part of the puzzle is figuring out
> that the wrinkly lump of plastic is a rubber raft. On the other hand, if
> you're envisioning IF as a story that requires reader participation,
> then the interesting thing is what will happen when the player inflates
> the raft and floats down the river.
>
> Two different philosophies of IF design. Neither is right or wrong.
>
> --JA
>
and get a list of the objects that were visible/knowable at this point
in the game at this location, so that the reddish-kiln-dried-masonry
may be referred to in a command as a brick, or as "chunk of wall" can
make a world of difference. Likewise, without giving clues away a
list of verbs, or adverbs and adjectives could help a player. If the
list also included words known that are not applicable to the puzzle,
but that the parser knows how to act upon, it adds a level of
immersion for the user to try, with some possibly comic/tragic
ramifications.
Josh Westbury
> To me the problem with the premise Jim put forward is that by doing
> something to interactive fiction, BUT STILL KEEPING IT INTERACTIVE
> FICTION, is that for the most part we live in too much of an instant
> gratification society. In general, people want lots of noise and
> pretty pictures, animations, etc.
$2 from a high lurk:post reader, this is a really interesting
discussion.
Important point in there somewhere - the million-download IF game has
to compete with entertainment that either just gives people instant
gratification, or worse - abuses the need for it, to keep them hooked.
Fruit machines are the most obvious example - they're designed by the
best behavioural scientists money can buy - but videogames do it too,
rewarding players in ways that keep them playing. Lost-style US TV
series with mysterious slow-burning plot arcs (personally I blame the
X-Files for this trend) are basically doing the same thing, rarely and
apparently randomly rewarding viewers with trickle-feed plot (for any
Skinnerians out there this is variable ratio operant conditioning)
thus you have to watch in case you miss the next twist.
The IF game will have to do this too in some way. There are benign
ways of doing this - to my mind there's nothing wrong with a learning
curve that rewards players for solving gradually more difficult
puzzles. Good writing remains important & IMHO this is one part of the
Harry Potter phenomenon - the reader is rewarded (for example with
new, humorous magical concepts) as they progress through the books.
It's also helpful to remember that HP (amongst other things) proves
you can still have addictive mass appeal with an essentially pure-
prose product so there's (some) hope for IF. Perhaps finding other
less benign ways to keep players hooked might be needed too. I have to
agree with the people that have talked about marketing strategies -
sadly this is likely to be crucial too.
Adam Thornton
Josh Westbury <joshua....@btopenworld.com> wrote:
>Lost-style US TV series with mysterious slow-burning plot arcs
>(personally I blame the X-Files for this trend)
Blame _Twin Peaks_.
Or, maybe, _The Prisoner_. Which isn't exactly US.
Adam
Ron Newcomb
<davidfis...@australiaonline.net.au> wrote:
> [other stuff about simulation]
>
> I'm OK with posting things without any particular response (no
> obligations!), but I am curious about anyone's reactions to these ideas ...
They are pretty much the same ideas I hold, about the likely benefits
of providing a baseline world sim. But I think the argument for/
against played out in February in response to your own Magnetic
Scrolls RAIF post, David, which set off the big Emergence thread on
Emily's blog.
I don't see anything significant has happened, or been built, in the
meantime to change anyone's opinion on the matter.
-Ron
http://emshort.wordpress.com/2008/02/13/emergent-puzzle-solutions/#more-216
Ron Newcomb
> couldn't be done with books or video games.
I've seen a lot of ideas thrown around in this thread, but I disagree
with the ones that want to make I-F just like some other form: a
graphical video game, a D&D play-by-email, etc. These suggest that I-
F can never be popular. I remember the same negativity said or
implied of video games a couple of decades ago, that they'll always be
for kids and never mainstream adults.
The popularity of books and email suggest reading, thinking, and
typing are not the barriers here. The computer's ability to
understand is the barrier, and "understand" touches a lot: the
parser's ability to understand the player's desire, the worldsim's
ability to carry it out, the AI's ability to compensate, etc.
Good stories are already told with the existing mechanics, but
strictly on the author's terms: the player's expressiveness -- his
allowed interactivity -- is curtailed to the point of being
negligible.
-R
Mike Roberts
>>I haven't had time to develop the idea at all yet, but I did do a little
>>looking around for existing systems. I couldn't find anything quite like
>>it, much to my surprise, so either (a) it's not interesting to anyone, or
>>(b) it just hasn't been tried yet.
>
> c) you're not looking in the right place.
>
> Seriously, what you describe sounds like a large part of the online
> component of D&D v4.
Thanks - I'll have to check it out.
--Mike Roberts
Mike Roberts
> Josh Westbury <joshua....@btopenworld.com> wrote:
>>Lost-style US TV series with mysterious slow-burning plot arcs
>>(personally I blame the X-Files for this trend)
>
> Blame _Twin Peaks_.
>
> Or, maybe, _The Prisoner_. Which isn't exactly US.
I'd personally lay the blame at the feet of one Charles Dickens.
--Mike Roberts
Jed
alert telling me that the magazine had been mentioned here.
Emily Short wrote:
> > So here's a question: those of you who read reviews of standard
> > fiction, or magazines of short stories, are there any of these venues
> > where you think the editors might be possible to interest in reviewing
> > or mentioning an IF game?
I can't speak for the SH reviews-department editors (our departments
are pretty autonomous), but I know that in the past they've been open
to reviewing computer games of various sorts, as well as other "new
media" forms. Not sure whether that's still true. I recommend
contacting them (http://www.strangehorizons.com/guidelines/
reviews.shtml) to see if they'd be interested in either reviewing an
IF piece or running a review written by one of you.
She added:
> > I've thought for a while that it would be interesting
> > if one of us approached the online but paying SF zineStrange Horizons
> > (http://www.strangehorizons.com/Guidelines.shtml) either with an IF
> > review or, more daringly, with an IF story as a submission to the
> > magazine.
and George Oliver replied:
> Yeah, I've had the same thought. I'm pretty sure that SH has never
> published a hypertext story, but from what I've heard they're still
> open to it.
Both true. We've only had about half a dozen hypertext submissions
over the nearly eight years we've been publishing the magazine, and so
far none of them have quite worked for us (though a couple have come
close). But we remain very interested in the idea.
In all the hypertext stories we've seen, the only interaction has been
clicking links, which is what we were expecting when we first said
we'd be interested in hypertext stories. I don't think we've ever
talked about running a real IF piece (of the sort involving typing in
commands and such); I'm open to the idea (not sure what my co-editors
would think), but as y'all have noted, there are various hurdles that
would need to be dealt with.
I've only intermittently dipped my toes into the IF world in the past
few years (grew up playing Colossal Cave, have enjoyed the few more-
recent IF pieces I've tried (like Gostak and Blue Chairs), but am
mostly out of touch), so I'm not up on the state of the art wrt things
like the Flash interpreter that George O. mentioned. Is there an
interpreter that has no copyright/licensing restrictions? How big is
it? (By which I really mean, is it a long wait to download it if
you're on a dialup line?) ...We at SH try to make sure that our
content is readable/viewable in all modern browsers, including text-
based ones and screen readers, and I don't think the magazine has ever
published anything that used Flash; but it's possible we could make an
exception for an IF piece we liked well enough. We would almost
certainly not want to publish something that required downloading the
IF work and an interpreter to the desktop (though we could consider
making a downloadable version available as an alternative in addition
to a browser-based version).
Two other potential difficulties, off the top of my head:
1. We usually pay for fiction by the word. I've never looked at the
total wordcount of all user-visible strings in an IF work, but I'm
guessing it's pretty large relative to the wordcount of all the
strings that any given user is likely to see in a single time
through. So an IF work that we could afford by our usual rates might
well be too short (in terms of playing time) to be interesting/fun.
But I'm talking through my hat; feel free to educate me.
2. Another interesting question is how to submit. We usually ask
hypertext-story authors to post the story in HTML on a password-
protected area of a web server and send us the URL and password. I'm
guessing that submitting IF to us would work similarly. But it's
conceivable that we could work out other arrangements if need be.
...I obviously don't normally read this newsgroup, so if you have
followup questions you want to ask, or responses to my post here,
please drop me a note in email. You can use the address on this
message for specific tech questions and answers. If you want to ask
about submitting something to us, though, or anything else having to
do with the magazine's fiction department, best to drop a note to
fic...@strangehorizons.com, with a subject line beginning with QUERY:
thanks,
--jed
Mike Roberts
>>Interactive Fiction, played on-line and facilitated by a human story
>>teller (or multiple story tellers) is a goal easily achieved within
>>the limitations of existing software.
>
> largely because people wanted to be able to play D&D without
> getting four or five people over at their house for several hours.
I think that's definitely part of it. The human-moderated IF idea I
outlined upthread really hinges upon getting a critical mass of players onto
the system so that at any given time you can log in and fairly quickly find
a pick-up game to join. *That* would be a key difference from in-person
D&D, since for that you do need to get several people's schedules to line
up. I haven't thought enough about the mathematics of it to know whether or
not this part is feasible, but my intuition is that it's not at all
far-fetched.
> I mean, in a very basic sense, what IF is about is computer-driven
> storytelling. If you're willing to take the computer out of it, then
> why not get together for a) an evening's storytelling, or, at the more
> structured end, b) a good old-fashioned pen-and-paper RPG?
Two answers. First is the above - it's the scheduling issue; you have to
schedule the in-person play, but with the on-line system you can play at
your convenience (assuming that the numbers do work out).
Second, there's a subtle but important difference in the nature of play
between IF as we know it and tabletop RPGs, at least to the extent I've
experienced them. In IF, the players solve the problems that advance the
plot; in RPGs, the player characters solve the problems that advance the
plot. This difference seems really important to me in shaping the type of
stories the two branches tend to deal in. IF gravitates more toward the
types of stories you find in static fiction, probably because the
protagonists are allowed to be exceptional - protagonists in IF are never
limited by their character's skill levels or intelligence points or Mythos
lore points or the like, since they can solve any puzzle that the player can
figure out. To do the same thing in an RPG you'd effectively have to max
out your PCs' skills/intelligence levels so that they'd get an automatic
"idea" or "notice clue" success every time their controlling player has an
idea or notices a clue. This is anathema to RPG play, but the idea that you
have to pass a statistics roll to even try a puzzle solution you just
thought of is equally anathema to IF play. None of this is to say that you
couldn't design an RPG with IF-like mechanics; it's just that people don't
seem to do that.
And anyway, IF as we know it isn't really computer-driven storytelling, is
it? It's more like computer-mediated storytelling: the story is still
driven by the author, and the computer is the device and medium for playing
back the story. So a networked human-moderated game doesn't seem to me to
change that much, except for substituting a more intelligent playback
device.
--Mike Roberts
stu...@animats.net
> Aaron A. Reed wrote:
>
> I'd be curious to learn more about this. I found JACL by googling. Its
> syntax looks ... rather primitive.
I prefer the description, "intentionally simplified." :)
Having a couple of games on the web though does give an interesting
insight into the playing habits of the casual visitor in the form of
the web logs. The average visitor stays for about five moves: four of
them are commands that current parsers can't possibly hope to
understand and the final one is usually obscene.
I couldn't agreed more with Victor's early five-star post about the
need to have the feeling of constant progress. It has got me thinking
about an IF translation of sudoku-type puzzles where every move you
make brings the final solution that one step closer and more obvious.
Prod at a puzzle and you should almost always be rewarded with more
information that makes the solution that little bit more obvious as
you accelerate your way towards final victory. This is what makes
Sudoku puzzles so much fun for so many people. With IF puzzles you
tend to try many-many ideas with little or no reward until the one
time you get it right. This is very discouraging and not much fun to
the average person. I'm not a bit fan of clues coming from outside
the game itself. I feel they are simply a band-aid fix for a problem
with the design of the puzzle in the first place. And yes, I am just
as guilty as anyone else of using this technique. I guess what this
translates to is that I feel that IF is still too much of an attempt
to simulate real life in some way where maybe abstracts puzzles
designed from the ground up to entertaining, then wrapped in a story,
would be better.
On different note, I also agree that leaving the keyboard to draw a
map is a frustrating loss of immersion. I, too, have thought a bit
about adding modules for auto-mapping, but I'm now tending towards
crafting geographies that are so intuitively laid out that a map just
isn't required at all. If fact I'm thinking about maybe even moving to
a system where the relationship between the locations isn't even told
to the player, they simply spend the whole game using "go to", or a
shorthand equivalent there of. As well as eliminating the need for a
map I think this also allows location descriptions to read more like a
novel and less like a game.
Tab completion, on the other hand, I now consider essential. Now that
I am used to having it I feel completely lost using interpreters
without it. Can you imagine the frustration of using bash without tab
completion? IF uses an almost identical interface in principle yet we
more often than not don't have it.
Stuart
Dannii
<davidfis...@australiaonline.net.au> wrote:
>
> I'm OK with posting things without any particular response (no
> obligations!), but I am curious about anyone's reactions to these ideas ...
>
> Here's another thought:
>
> I haven't tracked it down yet, but somewhere I read a great quote about the
> interface of IF implying a lot more than it provides -- the invitation to
> "type something" is a very broad promise to a new player. Non-IF games with
> much more restricted interfaces tend not to have boundaries that are as
> potentially frustrating as IF (to new players, at least), because they
> promise less.
>
> David Fisher
I too would think that the problem of what to do on the very first
turn is a big one. But I don't think having an interface to list all
the verbs etc would help. It should use the same parser with normal
verbs, just the introduction text would make it extremely obvious what
you should do. As in Myst when you're invited to click on the book.
The interface there is no different to anywhere else in the game, but
it works. I think changing the prompt text for the first few turns
might well help with this.
Michael Lodge
> This relates to the puzzle/story dichotomy, I think. If you're
> envisioning IF as a game full of puzzles, then "What could I possibly do
> with this huge wrinkly lump of plastic?" is part of the game. Having the
> software tell you that "inflate it with the pump" is a valid command
> really does spoil the fun, because part of the puzzle is figuring out
> that the wrinkly lump of plastic is a rubber raft. On the other hand, if
> you're envisioning IF as a story that requires reader participation,
> then the interesting thing is what will happen when the player inflates
> the raft and floats down the river.
>
> Two different philosophies of IF design. Neither is right or wrong.
>
Two thoughts here.
First, you need to be very careful about removing puzzles from the game
because if you remove them all it seems to me that you have also removed
the "game" aspect as well.
For example, imagine a first person shooter with a highly detailed and
interesting plot. You may know that to continue the plot you have to
find your way up to the top story of a building full of heavily-armed
enemies. Of course, you want several interesting things to happen on the
way.
The enemies provide a challenge. To get to your goal you have to work
out how to get past them. Since I said that this was a first person
shooter, this will probably require a certain amount of manual dexterity
to be able to move your character and shoot fast and accurately enough.
The challenge is what makes it a fun game. The payoff is the ongoing
storyline, and the promise of future challenges. If you remove all of
the enemies, and therefore the challenge, do you still have a game? When
playing a game without puzzles, I tend to feel like I am just going
through the motions and if there is no challenge to get the rest of the
story, you may as well just ask the player to turn the page or sit and
watch the next movie scene for all of the difference it makes.
I like the idea of a narrative at war with a crossword (thanks to Graham
Nelson). Each author chooses how much of either to put into their game,
but both are required to make a decent game. If you remove the puzzles,
or make them so obvious that there is nothing left to work out, then you
are back to static fiction. The obvious exception to this would be a
story that actually contained a dozen or so different narratives that
made a real difference. That may be interesting, but this would be MUCH
more work to write, and not something that I have seen done very much.
The other thought: (yes, all that above only counts as one)
The idea of revealing all accepted commands seems to be aimed at
eliminating the guess-the-verb puzzles. I have to ask, how often have
you really encountered guess-the-verb?
In my experience, the vast majority of times I encounter an apparent
guess-the-verb puzzle I am simply trying the wrong solution. I accept
that there are definitely some games that do have this problem but I
firmly believe that it is far more a problem of perception than reality.
People in some cases blame the parser or the game for their own
inability to work out the puzzle, and the puzzle is what drives the
program. I know this because I have made this mistake myself.