Conversation systems
Victor Gijsbers
experiences with teaching interactive fiction:
====
As I briefly alluded to earlier, menu-based conversations were seen as
horribly distracting and completely breaking of the story. One of my
authors referred to these as "gaping potholes in the narrative flow."
(One recent game we looked at was "Fate" and while people really enjoyed
the story, they didn't like the menu-based conversations. Although, that
said, they did recognize the challenge in trying to present how to
communicate with NPCs.)
====
Until now, I always thought that the general negative attitude to
conversation menus in the IF Community was simply a matter of having bad
prior experience with the "lawn mower" problem and/or being used to a
different style of conversation. But these people were not used to a
different style of conversation, and did not have bad prior "lawn mower"
experience. Apparently, if I understand it correctly, the very
appearance of a menu was felt as intrusion into the reading/playing
experience.
Another reason for my positive attitude towards conversation menus is
that the game which I consider to be not only the single best computer
game ever, but also the best extant work of interactive storytelling
(and the greatest influence on my own IF), the 1999 CRPG "Planescape:
Torment", makes extensive use of conversation menus. If the best uses
it, it must be good, no?
But perhaps there _is_ a difference between having a conversation menu
is a parser-based game where the player is constantly typing text, and
having a conversation menu in a game where all input is given by moving
and clicking the mouse. It might be the transition from free input to
choosing options that is jarring.
Is this the case? If not, is there another reason why nobody has ever
complained about the conversation system in Torment (or Baldur's Gate,
or any other of these types of games) but many people dislike
conversation systems in interactive fiction?
(Did people dislike the menus in Fate where you get to decide whether
you _really_ want to cut the pixie's wings, say, as much as they
disliked the conversation menus? What about the point in The Baron where
the game asks you, through a menu, to state your reason for killing the
young wolf?)
More practically, though, what am I going to do in my next game? Things
would be easy if there were some other conversation system that I really
liked, but there isn't. When I play a game that uses a standard ASK/TELL
system, I always have two problems: (1) I'm typing in a lot of things
that the NPC cannot converse about, which leads to irritation and (I
would argue) a more jarring sense of disconnection with the game than
the appearance of a conversation menu, and (2) I'm saying things I do
not want to say, because I can only choose topics, but I cannot choose
_what_ to say about that topic.
There is the style of conversation where you get hints about which
topics you could talk about; a particularly good example of this was
"The Elysium Enigma". The game makes a suggestion like "You could ask
Fred about nuclear war, his problems with Edmund, or whether he likes
first flush Darjeeling better than Earl Grey." This very much lessens
problem 1 above, since the player generally won't be trying to think of
something to say. Problem 2, however, is unaffected; and it also seems
to me that this approach is a step towards what people experience as the
limiting nature of conversation menus. You still can try to say other
things, but every time you choose one of the options from the list, you
are just navigating through a conversation menu with a bad interface.
(Typing "ask fred about about nuclear war" is worse than simply typing
"1", isn't it? Or is the increased sense of complicity worth the extra
typing effort?)
One possible solution to Problem 2, which Emily Short recently brought
up, is to include extra commands for the player so that they can
"recommend X", say, instead of "talk about X". There are many related
ways of doing this: we could let the player choose a mood,
Varicella-style; we could let the player "agree" and "disagree" whenever
the NPC made a claim; and so on.
The problem with this approach seems to be twofold. (1) Every one of
these additions will only be useful in a limited number of situations,
and will leave other points in the conversation where it would be
interesting to be able to choose between two things to say unresolved.
(2) Once implemented, you'd have to make sure that these commands work
correctly for _every_ point in _every_ conversation. That's... not good.
What I'm currently doing is implementing a hybrid ASK/TELL-Menu system.
The player can bring up topics, to which the character either responds
directly, or which bring up a conversation menu with further options.
The player can also use a general "talk to" command to start a
conversation menu about a predefined (by the author) topic, so that they
won't get stuck in trying to think of the right topic.
The idea here is that the main conversation menu is the only thing you
need to complete the game, but that there is a lot of (optional)
interesting and useful information 'hidden' behind the ASK/TELL system.
I am not sure that this is a good idea.
If the very appearance of a conversation menu is something bad, then
this is bad idea. Also, a hybrid system may be confusing; or it might
manage to bring the _bad_ aspects of its components together.
So I'm a bit stumped, and I would very much like to talk to you about
this problem.
Kind regards,
Victor
Jeff Nyman
news:47d553de$0$14347$e4fe...@news.xs4all.nl...
Unfortunately I have zero knowledge or experience with "Planescape: Torment"
and I'm curious as to its menu system
> But perhaps there _is_ a difference between having a conversation menu is
> a parser-based game where the player is constantly typing text, and having
> a conversation menu in a game where all input is given by moving and
> clicking the mouse. It might be the transition from free input to choosing
> options that is jarring.
Expectations, maybe? Perhaps it's like reading what you think is a linear
novel and all the sudden realizing, ten pages in, that it's a choose your
own adventure for a bit. But then it goes back to being linear for awhile.
So maybe the prose-heavy nature of textual IF sets the expectation that you
are "going along" with a story and thus when elements "intrude" (based on
some people's opinion), that seems to jar someone out of the story, ruin the
flow, whatever. I don't know. As I type that out it seems a bit simplistic
but maybe there's something to it. I would think it's almost certain that
styles of game that "read like a book" will engender different tolerances
for certain game mechanics like menus than would a game like "Planescape:
Torment." While I've never played that particular game, I can tell you the
game mechanics I would tolerate for something like "Deus Ex" (a graphical
shooter but with a fairly involved story line and narrative path) are very
different from what I would tolerate for a work of textual IF that tried to
tell a similar story. (Here I'm thinking of the conversation menus that you
could use in "Deus Ex" as well as certain conversation nodes that would
immediately commit you to an action -- and sometimes its consequences.)
> What about the point in The Baron where the game asks you, through a menu,
> to state your reason for killing the young wolf?)
Okay, I need to try that one. That sounds interesting because "stating a
reason" does speak to the player's motivation. What often came up in my
groups were the *player character's* motivation separate and distinct from
the player's. In other words, even if you -- the player -- force a certain
action (killing a wolf, in this case), how does your player character
respond to that, not only in idea but in action? Does how your player
character look at the world color the types of actions you -- as the
player -- take? If so, could that be incorporated into the narrative rather
than just made a stark menu choice?
I've found that a lot of what we dealt with, pretty much across the board,
had to do with telling stories differently by *not* conflating the player
and the player character, which then opened up more narrative possibilities
and thus more ways to at least think about handling things like this, such
as giving reasons for action. I'll definitely check out "The Baron."
> system, I always have two problems: (1) I'm typing in a lot of things that
> the NPC cannot converse about, which leads to irritation and (I would
> argue) a more jarring sense of disconnection with the game than the
> appearance of a conversation menu, and (2) I'm saying things I do not want
> to say, because I can only choose topics, but I cannot choose _what_ to
> say about that topic.
That first point did come up a lot, where an NPC just says "I don't know
about that." or, worse, the game itself just says "Bob doesn't appear
inclined to talk about that." The issue here was whether or not it was good
to have NPCs that could have some sort of "aggravation" factor. When the
factor was exceeded, they started responding in slightly different ways than
they would otherwise, eventually clamming up. This is, again, a simplistic
take on what was in fact weeks of discussion. But the idea was modeling NPCs
with things like an "attention span", a "mood", and so forth. Then making it
clear to the player in a graduated way that further asking along these lines
might be not only useless but maybe even counter-productive.
For example: "I didn't know Jack well, but from what I knew I was pretty
sure if I kept this up, he was going to stop talking to me at all. I
couldn't afford that."
So the player gets a contextually relevant "hint" as part of the narrative.
Although it was also agreed that the narrative couldn't "freeze up" such
that the NPC never decided to talk again. So there had to be a way of
reducing the "aggravation" factor. [Side note: this was the one area where
'simulationist' aspects were seen as welcome -- with people.]
I'm not sure about (2) in the sense that I don't know what you mean. But, in
case I do actually get it, I'll say that one example we did as part of my
classes was to take the game "Tapestry" and re-do some elements of it. What
people want to do was be able to ask Morningstar about a lot of topics
(death, fate, destiny, life, etc) and have him respond very contextually
about these topics, partly based on what you already asked him about. So if
you asked him about death *and then* your own death, he would respond
differently than if you asked about your death first *and then* about death
in general. Further, Morningstar would try to engage in some discussion
about topics up to a point -- until his "attention span" got too low or
"aggravation factor" got too high. The key is that the player could
gradually learn about these elements as well as how Morningstar viewed them.
Even further, the player character was allowed to have *his* own take on
what Morningstar was saying. How was "his own take" expressed to the player?
As part of the narrative.
In general I found people were much less willing to tolerate lack of context
to NPC responses and actions but were much more willing to tolerate an
obvious "game point" where the NPC stops responding. Everyone realized that
the game couldn't allow the NPC to talk about anything and everything,
always varying the response. So the trick was to concenrate on those parts
that most effectively kept the story idea clear in the player's head *and*
most effectively kept the narrative from slowing down too much. In other
words, the game mechanics didn't allow the player to stray too far but still
allowed them to explore some side avenues.
> One possible solution to Problem 2, which Emily Short recently brought up,
> is to include extra commands for the player so that they can "recommend
> X", say, instead of "talk about X". There are many related ways of doing
> this: we could let the player choose a mood, Varicella-style; we could let
> the player "agree" and "disagree" whenever the NPC made a claim; and so
> on.
As far as this, the "recommend X" might be okay but when I tried similar
solutions (like "think about") I found this just as intrusive for players.
They kept trying "recommend" or "think about" every single time anything
happened. (It was likened to pressing a refresh key to see if your browser
page has changed at all.) But -- that being said -- a way to have
recommendations appear as part of the narrative was generally welcomed. Or
even having the NPC volunteer hints about what to talk about as part of the
dialogue.
In other words, if the player is truly getting stuck about what to ask, the
narrative is failing. If the player is getting stuck about how to ask it,
the game mechanics are failing. The trick is to make the latter part of the
former. So once the player knows how to elicit information (or how to give
information, in turn eliciting more information), then the rest can flow
from the narrative. But, like you say, this is easier said than done. Part
of what people often wanted was to do things like this:
Discuss 'destiny' with Morningstar.
Keep talking about 'fate' with Morningstar. (Or even just: Keep talking
about 'fate'.)
But, other than allowing things like that, people were pretty happy with
"ASK ABOUT {topic}" or "TELL ABOUT {topic}". What mattered much more was how
the NPC responded, both when the topic was something that could be discussed
and when it was a dead-end and how much those responses seemed to keep the
story going.
> The idea here is that the main conversation menu is the only thing you
> need to complete the game, but that there is a lot of (optional)
> interesting and useful information 'hidden' behind the ASK/TELL system.
Yeah, this issue of "other knowledge" or "other story information" came up a
lot. Basically: what if I want the player to be able to experience more
detail if they work to ferret it out? Fine, but then if I allow the
narrative to make all that clear, doesn't that stop the narrative in just as
intrusive a way as a menu or a suggested topics?
This is really a problem that writers of novels run into as well. How much
information to present and when. (A master of this, I think, is Alastair
Reynolds but that's just me. I also think Stephen King does a fairly good
job.) So the trick here was seen as being able to allow this information to
come up almost no matter what, at least in some context, as part of the
narrative, but allowing the player to pursue those elements they wanted to.
The parts the player *needed* to pursue, however, to continue the narrative
had to be juggled in with that.
This is an area where I think knowledge of writing fiction can inform (no
pun intended) some different styles of textual IF. In other words, the use
of plotting, narrative structure, and pacing can at least provide ideas on
how to perhaps better convey story information and better structure a story
world with gating criteria that allow you to give the player a chance to
follow up on finding out things that they wanted to before, but didn't get a
chance.
In our group, this often came up as checking if certain things were
discussed to or by the player character. If they were, fine. Done. But if
not, contextual clues at certain parts of the narrative (game) -- where the
pacing would allow it -- would bring up the idea again, giving the player
another chance to get more information.
- Jeff
Jim Aikin
>
> More practically, though, what am I going to do in my next game? Things
> would be easy if there were some other conversation system that I really
> liked, but there isn't. When I play a game that uses a standard ASK/TELL
> system, I always have two problems: (1) I'm typing in a lot of things
> that the NPC cannot converse about, which leads to irritation and (I
> would argue) a more jarring sense of disconnection with the game than
> the appearance of a conversation menu, and (2) I'm saying things I do
> not want to say, because I can only choose topics, but I cannot choose
> _what_ to say about that topic.
I'm not a fan of menu-based conversation, so perhaps my bias will be
embedded in this answer.
With reference to item (1), the solution is, unfortunately, that the
author needs to do a lot of extra work to code unlikely or incidental
topics of conversation. That, and useful defaults for not-covered
topics. The TADS shuffled event list is a good resource for making the
defaults a _little_ less annoying, though the results still tend to be
rather artificial.
With reference to item (2), I just now had an idea based on something in
Jeff's post. If the PC is truly a character (and not simply the avatar
of the player), then it can be enlightening to learn what the PC is
thinking -- because it might _not_ be what the player is thinking. So
perhaps the command 'think about what to say to Mrs. McGillicuddy' would
be a useful general form. (Or simply 'think about what to say', provided
Mrs. McGillicuddy is in the room.)
In response, the game could do several things. It could sort possible
topics so as to highlight items that would be of urgent concern to the
PC (and omit items that the PC has no knowledge of). It could give a bit
of the PC's thought processes about such topics, which would deepen the
PC's character. And it could feed phrases to the player that go beyond
ask/tell about -- something like this: "You could ask the general about
the war (though he's already told you how badly it's going), about his
medals (if you don't mind him droning on for half an hour), about the
new tank brigade, about the spies, or how he plans to defeat the wily
Hun." [Or perhaps Wily the Hun ... there's probably a comic book
character in that.] This would alert the player that 'ask the general
how he plans to defeat the wily Hun" would currently be allowed as an input.
--JA
Bert Byfield
> then this is bad idea. Also, a hybrid system may be confusing; or
> it might manage to bring the _bad_ aspects of its components
> together.
> So I'm a bit stumped, and I would very much like to talk to you
> about this problem. Kind regards, Victor
You can't please everyone. I recommend you only count the people who
*do* like something, expecting any game to displease some amount of the
audience, but not being nonplussed by that.
thatwritterguy:D
walls of text x_X remember to space in further posts
Eriorg
> What I'm currently doing is implementing a hybrid ASK/TELL-Menu system.
> The player can bring up topics, to which the character either responds
> directly, or which bring up a conversation menu with further options.
> The player can also use a general "talk to" command to start a
> conversation menu about a predefined (by the author) topic, so that they
> won't get stuck in trying to think of the right topic.
>
> The idea here is that the main conversation menu is the only thing you
> need to complete the game, but that there is a lot of (optional)
> interesting and useful information 'hidden' behind the ASK/TELL system.
>
> I am not sure that this is a good idea.
>
> If the very appearance of a conversation menu is something bad, then
> this is bad idea. Also, a hybrid system may be confusing; or it might
> manage to bring the _bad_ aspects of its components together.
I'm not an IF author, but as a player, I'd like to mention that the
most satisfying conversation system I've seen in an IF game was the
hybrid ASK/TELL-Menu system in "City of Secrets". I thought it was the
best of both worlds, with a great feeling of freedom and richness:
like games with menus, it had flowing conversations instead of
successions of unrelated questions and their answers; but unlike games
with only menus, you could also often find unmentioned topics when you
wanted, which made the conversation feel less railroaded. Also,
although it's less important: in this game, the menus were fixed in
their own window at the bottom of the screen, so they didn't have to
be re-printed at every turn in the main window; I think it looks nicer
that way.
Of course, such a system, with so many different possibilities to
manage and so many different answers to write, must be an awful lot of
work for the author! And even with all that work, the conversations of
"City of Secrets" weren't always totally bug-free. Perhaps it'd be
easier to have several authors working full-time, one for each NPC?...
The TADS 3 conversation system, used in "The Elysium Enigma" and
others, is actually rather similar to the system from "City of
Secrets". I like it very much too, but not quite as much, because the
lists of suggestions are basically menus, but without the precision of
complete sentences, and it's just longer to type. (Also, in the TADS 3
system, I really don't like it when I have to type exactly the same
thing several times if I want to know everything the NPC can tell
about that topic.)
Mike Rozak
> As I briefly alluded to earlier, menu-based conversations were seen as
> horribly distracting and completely breaking of the story. One of my
> authors referred to these as "gaping potholes in the narrative flow." (One
> recent game we looked at was "Fate" and while people really enjoyed the
> story, they didn't like the menu-based conversations. Although, that said,
> they did recognize the challenge in trying to present how to communicate
> with NPCs.)
There are two competing UI issues at play:
1) Without a menu: Most of the things a player wants to say won't be
understood. When they try these and get a "I'm stupid. I don't understand"
response, they get frustrated.
2) With a menu: The player feels constrained because they have had options
removed from them. It's like saying, "You can choose any color so long as
it's black."
I found that menus are necessary, BUT I've created a "soft" menu, for both
commands and conversations.
Basically, you can type anything you want for commands and conversations. As
per usual in IF, sometimes it'll be understood, but more esoteric
commands/conversations won't.
There is an object context menu though, which is displays the most obvious
or commonly-used actions, such as "get", "drop", or "Hello."
Within the context menu, there are two options: "What else can I say?" and
"What else can I do?" that bring up really long lists of commands, so that
player knows what's possible and what isn't.
You can try at www.CircumReality.com .
Aaron A. Reed
Numbered menus immediately take me back to 1986 and BASIC. They seem
vaguely embarrassing and reek of primitive, unsophisticated code,
regardless of what actually is going on underneath the surface. Maybe
it's just the numbers that does it; maybe it's more of a set
convention in graphical adventure games. I'm not sure.
Plus, as others have observed, menus strip away the illusion. A much
more sensical way to write IF may well be to give the player a
numbered list of every currently available verb/noun combination the
author has implemented, but almost no successful games have been
written this way.
That said, there is certainly no easier way to offer complex dialog
options to the player. I certainly can't think of any other way to
offer the sort of complex conversational gambits that appear in "The
Baron" (or "Planescape: Torment") without menu options.
As Victor mentions, adding more verbs doesn't work very well either.
My current project once had several conversation verbs like IGNORE and
INTIMIDATE, but in a large game the sheer difficulty of dealing with
this response at every conceivable input prompt, without cop-outs like
"Now isn't a good time to intimidate Mrs. Davenport," led me to cut
them.
The style of saying "You could ask Mrs. Davenport about the soup, her
retirement plan, or recommendations for polka records" is jarring in a
different way. Normally, the narrative voice fades into the background
during play, only coming forward for an error, a disambiguation
message, or to present out-of-game data like a "Save successful"
message. But in this system, it feels like the narrator is constantly
interposing himself between you and your conversational partner.
Then there's the issue of how you teach new players to type "ask mrs
davenport about soup" rather than "Do you like it?"
In short, conversation is tough. When my game comes out, I'll be
interested to see what people think of my take on it.
--Aaron
R. Alan Monroe
news:c7c70559-29ad-44c6...@i7g2000prf.googlegroups.com:
> A much more sensical way to write IF may well be to give the player a
> numbered list of every currently available verb/noun combination the
> author has implemented, but almost no successful games have been
> written this way.
Can you name one (honest question)? I'm very interested in seeing such a
game.
Alan
Jeff Nyman
news:c7c70559-29ad-44c6...@i7g2000prf.googlegroups.com...
> That said, there is certainly no easier way to offer complex dialog
> options to the player. I certainly can't think of any other way to
> offer the sort of complex conversational gambits that appear in "The
> Baron" (or "Planescape: Torment") without menu options.
I haven't looked at "The Baron" yet but I'm planning to. I'm wondering if it
would be beneficial to try to craft those conversational gambits in another
way; see what can be done. Or maybe since the source to "Fate" is out there,
it could be tried with that game. In other words, trying other ways to see
how the conversational elements can happen. Victor mentioned a rationale to
be provided as part of "The Baron" (for killing a wolf) and it would be
interesting to see if that, too, could be incorporated other than as a stark
response choice.
> The style of saying "You could ask Mrs. Davenport about the soup, her
> retirement plan, or recommendations for polka records" is jarring in a
> different way. Normally, the narrative voice fades into the background
> during play, only coming forward for an error, a disambiguation
> message, or to present out-of-game data like a "Save successful"
> message.
From my point of view, I don't think of those messages (errors,
disambiguation) as the "narrative voice", per se. The voice of narrative is
part of the writing style and often doesn't relate to what's actually said
but rather how it's said, thus providing a sort of tone to the work as a
whole. It comes down to the persona telling the story. That persona should
develop from the personality and attitude of the narrator, all of which is
expressed by the author's choice of words and incidents. The game engine
isn't choosing the words and incidents.
So I don't liken it to the game being the persona telling the story
(although I can easily see how with slim narrative, that can seem to be the
case). For me, the messages you are referring to are somewhat like
"authorial intrusion" except it's the game engine intruding on the author's
behalf. By that same logic, though, I think this is part of why menus are
seen as jarring -- because they are *not* part of the narrative voice. They
are a straddle between the narrative and the game mechanics.
- Jeff
Kevin
conversations. When a menu-based conversation drops in, the game is
essentially saying, "Warning! Warning! You are at a crossroads!"
Pretty clearly, your choices will make some difference in the
development of the game. Therefore, my first reaction is always to
save the game and then mechanically go about trying each option. The
far more satisfying progress is to encounter an NPC and be at that
same crossroads, but without the warning signs. For example, perhaps
the NPC still has a hidden bank of pat responses, but rather than
triggering them by something so obvious as typing 1, 2, or 3, they are
triggered by whether you are holding the cigar, the notepad, or the
brick. Best yet are those situations in which the cause-and-effect
relationship of the triggering mechanism is not immediately obvious
even though clues about the cause-and-effect relationship might be
hinted at through careful reading and observation.
Jeff Nyman
news:8c8f7980-bce5-41e2...@59g2000hsb.googlegroups.com...
> I'd like to suggest a less artistic objection to menu-based
> conversations. When a menu-based conversation drops in, the game is
> essentially saying, "Warning! Warning! You are at a crossroads!"
> Pretty clearly, your choices will make some difference in the
> development of the game.
Well, on the last part ("make some difference"), I'll come back to that. The
other part ("crossroads"): very true. In fact, in my "Teaching Storytelling"
thread I mentioned this:
"The problem is that people found they 'left the narrative' when
encountering a menu because they had to stop and think what
the alternative choices might get them and they didn't know if
they could go back."
So, as I came to understand it, the two aspects were sort of combined:
people were taken out of the narrative not solely because of the *mechanics*
of a menu but rather of what the menu represented: a possible one-way choice
without clear indication of the consequences or if a plot element or parts
of the story would be missing otherwise.
As far as the "make some difference", that's often questionable. Sometimes
I've found the choice you make does have a substantive difference; other
times it just nets you a bit of dialogue or description that you wouldn't
otherwise have seen, which is questionable in terms of how substantive it
truly is. So I guess what I'm saying is that it wasn't "pretty clearly" the
case that your choice would actually matter. It often wasn't known one way
or the other. Even if there were hints of the consequences of one choice
over another, what I found was that people didn't often respond well to this
because it forced them into "I have to play again and try those choices."
Now that was interesting to me because if you think of novels, often people
re-read them to gain more the next time around -- even though you know
nothing has changed. But it's the re-reading in light your experience with
how the story played out -- and the eventual end -- that allows you to
sometimes pick up on or appreciate elements that at first passed you by.
This was an aspect that just about everyone in my groups felt was missing
from textual IF, a notable exception (and there were others) being "Babel"
which people found had replay (re-read) value.
So here the "make some difference" was not relegated to choices you could
make, or alternative solutions you could try, but rather to gaining a better
understanding of the story or the characters in light of your previous
playing/writing experience. A large topic of discussion was centered around
how to make game-stories rich enough to warrant a second play/read but
without using "artificial mechanisms" (like menus or emergent solutions) to
pull this off.
A side note to this was the thinking often stated to me that while no doubt
a lot of work goes into textual IF, the barrier to entry is relatively low,
compared to writing your own graphic game or writing your own novel. To this
extent, it was felt that textual IF would probably appeal to a certain
writer (whether or game or static fiction) who finally found a game format
they could write in that didn't require them to actually do a lot of the
stuff (graphics, sound, video) that other game formats require.
But the corollary to this was that people felt this tended to bring in
people who really didn't think about the writing, just about the game
aspects. Hence, you end up with things like menus. (I hate to just pick on
menus like this, but it just happens to be our focal point here.)
- Jeff
Blank
But in Planetscape: torment the *whole screen* didn't blank out and turn
into a menu - the conversation choices were in a tiny window with the
rest of the game still 'there' (i.e. visible) behind it, so there wasn't
that sudden disconnection (which is what I find disturbing about menus
as implemented in Inform).
Perhaps part of the issue with IF is that the "portal" onto the game
world is comparatively small (or perhaps "zoomed in") - often after
examining something the only text the player can see is the description
of that item. The rest of the room/activity has scrolled out of sight.
Perhaps part of the reason the format
(you could ask Deborah about the marigolds, the Cezanne, or the policeman)
>
has been better received is because it's more parsimonious with screen
real-estate than a traditional menu?
I wonder if a split-screen game (I'm thinking of the Douglas Adams
style, with the room description constantly visible in the upper pane)
would produce different responses?
One of the things I'm hoping for with I7 is that it will make it easier
to produce multi-window games without having to rummage elbow deep in glulx.
To be fair, I think in that situation you could normally just type "war"
or "tea".
>
> One possible solution to Problem 2, which Emily Short recently brought
> up, is to include extra commands for the player so that they can
> "recommend X", say, instead of "talk about X". There are many related
> ways of doing this: we could let the player choose a mood,
> Varicella-style; we could let the player "agree" and "disagree" whenever
> the NPC made a claim; and so on.
>
Unfortunately adding conversational verbs to the game just makes the
topics themselves harder to hit. It's bad enough for the player having
to discover that they need to ask Mr Ferguson for his aunt's marmalade
recipe in order to progress in the game. If they have to ask him for it
reassuringly (rather than demanding or wheedling), fat chance.
I think the reason that Torment worked so well is because they
understood that what makes good menu choices is presenting the player
with a moral dilemma, so that there is no "correct" answer, just
weighing different costs. The constriction of action at that point feels
quite natural to me.
Contrariwise, CYOA games I've played which said things like "the passage
forks here. Do you (1)go left, (2)go right" demonstrate how /not/ to use
menus, because it emphasises what should be a trivial exploration choice
too much.
> The problem with this approach seems to be twofold. (1) Every one of
> these additions will only be useful in a limited number of situations,
> and will leave other points in the conversation where it would be
> interesting to be able to choose between two things to say unresolved.
> (2) Once implemented, you'd have to make sure that these commands work
> correctly for _every_ point in _every_ conversation. That's... not good.
>
>
> What I'm currently doing is implementing a hybrid ASK/TELL-Menu system.
> The player can bring up topics, to which the character either responds
> directly, or which bring up a conversation menu with further options.
> The player can also use a general "talk to" command to start a
> conversation menu about a predefined (by the author) topic, so that they
> won't get stuck in trying to think of the right topic.
>
> The idea here is that the main conversation menu is the only thing you
> need to complete the game, but that there is a lot of (optional)
> interesting and useful information 'hidden' behind the ASK/TELL system.
>
> I am not sure that this is a good idea.
>
How about if the game tracked the essential information the player needs
to discover, which is available from very early on in the game (though
perhaps in tricksy side conversations) and if the player has reached the
point where they're stuck, an NPC tracks them down and *tells* them?
--jz
Blank
> Victor Gijsbers wrote:
>>
>> More practically, though, what am I going to do in my next game?
>> Things would be easy if there were some other conversation system that
>> I really liked, but there isn't. When I play a game that uses a
>> standard ASK/TELL system, I always have two problems: (1) I'm typing
>> in a lot of things that the NPC cannot converse about, which leads to
>> irritation and (I would argue) a more jarring sense of disconnection
>> with the game than the appearance of a conversation menu, and (2) I'm
>> saying things I do not want to say, because I can only choose topics,
>> but I cannot choose _what_ to say about that topic.
>
> I'm not a fan of menu-based conversation, so perhaps my bias will be
> embedded in this answer.
>
> With reference to item (1), the solution is, unfortunately, that the
> author needs to do a lot of extra work to code unlikely or incidental
> topics of conversation. That, and useful defaults for not-covered
> topics. The TADS shuffled event list is a good resource for making the
> defaults a _little_ less annoying, though the results still tend to be
> rather artificial.
>
but that artifice is also a useful signal to the player: "not
implemented: not a puzzle." If you make the responses too natural, the
player can batter away thinking that they just haven't asked the
question or set the situation up quite right. It's just like
implementing soft boundaries to the map: we want to stay in keeping with
the game while still tipping the player the wink so they don't get
frustrated.
> With reference to item (2), I just now had an idea based on something in
> Jeff's post. If the PC is truly a character (and not simply the avatar
> of the player), then it can be enlightening to learn what the PC is
> thinking -- because it might _not_ be what the player is thinking.
What about two screens: one showing the "objective reality" and the
other showing the PC's internal monologue? Just a random thought.
--jz
Jim Aikin
>
> I wonder if a split-screen game (I'm thinking of the Douglas Adams
> style, with the room description constantly visible in the upper pane)
> would produce different responses?
Without having played any split-screen games, I would be a little leery
of that option. I'd expect it would distract the player by forcing
his/her attention into two separate boxes at once -- like having a
conversation while listening to music. Hard work for the brain.
> How about if the game tracked the essential information the player needs
> to discover, which is available from very early on in the game (though
> perhaps in tricksy side conversations) and if the player has reached the
> point where they're stuck, an NPC tracks them down and *tells* them?
This is a fascinating idea. It would be a bit complex to implement, from
two directions. First, the software would need to track what the player
has already encountered (or has not yet encountered but needs to know
now). Second, the author would have to create a convincing NPC who is
able to roam the map and has basically good intentions toward the PC.
In some types of stories the latter might not be too difficult:
Inspector Chumley wanders into the room. "You know, I've been thinking
about those peculiar footprints," he says. "I may have an idea about
them. Could be rubbish, of course, my ideas are often rubbish. But if
you're at all curious, you need only ask me about the footprints.
The last sentence could be omitted, of course, if the player has chosen
"no, I'm not a beginner" at the start of the game.
--JA
Jeff Nyman
> How about if the game tracked the essential information the player needs
> to discover, which is available from very early on in the game (though
> perhaps in tricksy side conversations) and if the player has reached the
> point where they're stuck, an NPC tracks them down and *tells* them?
This is sort of going to what I mentioned earlier:
"In our group, this often came up as checking if certain things were
discussed to or by the player character. If they were, fine. Done. But if
not, contextual clues at certain parts of the narrative (game) -- where the
pacing would allow it -- would bring up the idea again, giving the player
another chance to get more information."
The "bring up the idea again" was based on coming up with mechanisms that
would allow the player chances to get the information needed.
The issue is often that most such conversations are required as gating items
(thus serving as a form of puzzle) that then allows the narrative to
continue when that information has been acquired. That's a convention that
has existed for a long time.
The investigation we were doing started centering on what you could, as part
of the narrative, to give the player chances to get the information they
need and then providing more overt aspects of the narrative if the player
clearly wasn't getting it or wasn't thinking of what to do to ask for it.
NPCs don't have to be the sole usage for this. Perhaps there are points
where let's say the player hasn't gotten a clue about some specific thing.
If the game recognizes that the player is getting near to where that
matters, perhaps something happens -- like you say, maybe an NPC tracks them
down; or perhaps when the player enters a room, a "memory" of what they saw
in a previous room is triggered by the game.
This, by the way, is where I think relations can be REALLY powerful in
Inform 7. The use of a mechanism that relates things together not just by
their physical or conceptual nature in the game, but by their narrative
nature as part of the story. I think this is a wide avenue of exploration
for textual IF and with Inform 7 in particular.
- Jeff
Marc Trius
> to discover, which is available from very early on in the game (though
> perhaps in tricksy side conversations) and if the player has reached the
> point where they're stuck, an NPC tracks them down and *tells* them?
> What about two screens: one showing the "objective reality" and the
> other showing the PC's internal monologue? Just a random thought.
>
> --jz
hear, hear
Blank
> Blank wrote:
>>
>> I wonder if a split-screen game (I'm thinking of the Douglas Adams
>> style, with the room description constantly visible in the upper pane)
>> would produce different responses?
>
> Without having played any split-screen games, I would be a little leery
> of that option. I'd expect it would distract the player by forcing
> his/her attention into two separate boxes at once -- like having a
> conversation while listening to music. Hard work for the brain.
>
I don't think that's an accurate analogy, because you're talking about
splitting the player's attention. Rather than intending the player to
pay attention to both panels simultaneously, I was thinking of the two
panels as "foreground" and "background".
I find when I'm playing IF I'm constantly typing "look" anyway to remind
me what else is there, where the exits are, yadda yadda. In the case of
the menu appearing in the bottom pane, the top pane would still show the
description of the location, which would act to keep the player
"connected" to the scene & the game (or maybe not - that's why I want to
try it out!)
>> How about if the game tracked the essential information the player
>> needs to discover, which is available from very early on in the game
>> (though perhaps in tricksy side conversations) and if the player has
>> reached the point where they're stuck, an NPC tracks them down and
>> *tells* them?
>
> This is a fascinating idea. It would be a bit complex to implement, from
> two directions. First, the software would need to track what the player
> has already encountered (or has not yet encountered but needs to know
> now). Second, the author would have to create a convincing NPC who is
> able to roam the map and has basically good intentions toward the PC.
>
Well, the npc doesn't actually need to know they're helping -
Every turn:
if the bees are known, and the violent argument has happened and the
interior of the clock has not been examined begin;
move mrs jensen to the location;
say "Mrs. Jensen staggers in with her hands over her ears. 'Oh, my
heavens!' she moans. 'The clock's gone crazy! It's striking four o'clock
over and over!'";
the crazy clock begins in one turn from now;
award no points for discovering the oily gloves; [because I'm mean]
end if;
or
The door crashed open and Jack strides in, his face set in a murderous
scowl. "Angelica says you've been nosing around asking questions. I
don't care what you've heard - I'm clean. Fuck up my career and I'll
fuck you up, got that?"
and then even if Jack doesn't actually give him the information in that
scene the player could now go to the sports center or Angelica with a
pretty good idea of what to ask about.
--jz
Jim Aikin
>
> Well, the npc doesn't actually need to know they're helping -
True enough.
> Every turn:
> if the bees are known, and the violent argument has happened and the
> interior of the clock has not been examined begin;
> move mrs jensen to the location;
> say "Mrs. Jensen staggers in with her hands over her ears. 'Oh, my
> heavens!' she moans. 'The clock's gone crazy! It's striking four o'clock
> over and over!'";
> the crazy clock begins in one turn from now;
> award no points for discovering the oily gloves; [because I'm mean]
> end if;
As written, that would have Mrs. Jensen entering the room every turn. I
know you're only giving an example here, not trying to write a working
game, but there are lots of fiddly little problems like that, all of
which would have to be handled gracefully.
Like, how many turns do you let the player flounder around before Mrs.
Jensen arrives?
And what is Mrs. Jensen doing in the meantime? In other words, what
would her function be in the story if the player was doing everything
right, and didn't need reminding?
--JA
Emily Short
> > With reference to item (2), I just now had an idea based on something in
> > Jeff's post. If the PC is truly a character (and not simply the avatar
> > of the player), then it can be enlightening to learn what the PC is
> > thinking -- because it might _not_ be what the player is thinking.
>
> What about two screens: one showing the "objective reality" and the
> other showing the PC's internal monologue? Just a random thought.
Have you tried Narcolepsy? It does something quite similar and might
be worth trying if you're interested in that idea.
Emily Short
> One of the things I'm hoping for with I7 is that it will make it easier
> to produce multi-window games without having to rummage elbow deep in glulx.
Should be already possible: try Jon Ingold's Flexible Windows, and
compare his game Dead Cities, which uses it. (If it needs more
refinement, that'd be worth knowing about, but from my experiments
with the extension it looks as though this functionality is already
all there.)
> How about if the game tracked the essential information the player needs
> to discover, which is available from very early on in the game (though
> perhaps in tricksy side conversations) and if the player has reached the
> point where they're stuck, an NPC tracks them down and *tells* them?
I tried something along these lines with City of Secrets (well,
broadly along these lines): there are a large number of scenes, some
of them very short, some more complicated, that exist purely to nudge
the player along if he seems to have gone too long without making
progress or finding out some important fact. This is where the idea
for the I7 scene object comes from initially, in fact. The triggers
for these are things like "if N turns have elapsed since the player
last did something story-advancing and he doesn't seem to have found
out about Y and he's in a shop, ...", and then they'd send in another
customer to have a conversation that he can overhear; or "the player
hasn't realized X and he's wandering around this region of the
map...", and they'll produce an encounter with a guard. That kind of
thing.
This did not work entirely: people routinely get stuck in CoS, and one
of the things that makes it hard to give them hints is precisely that
the game can get into many different states. As far as design goes, I
think there are several potential pitfalls.
One is that if it happens too much -- that is, if the player runs into
too many triggered scenes rather than acquiring information himself --
then he starts to feel as though the game is not under his control.
One of the challenges about CoS, and the reason that the midgame
pacing doesn't work, is that there's a huge area (in both geography
and conversation-space) to explore freely. It is easy for the player
to lose focus on what he's supposed to be doing or where he still
hasn't done enough exploring; and there the scenes might actually have
a detrimental effect by distracting the player from an idea he might
have been in the middle of pursuing, but which the game logic didn't
pick up on (so it decided that he was "stuck").
The second pitfall is that, as with any sort of soft/intuitive kind of
gating in IF, it takes a lot more testing (ideally with a lot of
different beta-testers) to make sure that everything works properly
and that you've chosen sensible numbers for e.g. the number of turns
at which you decide to provide each nudge. The more sophisticated your
criteria for detecting "stuckness", the more tuning they'll need. I
went through many, many testers on that project, and still ultimately
hadn't given it a thorough-enough testing out. It might have been
plausible to do more if I'd had a commercial QA department at my
disposal, but in terms of what an individual author will have time and
energy to do with the donated testing of the IF community, I felt like
I was pushing the boundaries. Certainly by the end of the project I
had put some months of more-than-full-time work into it and had
completely exhausted my own personal resources.
Now, I think all this might be a bit easier to arrange if the author
focuses the project better; CoS is also marred by trying to do
everything that I thought constituted "good IF", and as a result winds
up doing almost none of it well. If there were a structure with more
tightly defined goals, or if there were better abstract
representations of the player's degree of progress internal to the
game, it might be much easier to design this all to work well.
Victor Gijsbers
> I've found that a lot of what we dealt with, pretty much across the board,
> had to do with telling stories differently by *not* conflating the player
> and the player character, which then opened up more narrative possibilities
> and thus more ways to at least think about handling things like this, such
> as giving reasons for action. I'll definitely check out "The Baron."
I am curious about this. It seems to me that much of the interest of
interactive narrative comes through a partial identification of the
player with the player character, to be precise, that identification
that makes the player feel responsible for the actions of the player
character. If the player doesn't feel responsible for what happens in
the fiction, I doubt that IF can invoke emotions that static fiction cannot.
So how does _not_ conflating the player and the player character open up
more narrative possibilities?
> That first point did come up a lot, where an NPC just says "I don't know
> about that." or, worse, the game itself just says "Bob doesn't appear
> inclined to talk about that." The issue here was whether or not it was good
> to have NPCs that could have some sort of "aggravation" factor. When the
> factor was exceeded, they started responding in slightly different ways than
> they would otherwise, eventually clamming up. This is, again, a simplistic
> take on what was in fact weeks of discussion. But the idea was modeling NPCs
> with things like an "attention span", a "mood", and so forth. Then making it
> clear to the player in a graduated way that further asking along these lines
> might be not only useless but maybe even counter-productive.
>
> For example: "I didn't know Jack well, but from what I knew I was pretty
> sure if I kept this up, he was going to stop talking to me at all. I
> couldn't afford that."
Yes, but this would only feel natural if I were pestering Jack about
what really happened to his wife all those years ago, not when I was
trying to ask Jack about a couple of random topics that the game author
did not happen to think about--wouldn't it?
> I'm not sure about (2) in the sense that I don't know what you mean.
Suppose I type "TELL ALEX ABOUT FRIENDSHIP".
What on earth am I going to say if I type that? Is it:
1. "So Alex, this is how much you value our many years of friendship?
You betray me as soon as things get difficult, and even team up on me
with what you believed to be the winning side? It makes me sick that I
ever called you my friend--a random whore on the streets would have been
more faithful to me. And now you die."
or is it
2. "You know, Alex, if anybody else had done this to me, he would be
dead. He would be dead right now. Bullet in the head, no mercy. But I
won't kill you, Alex, because, you known, I'm not like you--I don't
abandon my friends. I stick with them, whatever happens, even if they
betray me. So I won't kill you. Now get out, before I change my mind."
That makes a lot of difference, obviously, and the ASK/TELL system just
doesn't seem to be able to present such choices very well.
Or does it?
Regards,
Victor
Victor Gijsbers
> Victor Gijsbers wrote:
>>
>> More practically, though, what am I going to do in my next game?
>> Things would be easy if there were some other conversation system that
>> I really liked, but there isn't. When I play a game that uses a
>> standard ASK/TELL system, I always have two problems: (1) I'm typing
>> in a lot of things that the NPC cannot converse about, which leads to
>> irritation and (I would argue) a more jarring sense of disconnection
>> with the game than the appearance of a conversation menu, and (2) I'm
>> saying things I do not want to say, because I can only choose topics,
>> but I cannot choose _what_ to say about that topic.
>
> I'm not a fan of menu-based conversation, so perhaps my bias will be
> embedded in this answer.
Could you elaborate on that a bit? Why do you dislike it?
> With reference to item (1), the solution is, unfortunately, that the
> author needs to do a lot of extra work to code unlikely or incidental
> topics of conversation.
Yes, but I think there are two reasons why this isn't really feasible.
1. As a game becomes larger, both the number of conceivable topics and
the number of NPCs increases--but since the two have to multiplied in
order to get the number of responses you have to write, things get bad
very quickly.
2. The deeper your implementation, the bigger the expectations that the
player will form about which topics she can converse about. If I don't
implement conversation topics for all the nouns in the room the NPC is
in, the player will notice this after a few abortive attempts at
conversation, and will stop asking about nouns. But if I do, the player
will start expecting that she can even talk about nouns in other
rooms--and so on.
It seems like a situations you cannot win--the more ground you cover,
the more ground people will try to tread.
> With reference to item (2), I just now had an idea based on something in
> Jeff's post. If the PC is truly a character (and not simply the avatar
> of the player), then it can be enlightening to learn what the PC is
> thinking -- because it might _not_ be what the player is thinking. So
> perhaps the command 'think about what to say to Mrs. McGillicuddy' would
> be a useful general form. (Or simply 'think about what to say', provided
> Mrs. McGillicuddy is in the room.)
I wonder--wouldn't players start doing this every time they saw an NPC,
and a few more times just for good measure? Wouldn't this, in the end,
come down to something very much like TADS-style in-conversation hints,
only less useful because not always visible?
I'm not sure whether it would, but I suspect it might. What do you think?
Regards,
Victor
Jeff Nyman
> I am curious about this. It seems to me that much of the interest of
> interactive narrative comes through a partial identification of the player
> with the player character, to be precise, that identification that makes
> the player feel responsible for the actions of the player character. If
> the player doesn't feel responsible for what happens in the fiction, I
> doubt that IF can invoke emotions that static fiction cannot.
From what I've seen, including doing a lot of testing in game-related
companies, this "partial identification" isn't really sought all that much,
but the idea that it is seems to be very much a part of textual IF. (This is
originally why I started the "Second Person" thread -- many, many months
ago -- wondering if the second person is the reason for this, or just an
outgrowth of it.) Like I had mentioned about "The Longest Journey", this was
a very popular graphical adventure game and it was a highly story-driven
one. The enjoyment that came from this was not due to partial identification
with the protagonist whom you played, but rather with the way the story was
told, including background information into April Ryan's life. To many
people, that game did evoke emotions, at least at some level.
Likewise with novels. I don't think you're too often expected to have any
sort of identification with the protagonist except insofar as you can relate
to (sympathize, empathize, despise, etc) the thoughts and/or actions taken
by the protagonist as they solve their story problems. Many novels evoke
emotions in readers.
So there are two forms of storytelling -- graphical adventure games and
novels -- that can evoke emotions but that do not require the reader/player
feeling responsible in some sense.
If I want my emotions engaged, conflating myself with a protagonist can only
hurt because then I'm just looking at my own view of the world and my own
reactions, and I already know that view and those reactions. What I want to
do is be forced to consider alternative viewpoints and alternative actions
that might be taken.
This is where I think textual IF can benefit: by looking at how storytellers
in general evoke emotion, whether that be film or novel, and yet still don't
require a conflation of the viewer/reader with the actual characters on the
screen/page.
> So how does _not_ conflating the player and the player character open up
> more narrative possibilities?
One way is because then the player character (the protagonist) is definitely
*not* you. If I read every novel that had a character like me (or a
cardboard character that I could assume was like me), the narrative
possibilities are quite a bit more limited from the author's standpoint.
For me when you write a story, you're saying "Okay, reader, here you are.
Given these characters and this situation, human nature is such that things
will end up this way." This is your *truth* as an author. You are asking the
reader to accept that truth -- but textual IF gives the reader a chance to
subvert that a bit and try to impose their own ideas over the author's. How
much will the story/game let them? Well, clearly the game will constrain
what can be done to allow the narrative to go on. But the extent that the
player is forced to think about how *they* would react in that situation but
then what they have to do to get the *player character* to react in that
situation sets up a dynamic. (At least in my view.)
When depicting the life of a fictional character, a novelist usually has to
choose to include only those impressions, thoughts, reflections, sensations,
feelings, desires, and so on, that bear on the character's motivations,
development, and decision-making abilities. In other words, those aspects of
character that will affect the way in which the character copes with the
dilemmas they're going to face in the course of the story. But that implies
there are impressions, thoughts, reflections, sensations, and so forth. I
don't just want my impressions or thoughts; I can get to those without
playing a game. What I want is to investigate a situation that allows me to
play as a different person: and then see how I might have to react given
that I am this different person. (Who knows? Maybe I'll find out I can
conflate myself with the protagonist -- but maybe not.) The point is I have
more narrative possibilities when I can open up a character that is very
different from the reader.
Likewise, so-called "rounded characters" are said to have complex motives
and conflicting desires and are subject to passions and ambitions. Those may
be motives I disagree with; desires I don't have; passions and ambitions I
don't share or never even thought about. In static fiction, readers enjoy
the "intimacy" with such characters because they're seen as worth knowing.
They're worth spending time with and living vicariously through their
experiences because they force me to look at myself through a lens that is
very much not myself.
Another thing we talked about a lot is that unless a novelist understands
the dynamics of a given character's development, the character's motivations
are usually not fully understood. It's the characters' motivations that
produce the conflicts and generate the narrative tension that a novel must
have if it's going to succeed in holding the reader's attention. Narrative
possibilities open up to me when I can present to hte reader motivations
that force them to consider what they would really do in that situation.
What if the situation is a slowly dawning realization that they have to do
something they really don't like?
As far as narrative possibilities, one major point here is the narrative
voice. This is given by the persona of the person telling the story. If this
is the protagonist, you can tell the story in a way that doesn't ask you to
*be* the character, but asks you to try to *understand* the character and
then act as such a person might. (Not as *you* might; as *such a character*
might. Then when the game is over, you can see if you agree that such a
character might act that way.)
>> For example: "I didn't know Jack well, but from what I knew I was pretty
>> sure if I kept this up, he was going to stop talking to me at all. I
>> couldn't afford that."
>
> Yes, but this would only feel natural if I were pestering Jack about what
> really happened to his wife all those years ago, not when I was trying to
> ask Jack about a couple of random topics that the game author did not
> happen to think about--wouldn't it?
Yes, and in that case, the above response of mine probably wouldn't be
issued. If it was random topics, you could perhaps contextualize a bit on
standard non-replies, as it were. Like I said, I didn't find much resistance
to people when they saw a clear "game"-imposed boundary on how much an NPC
would respond to numerous topics. Everyone understands it's just an
interactive element that is not a real person. They don't expect this NPC to
act like a real person (just like readers of novels don't expect dialogue
that matches how people "really" talk).
What was found to be more unforgivable is when the NPC didn't contextualize
based on obvious elements key to the story *or* key to character
development: helping the reader/player better understand characters and
their motivations, including that of the player character. For example, in
the above case, if it was important that Jack explain what happen to his
wife all those years ago, it would be disappointing if questions along those
lines just had Jack always refusing to answer, without some context to his
action. Context might be: "Jack stormed off. Apparently I had pushed that
topic too far." If it wasn't important about what happened to Jack's wife
years ago, then it probably should have never been brought up in narrative
in the first place, unless you are using it as a red herring. (But even then
the red herring does play a key point in the narrative.)
Now, a more telling idea is: what if the player character lost *his* wife
years ago. How does *that* impact his motivations as he presents his
impressions of the world to *you*, the player? How does that change how the
*player character* goes about characterizing Jack to *you* the player? When
you, as the player, are having to figure out the protagonist/player
character as much as you are other elements of the game, I do think this is
another area where narrative possibilities can open up.
>> I'm not sure about (2) in the sense that I don't know what you mean.
>
> Suppose I type "TELL ALEX ABOUT FRIENDSHIP".
>
> What on earth am I going to say if I type that? Is it:
Or if the game allowed you to be a bit more specific for a situation. What
about >REMIND ALEX OF OUR FRIENDSHIP.
That at least provides the possibility of a situationally useful response.
My group often wanted to implement things like this, rather than just
relying on ASK/TELL. The trick here was making sure the player knew such
possibilities existed. Now you could present that in a menu, of course, like
this:
(1) Beg for my life.
(2) Remind Alex of our friendship.
To me I can see the benefit of that menu approach, but a lot of people just
found that to be totally breaking of the narrative. My problem was, "Okay,
but then how do I let the player know the possibilities?" For example, do I
have this:
= = = = = =
Alex stood over me, gun in hand. I knew was having a debate as to whether or
not I was a liability. The look in his eye told me which way that debate was
going to go.
"Gonna beg for your life?"
I looked up at him. I guess that was an option. I could grovel and beg. Is
that what I wanted to do? Would it even matter if I reminded him of our
years of friendship; how I stayed with him when his son died. The son he now
blamed me for killing.
> ?????
= = = = = =
It's tricky. Is it "obvious" from that the player can type "beg for my life"
or just "beg"? Is it "obvious" they could say "remind alex of our
friendship" or maybe even "apologize for son's death"? I actually agree that
menus can make that a bit more explicit. But the problem is: what if the
player wants to try all of them (assuming Alex doesn't blast them in the
meantime -- which narrative should indicate is coming)? Does the menu just
keep showing up until the player tries all combinations? If not, what if the
player is curious to have tried another option with Alex but now can't?
- Jeff
Emily Short
<jeffnyman_nospam@_pleasenospam_gmail.com> wrote:
<snipping quite a lot and leaving only a representative portion>
> So there are two forms of storytelling -- graphical adventure games and
> novels -- that can evoke emotions but that do not require the reader/player
> feeling responsible in some sense.
>
> If I want my emotions engaged, conflating myself with a protagonist can only
> hurt because then I'm just looking at my own view of the world and my own
> reactions, and I already know that view and those reactions. What I want to
> do is be forced to consider alternative viewpoints and alternative actions
> that might be taken.
I find this interesting, but I kind of feel (sorry, I hope this
doesn't sound rude) as though the discussion is starting on the wrong
footing, in that it seems largely to be assuming that there is *a*
right/best/most effective/most emotionally gripping possible
relationship between the player and the protagonist, when the history
of modern IF seems to suggest just the opposite. (And it's also
possible to make another split, protagonist/viewpoint character, which
is occasionally meaningful.)
Jeremy Douglass' dissertation goes into this topic quite a bit,
arguing in fact that we consider abolishing the term "player
character" entirely because it's misleading -- it blends together a
host of possible interactor/character relationships in such a way that
it makes it harder to hold clear and comprehensible discussions of
them. While I'm not quite ready to sack the term myself -- because
it's easy shorthand, among other things -- I think his discussion
might be enlightening. He also points out a number of ways existing
games have tweaked the standard in one way or another.
At any rate, I am sympathetic to the argument that IF has not yet
explored the avenues of player vs. character disassociation as far as
it could, or played out every combination of viewpoint/agency, but
there's more out there already than I think people are considering in
this conversation. (I'd offer a list here, but Douglass already talks
about this a good deal, and also I'm running late for dinner...)
Jeff Nyman
news:0f63bde5-9185-4b51...@x41g2000hsb.googlegroups.com...
> I find this interesting, but I kind of feel (sorry, I hope this
> doesn't sound rude) as though the discussion is starting on the wrong
> footing, in that it seems largely to be assuming that there is *a*
> right/best/most effective/most emotionally gripping possible
> relationship between the player and the protagonist, when the history
> of modern IF seems to suggest just the opposite. (And it's also
> possible to make another split, protagonist/viewpoint character, which
> is occasionally meaningful.)
It definitely doesn't sound rude and I can see your point. I think this was
kind of my point, but I'm probably not making it well. I think there are
many alternatives to presenting the story in such a way that does not
conflate the player character with the player and many alternatives that may
do just that.
First, part of my original response to Victor was based on what people told
me and what I observed. The paper thin and/or cardboard nature of many (not
all) IF protagonists was seen much more often than not. (This could have
been due to our selection process of games; more on that below.) Further,
the notion of people writing as if "you" were experiencing the events, often
seemed to constrain how much liberty the author took with giving a
personality to the protagonist that could be discovered along with the other
game elements. Further, a major benefit that many authors saw to textual IF
and its interactivity was to allow the player character (protagonist) to be
in some form of contention with the player, to some extent, opening up
different possibilities.
Second, part of what I'm basing this on is three very successful other areas
of endeavor: films, novels, and graphical adventure games. While there may
be exceptions here and there, in very few cases is there any presumption of
conflation. Further, the contention is often that people don't want to see
movies or read books about themselves -- meaning, about a character that is
just like themself. We read and see films to experience other characters in
dramatic situations (or at least more dramatic than we are used to). This
doesn't mean, however, that we can't relate to these people. That's sort of
the whole point, in fact. Having a character that is not you but that, in
some particulars, you can relate to in some fashion. This gives you license
to feel some responsibility when you agree with the protagonist (who may be
doing bad things) and some license to feel better when you find an
alternative action to take that your protagonist in the game was clearly not
advocating.
One example we dealt with as a practice idea: the protagonist (PC) wanted to
slaughter someone to get through an area. You, as the player, didn't want to
do that. You found an alternate solution. The narrative allows the player
character to realize that course of action was effective and customizes the
player character's response the next time a similar situation comes up. In a
sense, "you" have engendered a (minimal) form of "growth" in a character.
You didn't just read an author who allows his characters to grow; you
effectively work to do this as well; you effectively see what boundaries
your protagonist has and what you can do to operate within those. If you
chose to act just as the protagonist wants or suggests (via narrative),
then, yeah, you have conflated the player character and player. But if not
... that's where people saw possibilities.
Regarding the history of modern IF, that's something I've long been trying
to figure out. I'm not sure yet of the selection effects that have been
present in modern IF that have led to the history as it is. Getting a bunch
of writers involved -- who were not inclined to write for an existing
community -- really had an effect of showing me that a different set of eyes
looked at IF very differently in terms of its potential and possibilities.
Put another way, if I come to IF as a writer but also emerse myself in the
community, I'm likely to automatically adopt some of the conventions of that
community in order to make sure my game is played. What I wanted was a group
that were interested in storytelling, but that were not emersed in the
community (the history of IF to date) and so who would not look at it with
the same filters and/or selection effects being operative.
That's a lot of generalizing there and I'm definitely not trying to claim
that people only write games that will conform to a supposed "community
acceptance" specification nor am I assuming that people in that community
could not break out of it if they so chose. Rather, I just noticed that
people who were not enmeshed, even peripherally in textual IF, had a very
different view of the possibilities they saw and the means by which they saw
those possibilities becoming realities. So I wanted to learn more about
that. I mention all that just to reinforce that I believe I saw what you
describe ("assuming that there is *a* right/best/most effective/most
emotionally gripping possible relationship between the player and the
protagonist") as being in place or at least apparently in place among the IF
community and I was trying to see why others differed.
This being said, your point is not lost on me. That assumption that I
believed existed was most likely partly due to not having enough exposure to
the possibilities (due, again, to the selection process of games; which I'm
getting to below).
> At any rate, I am sympathetic to the argument that IF has not yet
> explored the avenues of player vs. character disassociation as far as
> it could, or played out every combination of viewpoint/agency, but
> there's more out there already than I think people are considering in
> this conversation.
I found the dissertation so I'll read that. I also agree that there are
probably many things out there that showcase some or all of what we're
considering. I can tell you that one problem we had as a group was deciding
what games to try. We were trying to find concise summaries that would allow
us to see what was unique about a given game, or what viewpoint it used (or
whether it switched viewpoints), or what techniques it specifically used
(menu-based conversations, for example), and so forth. We often had to try
to read reviews to find what we were looking for but given the quantity of
material available and the differing detail that reviews provided, it was
sometimes hard to find specific games that we could compare against each
other except by trying them and seeing what they did.
- Jeff
Adam Thornton
>I wonder if a split-screen game (I'm thinking of the Douglas Adams
>style, with the room description constantly visible in the upper pane)
>would produce different responses?
There's an easy way to tell.
Of course, the responders have to play _Stiffy Makane: The Undiscovered
Country_, which might be an awfully high price to pay.
To clarify: the Stiffy Makane conversation menus happen in a different
window than the main game text. Also, I think you mean Scott, rather
than Douglas, Adams.
>One of the things I'm hoping for with I7 is that it will make it easier
>to produce multi-window games without having to rummage elbow deep in glulx.
If there were a GWindows Extension for I7 that would be really cool.
>I think the reason that Torment worked so well is because they
>understood that what makes good menu choices is presenting the player
>with a moral dilemma, so that there is no "correct" answer, just
>weighing different costs. The constriction of action at that point feels
>quite natural to me.
One of my favorite techniques in _Torment_ was to have identical
conversation choices, one of which was followed by "(lie)". I used a
variant of this in _Chicken and Egg_ but I don't know if anyone a)
noticed, b) found it amusing, or c) discovered how pointless the actual
choice was.
Adam
Emily Short
<jeffnyman_nospam@_pleasenospam_gmail.com> wrote:
> One example we dealt with as a practice idea: the protagonist (PC) wanted to
> slaughter someone to get through an area. You, as the player, didn't want to
> do that. You found an alternate solution. The narrative allows the player
> character to realize that course of action was effective and customizes the
> player character's response the next time a similar situation comes up.
Hm, interesting.
> I just noticed that
> people who were not enmeshed, even peripherally in textual IF, had a very
> different view of the possibilities they saw and the means by which they saw
> those possibilities becoming realities. So I wanted to learn more about
> that.
Yes, I find that very useful input as well.
> I found the dissertation so I'll read that. I also agree that there are
> probably many things out there that showcase some or all of what we're
> considering. I can tell you that one problem we had as a group was deciding
> what games to try. We were trying to find concise summaries that would allow
> us to see what was unique about a given game, or what viewpoint it used (or
> whether it switched viewpoints), or what techniques it specifically used
> (menu-based conversations, for example), and so forth.
Some of this stuff can be coaxed out Baf's Guide; some of it I've
tried to address with lists on my website (e.g., lists of games that
do unusual stuff with the player/protagonist relation). But you're
right that it's not entirely easy to get at this kind of information
if you don't already know exactly what you're looking for.
Anyway, it sounds like we're mostly agreeing here! I mostly just
wanted to point out that recent IF contains a number of examples
already that might be useful for examining this question, and suggest
Jeremy's dissertation as a place to find both some critical
terminology and some game suggestions.
One thing I observed recently which is maybe relevant: games where the
player knows *more* than the viewpoint character, through knowing more
about how the world works than the VP character does, have had several
prominent recent examples: Lost Pig (vp char is not that bright),
Child's Play (vp char is a baby), and Treasures of a Slaver's Kingdom
(vp char is a stereotypical Conan-type hero, a little bit dumb).
ToaSK is an especially interesting case because it thrives on making
the hero comical to the player, through parody. It distances the
player and the protagonist with obvious don't-take-this-seriously
passages, like "You regard your rippling muscles in the mirror..." and
so on. I'm paraphrasing, and badly, because I don't have the game open
at the moment -- it's got very clever writing throughout -- but the
idea is that it is constantly telegraphing that the protagonist is a
specific sort of genre stereotype, and both the player and the author
know that and perceive him as an object of a certain amount of
mockery.
I found that that affected my goals as a player: I was more interested
in controlling the PC as a puppet in a way that made an entertaining
story, than in controlling the PC as though I shared all his goals and
identified with him.
So that's one direction this kind of dissociation can go. I didn't
feel as though the PC were "me" in any important sense at all. In
fact-- well, cut for spoilers, and if you don't want to read ahead
I'll understand, because it partly reveals one of the coolest moments
in the game.
S
P
O
I
L
E
R
S
At one point about 1/2 to 2/3 of the way through the game, there is a
situation where the player can see the PC is falling into a trap. This
is telegraphed by details in the PC's environment he notices but is
too stupid to grasp the meaning of. The game has laid careful
groundwork ahead of time with some references to a specific kind of
monster-human -- this appeared off-hand at the time, but it prepared
the player to notice when one of these actually turns up,
ineffectively disguised.
There is, I think, a point at which the player could try to extricate
the PC from the trap before things get worse. I keep meaning to replay
and see what happens if you try to flee at that juncture.
Now here's what I thought was interesting: when I played this, I did
*not* try to escape. Instead, I thought, Bwahaha, my idiot protagonist
is totally going to fall for this. *And I went along with it.*
Of course, at that point the game had already earned my trust, both by
repeatedly stating that the game couldn't be made unwinnable and by
being very well written, structured, etc., so that I felt I could
follow along wherever it went and still wind up somewhere enjoyable. I
don't think this gimmick would work nearly so well without that kind
of author-trust establishment.
Jim Aikin
>>
>> I'm not a fan of menu-based conversation, so perhaps my bias will be
>> embedded in this answer.
>
> Could you elaborate on that a bit? Why do you dislike it?
First, because it pulls the player away from the narrative flow of the
story. A conversational menu is like a big flashing neon sign that says,
"THIS IS SOFTWARE. THIS IS A GAME."
Second, because (as other posters have mentioned), it turns the alleged
conversation into a branching tree experience. The best way to proceed
is to save the game and then start over from the saved game file in
order to make sure you've explored every branch. This methodology pulls
you out of the flow of the story.
Third, because some of the branches may disappear for reasons that
aren't clear. If "ask Mrs. Warburton why she likes darjeeling so much?"
is available to me now, as a menu item, but not later, after I've asked
her about the health of her daughter Imelda, I'm pissed off. I would
much rather have a basic "ask her about darjeeling" command that will
remain available later, should I choose to use it.
>> With reference to item (1), the solution is, unfortunately, that the
>> author needs to do a lot of extra work to code unlikely or incidental
>> topics of conversation.
>
> Yes, but I think there are two reasons why this isn't really feasible.
>
> 1. As a game becomes larger, both the number of conceivable topics and
> the number of NPCs increases--but since the two have to multiplied in
> order to get the number of responses you have to write, things get bad
> very quickly.
>
> 2. The deeper your implementation, the bigger the expectations that the
> player will form about which topics she can converse about. If I don't
> implement conversation topics for all the nouns in the room the NPC is
> in, the player will notice this after a few abortive attempts at
> conversation, and will stop asking about nouns. But if I do, the player
> will start expecting that she can even talk about nouns in other
> rooms--and so on.
This is true to a considerable extent, yes -- but that doesn't make it
non-feasible. It's still feasible. It's just a lot of work.
I very much doubt that ANY of the players who have tried Lydia's Heart
has discovered all of the things that the NPCs can converse about. A
couple of them are very taciturn, while others are abundantly chatty.
>> With reference to item (2), I just now had an idea based on something
>> in Jeff's post. If the PC is truly a character (and not simply the
>> avatar of the player), then it can be enlightening to learn what the
>> PC is thinking -- because it might _not_ be what the player is
>> thinking. So perhaps the command 'think about what to say to Mrs.
>> McGillicuddy' would be a useful general form. (Or simply 'think about
>> what to say', provided Mrs. McGillicuddy is in the room.)
>
> I wonder--wouldn't players start doing this every time they saw an NPC,
> and a few more times just for good measure?
Sure, why not?
> Wouldn't this, in the end,
> come down to something very much like TADS-style in-conversation hints,
> only less useful because not always visible?
If you want to view the conversation topics, type the command that makes
them visible. That's not difficult. The point, to me (and I'm mainly a
writer) is to preserve the integrity of the narrative flow. A TADS-style
"You could ask her about her buttercups, or about World War III" seems
very intrusive to me, unless I've directed the PC to reflect on what
sort of conversations might be appropriate.
--JA
Blank
Perhaps we could extend the prompt a little, so that "> " is the
standard prompt, but say "+> " indicates there are additional, special,
commands available at this point. Those commands would be indicated in
the ordinary game text like you've done above, but will be listed
explicitly (either in sentence list form or as a menu) if the player
enters "special", "additional", "hint" or "+" at that point. (I'd have
"help" work slightly differently, calling up the standard help menu as
well as the extras.)
> But the problem is: what if the
> player wants to try all of them (assuming Alex doesn't blast them in the
> meantime -- which narrative should indicate is coming)? Does the menu just
> keep showing up until the player tries all combinations? If not, what if the
> player is curious to have tried another option with Alex but now can't?
>
> - Jeff
>
>
I'd like to be able to implement an auto-save at this point, as a
(hopefully) subtle meta-message that "it's okay to choose here, you can
easily come back and investigate other path(s) if you want to later."
Unfortunately I remember from a different thread that this had problems.
--jz
Smoov...@gmail.com
likes menus when used well in an IF work. I thought I'd chime in.
I do not think that the Ask/Tell system can believably sustain more
than a handful of NPC's in a game, especially in terms of good use of
the writer's time. If you look at the vast majority of IF, there is a
suspicious lack of people in most of them. The vacation spot/bar/
store/town after closing time, the edge of the party you've left
because of a headache, the mostly-empty dungeon full of traps, the
derelict spaceship, the mysterious village after sundown when all the
villagers lock themselves inside. More NPCs = a ton more work
crafting believability.
Skillful designers work around this. They use the sparseness as a
tool for generating mystery, to encourage inspiration. In at least
one excellent case the trick is worked it into the narrative
(Slouching Towards Bedlam). In another case there's the Plotkin game
(the name of which escapes me) where the character is in a theater but
obviously can't interact with anyone because there's a play on and
he's distracted by something else. Using the limitations of the
medium to inform the plot is great. But I argue that as time goes on
it's generating staleness in the medium.
And you could say that a handful of very well-fleshed-out characters
is better than many not-fleshed out characters, but this should not be
a limiting factor in design.
One of the benefits of menus is that they can support a larger cast.
I can't imagine playing City of Secrets without a partial menu system;
"inclusion in the narrative" would have amounted to infodumps and the
amount of ask/tell involved afterwards *still* would have driven me
batshit insane. Or more likely, driven me to type "quit".
In Photopia, menus were embedded in the structure itself (more on that
in a second). The menus allowed the relatively large number of scene
and viewpoint changes to feel less jarring (to me). Ask/tell would
have made the broken the overall flow of the narrative and lengthened
an IF that shouldn't be lengthened much (again, my opinion). Or it
would have made the IF into complete static fiction by removing the
main feling of interactivity I understand (from the years and years
of discussion) that a number people do not like the game for that
exact reason (the fact that the interactivity is largely illusion from
a choice perspective).
BUT: Photopia is the single best IF I've found to give to beginners.
Of the 40 or so people I've arm-twisted into playing it, 9 out of 10
have loved it. Do these people go on to play more involved IF? For
the most part no, but most of them would not have in any case. That's
worth examining. Perhaps not any valid evidence if someone's first
experience is with menus (to be fair, I've tried other works first),
but this still puts my personal observations in direct contradiction
to Jeff Nyman's observations.
And what I meant by "embedded in the structure" was this: since the
menus make up the bulk of the interaction, they don't feel as jarring
(to me). And I agree that the placement of the menus in a separate
window helped immensely in City of Secrets. I did not notice them
much in "The Baron" either, I suspect because they were present from
near the get-go and then for much of the rest of the game. So I think
that *if* menus are used, they should be considered in the overall
scheme of the design carefully.
I'd love to see more ask/tell menu integration (and have been
tinkering on and off with something like that for months). I'd also
like to see less "all over the map" choices in conversational menus.
If you want to establish part of the character's motivation/
personality by offering wildly different choices near the beginning of
the game, fine. Once it's established that the character is a certain
way, offer choices that deepen established traits. Don't keep
offering my rogue the chance to suddenly start talking like a saint
for no reason.
I obviously don't think menus are appropriate everywhere. For
example, the idea of the Zork games having a menu system makes me
throw up in the back of the throat a little. And I don't hate Ask/
Tell, but I hate getting stuck on Ask/Tell *way* more than I hate
menus. And standardized responses to what I consider reasonable
topics throws me out of the game more than menus unless very cleverly
done (the toolman pointing at the guidebook in For A Change). I'm
intrigued by the notion of a wide variety of responses based on mood
and current game state, or using those responses as hints, but
cautiously so.
I don't know. Maybe it's because I'm a younger player who grew up on
Lucas Arts games and had to play catch-up with Infocom. Maybe it's
just that I've played games with menus my entire life.
Other stuff:
-Bringing in a deus ex machina when the player is stuck better be
damned seamless unless the game is very lighthearted. Emily's covered
my concern about time and resources with this one.
-Jeff, I agree with your clarification of character identification,
which seems to be what most people mean when saying "character
identification". I don't think anyone was seriously saying the
character *had* to be "just you", although as I understand that was
part of the draw of the early text adventures (surprise at seeing the
adventurer's appearance in Enchanter, etc.). I think there's a time
and place for more in-depth player-as-character, and have enjoyed a
number of works that would have suffered from being more like novels
in terms of protagonist characterization. Not ironically these are
usually the more game-like or puzzle-focused. Alternately it was a
different kind of characterization. In "Hunter in Darkness", the
protagonist had no character, and the game would have suffered for it
because the identification came from imagining yourself in the
horrible physical situations of that work. It wasn't the *character*
stuck in a hole bleeding to death, it was *you*. This is the
potential strength of second person. Is your narrative style still
applicable when there are no other characters?
Blank
Ooo, thanks! Noted.
--jz
Blank
> Blank wrote:
>>
>> Well, the npc doesn't actually need to know they're helping -
>
> True enough.
>
>> Every turn:
>> if the bees are known, and the violent argument has happened and the
>> interior of the clock has not been examined begin;
>> move mrs jensen to the location;
>> say "Mrs. Jensen staggers in with her hands over her ears. 'Oh, my
>> heavens!' she moans. 'The clock's gone crazy! It's striking four
>> o'clock over and over!'";
>> the crazy clock begins in one turn from now;
>> award no points for discovering the oily gloves; [because I'm mean]
>> end if;
>
> As written, that would have Mrs. Jensen entering the room every turn. I
> know you're only giving an example here, not trying to write a working
> game, but there are lots of fiddly little problems like that, all of
> which would have to be handled gracefully.
>
True enough.
> Like, how many turns do you let the player flounder around before Mrs.
> Jensen arrives?
>
I'd pick a number and then tweak it in beta.
> And what is Mrs. Jensen doing in the meantime? In other words, what
> would her function be in the story if the player was doing everything
> right, and didn't need reminding?
>
> --JA
Well, either Mrs. Jensen is a natural candidate: it's her house and the
story is about her, or I'd write the situation to include her as
background. So in that case her "function" acutally would be as the
safety-net and her visible activities would just be make-work. Of course
it's all situational: if you have a game where the player can fly to
Venus in a bathtub then having Mrs. J stagger onstage in the Venusian
lot would blow the gaff rather. But hell, that's a design problem and
the solution will be different in each case.
--jz
Jeff Nyman
news:610869e7-4a1a-40a9...@h11g2000prf.googlegroups.com...
> I do not think that the Ask/Tell system can believably sustain more
> than a handful of NPC's in a game, especially in terms of good use of
> the writer's time.
I completely agree. This matches most of what you have in all storytelling
situations: a relative minimum of well-rounded characters, the rest being
more or less flat in nature.
> More NPCs = a ton more work crafting believability.
I agree. And that believability factor me is based on incorporating the NPCs
intelligently into the plot and making dialogue options effective and
efficient, both from a storytelling perspective (narrative drivers,
characterization) and from a player perspective (knowing how to interact
with the NPCs).
> Using the limitations of the
> medium to inform the plot is great. But I argue that as time goes on
> it's generating staleness in the medium.
This was something that my group felt as well. There was this feeling that
there was a certain "sameness" to textual IF, as if the same author was just
trying new story genres. Although "Slouching Towards Bedlam" was a great
counter-example to this as was "Babel." There were others as well.
> In Photopia, menus were embedded in the structure itself (more on that
> in a second). The menus allowed the relatively large number of scene
> and viewpoint changes to feel less jarring (to me). Ask/tell would
> have made the broken the overall flow of the narrative and lengthened
> an IF that shouldn't be lengthened much (again, my opinion).
Agreed on this and we did look at "Photopia." Most people in my group didn't
like it. By itself, that means nothing of course since that's an "eye of the
beholder" thing but what they didn't like was that it -- and this is a
direct quote from one author, based on my notes -- "seemed like it wanted to
tell a story but couldn't figure out how." It was felt that "Photopia" would
have been better as static fiction and not textual IF, which goes to your
point: "Or it would have made the IF into complete static fiction".
> BUT: Photopia is the single best IF I've found to give to beginners.
> Of the 40 or so people I've arm-twisted into playing it, 9 out of 10
> have loved it.
Interesting -- because as you later say -- this directly contradicts what I
found. When you say they "loved it", what was it that they indicate was
enjoyable?
> And what I meant by "embedded in the structure" was this: since the
> menus make up the bulk of the interaction, they don't feel as jarring
> (to me). And I agree that the placement of the menus in a separate
> window helped immensely in City of Secrets. I did not notice them
> much in "The Baron" either, I suspect because they were present from
> near the get-go and then for much of the rest of the game. So I think
> that *if* menus are used, they should be considered in the overall
> scheme of the design carefully.
Yeah, this was something we considered as well. We didn't have "City of
Secrets" as part of our list (and probably should have, by the sound of it)
but we did go over the idea of how menus (and other elements, like a running
inventory list that's always visible) could be presented. We did use the
"Spellcasting" games and "Timequest" game from Legend. These were found to
be, almost unanimously, cumbersome and distracting. People didn't like the
separate windows because it, again, distracted from the narrative. Some of
my programmers in the group felt that it "exposed the implementation
model" -- which it didn't, but I got their point, which was that it exposed
the game mechanics more obviously than a menu (or other element) in-game
would have done.
But, please keep in mind, too, that this group had very little exposure to
textual IF. Some had played it years ago but I was careful to select people
who didn't have any exposure in the last decade or so, beyond hearing about
it or knowing roughly what it was.
> Once it's established that the character is a certain
> way, offer choices that deepen established traits. Don't keep
> offering my rogue the chance to suddenly start talking like a saint
> for no reason.
This was definitely an area we looked at and that I've alluded to in various
posts here: the idea to "grow" the protagonist (player character) in a
certain way, such that you eventually start to constrain how the protagonist
presents the world to you, the player. In other words, you might make the
guy such a jerk that his descriptions of locations or objects or people are
now colored by that mood or that attitude. This was really tricky to do well
because it involved the player getting to know the player character but also
participating in their growth (or decline) as a person.
> throw up in the back of the throat a little. And I don't hate Ask/
> Tell, but I hate getting stuck on Ask/Tell *way* more than I hate
> menus.
I think this is an excellent way to put it and I do think this matches what
I found as well. As I had mentioned in the other thread (about teaching
storytelling), most people in my group *did* see how menus could be
effective -- *given the way that NPCs and conversations were currently
handled*. What was wondered is if better narrative or more integrating of
the NPCs as "puzzles" in their own right -- as solutions to be gained just
as much as opening a treasure chest -- would alleviate some of the ASK/TELL
frustrations or the ability to get stuck where you either give up asking or
you try everything you can think of, thereby sitting through a hundred
"don't know, don't care" type responses from the NPCs.
> usually the more game-like or puzzle-focused. Alternately it was a
> different kind of characterization. In "Hunter in Darkness", the
> protagonist had no character, and the game would have suffered for it
> because the identification came from imagining yourself in the
> horrible physical situations of that work. It wasn't the *character*
> stuck in a hole bleeding to death, it was *you*.
Absolutely! I think that's a great point to keep in mind. There are
definitely some games where you probably do want the player to imagine they
are truly in this situation. One of my authors wanted to work out a "hostage
scenario" concept wherein the descriptions of what was happening were
extremely brutal -- and quite frankly, very hard to "play through" -- but,
damn, was it effective! And my belief is that it was effective because it
made the reality so stark and forced you to consider that this situation was
happening to *you*.
However another author pointed out that this could be made just as bad by
making *you* (as player) relate to a character that has to do some
despicable things. To this author, the strength came from you being
complicit in the acts of someone else and having to decide whether you
agreed or not but also forcing you to empathize and trying to act as *you*
would but also dealing with how your *player character* wants to.
By the way, if you want to see what this last example was based on, read
"Survivor" by J.F. Gonzalez.
(http://www.amazon.com/gp/product/0843955678/ref=wl_it_dp?ie=UTF8&coliid=I1GJNGPE6G4K43&colid=3KRG5DTE2RW6F)
However, be prepared for some pretty graphic stuff.
- Jeff
miket...@embarqmail.com
one) it seemed to me that Jeff's study group was very much against the
idea of player as player character. They wanted the protagonist to be
well defined (by the author), but not necessarily clearly detailed (to
the reader), at the outset. They also didn't seem to be interested in
emergent solutions, etc. This was a little depressing to me, because
like a few others around here (Victor?), I love the RPG genre and have
always wondered about and been a little dismayed at the relative lack
of them in textual IF. I had assumed this was because they were
difficult to do well, and that as a result, very few succeed. (I'm
not talking about any specific games, just a general impression). Now
I'm faced with the realization that people may not even want them.
Which brings us to:
On Mar 12, 10:07 am, "Jeff Nyman" >
> This was definitely an area we looked at and that I've alluded to in various
> posts here: the idea to "grow" the protagonist (player character) in a
> certain way, such that you eventually start to constrain how the protagonist
> presents the world to you, the player. In other words, you might make the
> guy such a jerk that his descriptions of locations or objects or people are
> now colored by that mood or that attitude. This was really tricky to do well
> because it involved the player getting to know the player character but also
> participating in their growth (or decline) as a person.
>
Now we're talking. You see, I'd alway dreamed of a text IF - style
interface CRPG where the player actually created the character in the
course of the game (as opposed to picking classes and attibutes at the
start). In a sense, it would be more of a role *creating* game (or
RCG, for those of you who don't think we have enough acronyms).
However what I was thinking about was not growing a character simply
by leveling - up with combat and treasure hunting. I wanted a game
where a character's motivations and challenges are tailored to the way
a particular player wanted to go. I realize this is probably not what
Jeff was suggesting in terms of the scope of open - endedness. I also
know it would be incredibly naive to suggest that something like this
could be accomplished overnight, but the issues we're discussing here
would go a long way to aiding that quest (so to speak). Two examples:
- The idea of tailoring textual output based on earlier choices the
player has made seems to be a relatively easy one from a coding
aspect. Obviously an author would still have to write the different
descriptions and anticipate possible directions. Of course, we
already have to do that every time we implement a new object.
- The idea that Emily has posted about and IIRC even incorporated in
one of her games of tailoring multiple solutions to differing styles
of play.
Obviously, these concepts have other, more typical IF (if there is
such a thing) applications. They should be definitely be pursued.
Skinny Mike
Adam Thornton
<miket...@embarqmail.com> wrote:
>Now we're talking. You see, I'd alway dreamed of a text IF - style
>interface CRPG where the player actually created the character in the
>course of the game (as opposed to picking classes and attibutes at the
>start). In a sense, it would be more of a role *creating* game (or
>RCG, for those of you who don't think we have enough acronyms).
>However what I was thinking about was not growing a character simply
>by leveling - up with combat and treasure hunting. I wanted a game
>where a character's motivations and challenges are tailored to the way
>a particular player wanted to go.
You *have* played _Planescape: Torment_, haven't you?
If you haven't, you really should.
It's not really all *that* freeform--in fact, I don't think the quests
change at all, although whether you accept them and how you choose to
complete them is fairly open--but there is quite a lot of what you call
role creation in it.
Adam
Jim Aikin
>
> Perhaps we could extend the prompt a little, so that "> " is the
> standard prompt, but say "+> " indicates there are additional, special,
> commands available at this point. Those commands would be indicated in
> the ordinary game text like you've done above, but will be listed
> explicitly (either in sentence list form or as a menu) if the player
> enters "special", "additional", "hint" or "+" at that point.
Or "commands". This is a great suggestion! Powerful, yet unobtrusive. I
could imagine situations in which several different "prompt prompts"
might appear within a game -- such as +> and => and :> and ]> . (I
haven't yet imagined the situations, but I _could_.)
--JA
miket...@embarqmail.com
>
> You *have* played _Planescape: Torment_, haven't you?
>
> If you haven't, you really should.
>
No, I'm a little embarassed to say, I have not. The last graphical
adventure game I played was some vampire thing on my old roommate's
PS. Before that, one of the Ultimas I guess. I was really referring
to purely text games. Others have recommended it and I will check it
out, though. Thanks!
Skinny Mike
David Fisher
news:1ed04f0a-e799-458d...@c33g2000hsd.googlegroups.com...
> On Mar 12, 10:07 am, "Jeff Nyman" wrote:
>>
>> This was definitely an area we looked at and that I've alluded to in
>> various
>> posts here: the idea to "grow" the protagonist (player character) in a
>> certain way, such that you eventually start to constrain how the
>> protagonist
>> presents the world to you, the player.
>
> interface CRPG where the player actually created the character in
> the course of the game
> I wanted a game where a character's motivations and challenges
> are tailored to the way a particular player wanted to go.
I love the idea of a game which observes the player's choices and
adjusts itself accordingly ...
Non-IF games which do this usually seem to be focused on moral
choices (choosing the dark side or the light side). For me, it
would be more interesting to be able to influence the PC's
personality in a more general way.
Here are some personality-related words that might be inspirational:
Confrontational, gentle, vulnerable, trusting, naive
Courageous, foolhardy, tough, careful, overcautious
Intimidating, indecisive, manipulative, calm, irrational
Cunning, secretive, obnoxious, tenacious
Cooperative, independent, non-conformist
Scrupulous, forgetful, messy, meticulous
Eccentric, paranoid, implusive, obsessive, confused
Respectful, self-indulgent, greedy, lazy, petty
Compassionate, sociable, uncommunicative
Playful, clueless, romantic
A game that could work out which of these attributes was consistent
with the player's choices (and characterised the PC that way for
the rest of the game) would be amazing ...
David Fisher
David Fisher
news:1ed04f0a-e799-458d...@c33g2000hsd.googlegroups.com...
>>
>> This was definitely an area we looked at and that I've alluded to in
>> various
>> posts here: the idea to "grow" the protagonist (player character) in a
>> certain way, such that you eventually start to constrain how the
>> protagonist
>> presents the world to you, the player.
>
> interface CRPG where the player actually created the character in
> the course of the game
> I wanted a game where a character's motivations and challenges
> are tailored to the way a particular player wanted to go.
I love the idea of a game which observes the player's choices and
adjusts itself accordingly ...
Non-IF games which do this usually seem to be focused on moral
choices (choosing the dark side or the light side). For me, it
would be more interesting to be able to influence the PC's
personality in a more general way.
Here are some personality related words that might be inspirational:
Victor Gijsbers
<jeffnyman_nospam@_pleasenospam_gmail.com> wrote:
> So there are two forms of storytelling -- graphical adventure games and
> novels -- that can evoke emotions but that do not require the reader/player
> feeling responsible in some sense.
Well, of course. I don't dispute that; and interactive fiction can
evoke emotions without making you feel responsible or anything like
that. And these techniques might be enough if what you are interested
in is simply writing a game with an emotionally engaging story. But
they are _not_ enough if you want to explore the possibilities that
interactive fiction offers above and beyond static (or dynamic but non-
interactive) narrative. I mean, why make a fuss about interactive
narrative if you're really just going to use the exact same techniques
that are already possible in non-interactive narrative?
> If I want my emotions engaged, conflating myself with a protagonist can only
> hurt because then I'm just looking at my own view of the world and my own
> reactions, and I already know that view and those reactions. What I want to
> do is be forced to consider alternative viewpoints and alternative actions
> that might be taken.
I think there might be some terminological confusion here. When I am
talking about identification with the player character, I mean taking
on the role of the protagonist, something which partly consists of
adopting (to a certain extent) the points of view of that character. I
don't think it would be useful, interesting and perhaps not even
possible to do that without also hanging on (again to a certain
extent) to your own viewpoints--but it is certainly something
radically different from just 'being yourself', just taking an
external position.
So as I am using the term, identification with a protagonist _forces_
you to consider an alternative viewpoint. If you are using the term in
such a way that identification with a protagonist _precludes_
considering an alternative viewpoint, we must be talking about
something different. :)
I am not as sure as you are that you already know your own viewpoints
and reactions--however, this might be better reserved to a discussion
of its own.
I'm going to say more in reaction to this post, and to other posts in
this thread as well, as soon as I can.
Kind regards,
Victor
Smoov...@gmail.com
through my post, it sounded a bit more confrontational than I meant it
to, so apologies for that. The things you're bringing up need to be
brought up, and I think it will result in pure gold down the line. A
couple of thoughts:
> I completely agree. This matches most of what you have in all storytelling
> situations: a relative minimum of well-rounded characters, the rest being
> more or less flat in nature.
>
Agreed, except that in IF a flat character can look *very* flat. In a
book, a supporting character can have one line of description and
that's fine, because they're never going to show up again. The author
has the power to move your attention away from that character. In IF,
you often have the ability to sit there hammering the character with
all sorts of conversation/action gambits, and that shows the holes
very quickly. If the character is not around for very long it might
work fine. This is harder than it sounds. Consider the number of
times where an NPC is in one scene, but the scene doesn't change until
the player does something non-conversation-related.
> > BUT: Photopia is the single best IF I've found to give to beginners.
> > Of the 40 or so people I've arm-twisted into playing it, 9 out of 10
> > have loved it.
>
> Interesting -- because as you later say -- this directly contradicts what I
> found. When you say they "loved it", what was it that they indicate was
> enjoyable?
>
After you brought this up, I went back and asked a few of them (6 have
responded so far, a couple of others are aquaintances I don't talk to
much these days). I also asked about the menus, and got some
interesting answers. A few of them felt that the menu/banal real
world to freeform/fantasy world dichotomy worked (in a sense, the
"ugliness/ lack of choices" with the menus actually made the color
moments stand out more in a pleasing way). They all mentioned the
blue maze puzzle solution as being their favorite part of the game.
They all said the menus were "not a big deal".
I suspect that the similarity with static fiction was actually a bonus
with a lot of the people I pushed it on. I remember making no real
attempts to explain what IF was, just "You're bored, this is fun and
easy, these are some helpful commands in some of the bits".
In one case (my brother), I tried moving him on to other IF works. He
says he tried Slouching and Anchorhead but they "made his head
explode" (my brother is far from stupid and reads a book a week,
clarification is that he lost interest part of the way through),
although he apparently enjoyed Enchanter and more recently Lost Pig
(all four of these were some of many recommendations I made).
If I had presented games in a classroom setting as part of a
comparative analysis, I think I might have had different results. But
instead I'm dealing with casual gamers/ readers, many of whom will
never actually finish a truly involved work of IF because learning the
interactions of the medium are not worth it to them for one reason or
another. Every time I've tried giving someone new to IF a more
involved work in the first place, they never actually finish it or ask
if if there's anything "simpler". The problem is, many games
explicitly designed as beginner's fair are usually effing boring, or
written "for 10 and up" which translates as effing boring to an
intelligent 20-year-old.
From the literary angle, I think your data may result in better book-
lover to IF reader conversions in the medium (as opposed to my end,
which is often "literate gamer to IF reader"). If you're bringing
people in from this angle the menus probably will not work as well.
> Yeah, this was something we considered as well. We didn't have "City of
> Secrets" as part of our list (and probably should have, by the sound of it)
> but we did go over the idea of how menus (and other elements, like a running
> inventory list that's always visible) could be presented. We did use the
> "Spellcasting" games and "Timequest" game from Legend. These were found to
> be, almost unanimously, cumbersome and distracting. People didn't like the
> separate windows because it, again, distracted from the narrative. Some of
> my programmers in the group felt that it "exposed the implementation
> model" -- which it didn't, but I got their point, which was that it exposed
> the game mechanics more obviously than a menu (or other element) in-game
> would have done.
>
Agreed to a degree. I didn't mind the multiple windows in Narcolepsy,
perhaps because pains were taken to make it look aesthetically
pleasing. If it had just been one text window grafted onto another
text window, I don't think I would have enjoyed it as much.
It's actually for a similar reason that I don't like the idea of
"think about conversation topics". Every time I have to type a meta-
command, it throws me out of the experience and exposes the puppet
strings. It's something about the act of actually typing the command
that does it. Inventory checks are okay, given that I can visualize
the protagonist looking or feeling around in his/her pockets. "What
is" can occasionally slide by if the answers fit very well in tone
with the rest of the story.
>
> This was definitely an area we looked at and that I've alluded to in various
> posts here: the idea to "grow" the protagonist (player character) in a
> certain way, such that you eventually start to constrain how the protagonist
> presents the world to you, the player. In other words, you might make the
> guy such a jerk that his descriptions of locations or objects or people are
> now colored by that mood or that attitude. This was really tricky to do well
> because it involved the player getting to know the player character but also
> participating in their growth (or decline) as a person.
Agreed, but it's something that just plain needs to happen more.
>What was wondered is if better narrative or more integrating of
> the NPCs as "puzzles" in their own right -- as solutions to be gained just
> as much as opening a treasure chest -- would alleviate some of the ASK/TELL
> frustrations or the ability to get stuck where you either give up asking or
> you try everything you can think of, thereby sitting through a hundred
> "don't know, don't care" type responses from the NPCs.
I think it *might* in many cases. Although I'm not sure what you mean
by making the NPC more like a puzzle; to me the ASK/TELL NPC already
seems like a puzzle. I didn't mention this in my last post, but
better integration of the narrative would improve the quality of IF in
general regardless of fixing this problem. And I agree that it's
certainly worth exploring.
I have some thoughts on avoiding the problem of "save and see all of
the options" with menus, but I'll save those thoughts for another
thread once I've codified 'em a little.
I'd like to mention that I missed the other thread before reading this
one originally; discussions of conversation menus always bring me out
of the woodwork precisely because I'm trying to find a better way to
do it. Or failing that, throw them out and use a better option.
Jeff Nyman
news:c199b478-c433-49e4...@s13g2000prd.googlegroups.com...
> Agreed, except that in IF a flat character can look *very* flat. In a
> book, a supporting character can have one line of description and
> that's fine, because they're never going to show up again. The author
> has the power to move your attention away from that character. In IF,
> you often have the ability to sit there hammering the character with
> all sorts of conversation/action gambits, and that shows the holes
> very quickly.
Which then might suggest that this NPC shouldn't be left around as part of
the narrative. In other words, the fact that the player can hammer the
non-responsive NPC might mean that either (1) the NPC shouldn't be there or
(2) a "plot manager" of some sort should remove the NPC.
> If the character is not around for very long it might
> work fine. This is harder than it sounds. Consider the number of
> times where an NPC is in one scene, but the scene doesn't change until
> the player does something non-conversation-related.
That's true. What we were trying to look at was "plot managers" (I know some
call them "drama managers" but I'm not big on that term; but whatever, I
suppose). The idea here is that the scene does change based on the dictates
of the plot, not just on the player actions. Of course, this (usually) isn't
real-time so to some extent it is, in fact, based on player actions.
But the plot manager can make sure, on the next action, to contextually
decide how to deal with things like a non-responsive NPC. If the player
*doesn't* try to talk to them and the NPC is useless or is just filler,
maybe a description of them moving off? If the player *does* try to them and
the NPC is usless or is just filler, still hvae them walk off, but the
description might now be different. The key is that the plot manager knows
what has to happen in a given scene and works to make those things happen,
at least to some extent.
There's a whole lot hiding behind this kind of idea so I realize what I'm
describing here is not as simplistic as I'm describing it.
> From the literary angle, I think your data may result in better book-
> lover to IF reader conversions in the medium (as opposed to my end,
> which is often "literate gamer to IF reader"). If you're bringing
> people in from this angle the menus probably will not work as well.
Yeah, and therein lies a little of my own problem. Clearly I had a selection
effect at work. Now, granted, it's one I imposed on myself. But one idea
that's fluttered around in my head is that maybe textual IF could appeal to
a wider "readership" if they weren't so much like games and more like
stories or novels, in the sense that you have some of the same elements that
people who read novels enjoy. The trick is: how do you keep the elements
that make textual IF interactive (and thus challenging) while also allowing
for the static elements (i.e., non-interactive) in a novel?
It's tricky. At least to me. So I was hoping that a group of writers could
help me see the form with different eyes. I come from a minimal writing
background myself, but I haven't published my first novel yet -- and who
knows if I will -- so I didn't lump myself in as "writer." In fact, I have
just enough distance from textual IF and novel writing that I was hoping my
relative ignorance would actually help. The jury's still out on that one.
Anyway, I strayed a bit here. The point was that there's a wide readership
of novels out there. People like good stories. People like engaging
characters. It seems like a ready-made audience, at least potentially, for
textual IF. If you can convince people that it's not "just" a game, but
really an Interactive Novel -- well, maybe that would work as a selling
point. Then you have to explain what "Interactive" in this context means --
but that also means understand what level of interactivity works while still
not destroying the novel-like aspect.
I'm making less sense as I go here so I'll stop. (This stuff is clear in my
mind; it just usually can't get any futher. Perhaps I need to rethink this
"writer" thing, huh?)
> It's actually for a similar reason that I don't like the idea of
> "think about conversation topics". Every time I have to type a meta-
> command, it throws me out of the experience and exposes the puppet
> strings. It's something about the act of actually typing the command
> that does it. Inventory checks are okay, given that I can visualize
> the protagonist looking or feeling around in his/her pockets. "What
> is" can occasionally slide by if the answers fit very well in tone
> with the rest of the story.
Okay, now here's an interesting point. One thing I did notice that people in
my group didn't mind was a separate "inventory window" and sometimes even a
separate "room description" window. Here was the logic I got when I
questioned them on this. They said that they could see that as being things
the protagonist would have "in memory" and thus would always be available
"in their mind." So having the inventory and the room description always
displayed in some way "made sense" to people because this was like a
snapshot of the protagonist's memory made available for the player.
This is where people with a writing focus saw the mechanics of the game
format being useful.
So what we did, in one case ,was experiment with having a separate window
for room descriptions. If that description changed, something in the window
changed to indicate that. This let the reader/player know that they could
look at that description again, as opposed to always checking it to see if
there was something new.
The wider point here being that I found certain elements were found less
intrusive to the "narrative experience" if those elements could be worked
into the metaphor of the player "playing a book." That's vague, I know, but
the two concrete examples were the idea of the always-available inventory
and the always-available room description. Another idea we kind of played
around with was a "relations window", which is similar to the idea of
'suggested topics' but it was a window that kept in mind the relationships
that the player/protagonist had observed. Something like this:
John ==> Kissed Jodi
Karen ==> Saw John kiss Jodi, told Susan
Susan == Brought up "stolen emerald"
Susan ==> Suspects John
So this suggests that Susan suspects John of something. Stealing the emerald
perhaps? But Susan heard from Karen about the kiss. So is she just ticked
off at John and mentioned her suspicious to "get back at him"?
Okay, rotten example, but this is the kind of thing we were trying to focus
on. Now, again, these are relations that the protagonist would have
experienced in some way (either via being told or seeing) and would have
those in his "memory" (thus being okay in a separate window as non-narrative
breaking). This is also why I've said that I really think the relations of
Inform 7 could be used much more than I think they currently are.
> I think it *might* in many cases. Although I'm not sure what you mean
> by making the NPC more like a puzzle; to me the ASK/TELL NPC already
> seems like a puzzle.
I guess what I meant here is that you have to figure out the characters:
what makes them tick? What will motivate them to take an action on your
behalf? If you've already ticked them off, what can you do to make it
better? If you killed them and later realized you really needed a piece of
information they had, what can you do about it? Can you turn one NPC against
another? If so, to what end? Will telling "Jack" what "Susan" just said
about him make him fly over the edge? Will it lead to a confrontation? But
what if you want to avoid a confrontation -- at least until you've found a
bit of evidence? Can you do that? What's the best way to keep "Jack" and
"Susan" from getting mad at each other?
In a good novel you generally want to explore the characters just as much as
the setting. It's the characters that are memorable and that drive a lot of
the action. They do what they do for reasons; and what they do has
ramifications on other characters and on the protagonist. The protagonist
then has to deal with those ramifications, not just by getting past a locked
door, but getting the NPC to unlock the door for them.
So I think you're correct: ASK/TELL becomes a "puzzle" in some ways, but of
a minimal (or perhaps weak) sort. It's a puzzle largely because of its
implementation and the fact that often people are "encouraged" to just
ASK/TELL about anything and everything. So I think that (at least in part)
led to the idea of menus and then eventually 'suggested topics', which could
constrain that "encouragement" a bit and focus the conversations.
- Jeff
Jeff Nyman
> John ==> Kissed Jodi
> Karen ==> Saw John kiss Jodi, told Susan
> Susan == Brought up "stolen emerald"
> Susan ==> Suspects John
>
> So this suggests that Susan suspects John of something. Stealing the
> emerald perhaps? But Susan heard from Karen about the kiss. So is she just
> ticked off at John and mentioned her suspicious to "get back at him"?
One thing I failed to make clear is that this list above can also provide
the player with ideas to ask about and, if done well, maybe suggest who
would and would not be able to speak to some things. Further, the
"relations" may provide clues that the protagonist picked up on but the
player did not: thus a gentle nudge but without having to ask for a "hint."
This was an area I was trying to get the group to focus on: the idea of a
sort of collective knowledge base of the world (game) being brought up such
that the player could reasonably consider things and plan out courses of
action. Further, perhaps it becomes clear to the *player* that the
*protagonist* missed an obvious clue, based on the relations available. The
protagonist described a scene or a confrontation a certain way, as part of
the narrative. The player, in looking over the relations list (and that's a
bad name), realizes that maybe the protagonist didn't quite get what really
happened.
This goes back to the idea of the player having to "worth with" the
protagonist to some extent; to not just see the world through their eyes but
to see through their motivations, biases, perceptions, etc.
- Jeff
Jeff Nyman
embarrassingly wrote in message
news:g_qdnYoRxJfTFUfa...@comcast.com...
> This goes back to the idea of the player having to "worth with" the ...
Good Lord! Make that "work with".
Sheesh. I need to lay off the caffeine in the morning.
Victor Gijsbers
> When depicting the life of a fictional character, a novelist usually has to
> choose to include only those impressions, thoughts, reflections, sensations,
> feelings, desires, and so on, that bear on the character's motivations,
> development, and decision-making abilities. In other words, those aspects of
> character that will affect the way in which the character copes with the
> dilemmas they're going to face in the course of the story. But that implies
> there are impressions, thoughts, reflections, sensations, and so forth. I
> don't just want my impressions or thoughts; I can get to those without
> playing a game. What I want is to investigate a situation that allows me to
> play as a different person: and then see how I might have to react given
> that I am this different person. (Who knows? Maybe I'll find out I can
> conflate myself with the protagonist -- but maybe not.) The point is I have
> more narrative possibilities when I can open up a character that is very
> different from the reader.
I agree with almost everything you say; so let's assume (as I did in my
previous post) that we were talking past each other on the
identification issue. We both believe that strong characterisation is
(generally) better than having an "empty slot" protagonist, a void that
the player has to fill with himself. (I can see how you might be tempted
to call that process "identification with the protagonist". Personally,
I was using that phrase to designate something like crawling into the
skin of a strongly characterised fictional person.)
I don't agree with _all_ the detail of your story; especially not with
the part where you seem to argue that the aim of fiction is to
faithfully recreate how certain people act / how human nature works. I
believe that fiction is the realm of possibility, not of necessity; of
redemption, not of knowledge. The psychologists can tell me how people
_do_ react in certain situations. I want the novelist to tell me how
people _could_ react in those situations, and not just any 'could', but
a 'could' that grants me a glimpse--however vague and transitory--of new
and unknown heights of compassion, love, beauty, tenderness, humanity.
What we take away from "Anna Karenina" is NOT the knowledge that a
certain type of woman in a certain type of circumstances will do things
that will make them profoundly unhappy. What we take away from "Anna
Karenina" is a vague and transitory conviction that a state of grace
might be reached and just a little insight into the demonic forces that
might keep us away from it. (And part of Tolstoy's greatness is that
although he conceives of this in explicitly religious terms, he does not
force us to do so.)
But if we want to discuss this topic further, I think we should do so in
another thread.
> Yes, and in that case, the above response of mine probably wouldn't be
> issued. If it was random topics, you could perhaps contextualize a bit on
> standard non-replies, as it were. Like I said, I didn't find much resistance
> to people when they saw a clear "game"-imposed boundary on how much an NPC
> would respond to numerous topics. Everyone understands it's just an
> interactive element that is not a real person. They don't expect this NPC to
> act like a real person (just like readers of novels don't expect dialogue
> that matches how people "really" talk).
Okay. That is good to know.
> What was found to be more unforgivable is when the NPC didn't contextualize
> based on obvious elements key to the story *or* key to character
> development: helping the reader/player better understand characters and
> their motivations, including that of the player character. For example, in
> the above case, if it was important that Jack explain what happen to his
> wife all those years ago, it would be disappointing if questions along those
> lines just had Jack always refusing to answer, without some context to his
> action. Context might be: "Jack stormed off. Apparently I had pushed that
> topic too far." If it wasn't important about what happened to Jack's wife
> years ago, then it probably should have never been brought up in narrative
> in the first place, unless you are using it as a red herring. (But even then
> the red herring does play a key point in the narrative.)
> Or if the game allowed you to be a bit more specific for a situation. What
> about >REMIND ALEX OF OUR FRIENDSHIP.
Yes, but either I'll have to implement a huge number of 'remind'
responses, or I have to contextually tell the player that she can now
use the verb 'remind'. Your suggestions as to how to do the latter are
interesting, but call for a very specific style of writing--almost
necessarily first person, very self-conscious and reflective; a style of
writing that I can see working for a short noir detective story,
perhaps, but not for many other projects.
> It's tricky. Is it "obvious" from that the player can type "beg for my life"
> or just "beg"? Is it "obvious" they could say "remind alex of our
> friendship" or maybe even "apologize for son's death"? I actually agree that
> menus can make that a bit more explicit. But the problem is: what if the
> player wants to try all of them (assuming Alex doesn't blast them in the
> meantime -- which narrative should indicate is coming)? Does the menu just
> keep showing up until the player tries all combinations? If not, what if the
> player is curious to have tried another option with Alex but now can't?
This latter 'problem' is something that Jim Aikin also brings up. I am
not 100% sure I understand it, though. There seems to be a tension. On
the one hand, the player (presumably) wants to steer the story one way
rather than another. (Otherwise, why play interactive narratives?) But
on the other hand, if I understand you and Jim right, the player wants
to see _all the text_, or _everything that happens_, or _all the
possibilities of the game_.
But these two are simply incompatible (unless you are willing to do
replays). So what _does_ the player want? Does he want choices with
consequences, or does he want to be able to read everything? I cannot
give him both, as a matter of logic.
Regards,
Victor
Conrad
<jeffnyman_nospam@_pleasenospam_gmail.com> wrote:
> "Jeff Nyman" <jeffnyman_nospam@_pleasenospam_gmail.com> incorrectly andembarrassingly wrote in message
>
> news:g_qdnYoRxJfTFUfa...@comcast.com...
>
> > This goes back to the idea of the player having to "worth with" the ...
>
> Good Lord! Make that "work with".
>
> Sheesh. I need to lay off the caffeine in the morning.
Jeff, this has been a very productive couple of threads; I want to
thank you for it.
And, I have an implimentation question about conversations in TADS 3
-- I realize this thread has been mosty theory, and I welcome any
theoretical/gut responses to this... But it's also a real example I'm
working on.
I'm trying to impliment a multi-verb ask/tell conversational system
where the player's speech is not reported.
For example:
You are in a small concrete cell with metal bars.
The guard sits on a stool outside.
> ASK GUARD ABOUT KEY
"Shaddup," the guard says. "Nobody's going to criticize me for
talking to a prisoner."
> CRITICIZE GUARD
A tear forms in the guard's eye. "Now don't you turn on me."
> _
-- Has anything like this been done in TADS 3? Anything I can ...
steal?
Conrad.
Victor Gijsbers
> Man. This post makes me feel like the only *player* of IF who actually
> likes menus when used well in an IF work. I thought I'd chime in.
Thanks, that's useful input as well.
We shouldn't be forgetting that Jeff's sample is not just biased in
terms of all of them being more or less new to interactive fiction; it
is also biased in that all of them are writers. Now menu's are used a
lot in computer games, especially RPGs and graphical adventures, while
they are never ever used in static fiction. So perhaps his sample of
people was biased _against_ menus through a mechanism I hadn't
originally thought of. And if that is the case, I shouldn't worry too much.
Still--I want to think through the possibilities very, very carefully.
By the way, your idea about menu's being present from the beginning or
not cannot explain Jeff's students' negative reaction to Fate: that has
menus present from the start as well.
Regards,
Victor
Victor Gijsbers
> Victor Gijsbers wrote:
>>>
>>> I'm not a fan of menu-based conversation, so perhaps my bias will be
>>> embedded in this answer.
>>
>> Could you elaborate on that a bit? Why do you dislike it?
>
> First, because it pulls the player away from the narrative flow of the
> story. A conversational menu is like a big flashing neon sign that says,
> "THIS IS SOFTWARE. THIS IS A GAME."
But surely a piece of interactive fiction _is_ software and _is_ a game?
I mean, your works contain a lot of puzzles; thus, they are obviously
games, and people are constantly being reminded that they are games. I
even had to get some pencil and paper to solve the secret code in "Mrs.
Pepper"--that certainly breaks the narrative flow and reminds me that I
am playing a game.
Which, to me, seems to be no problem at all. I _am_ playing a game, so
why would I want to forget about that?
In my work in progress, people will be constantly reminded that they are
playing a game because they will have to device cunning tactics in order
to win difficult fights. Is that a bad thing? I don't think so--if you
don't like playing a tactical game, you shouldn't pick up this work,
just as you shouldn't pick up "Mrs. Pepper" if you don't like playing a
puzzle game.
But do menus somehow do this in a worse way than puzzles or tactical
challenges?
> Second, because (as other posters have mentioned), it turns the alleged
> conversation into a branching tree experience. The best way to proceed
> is to save the game and then start over from the saved game file in
> order to make sure you've explored every branch. This methodology pulls
> you out of the flow of the story.
Okay, this seems to me really important, and I _really_ want to
understand it. Why is "the best way to proceed" saving the game and
exploring every branch by restoring multiple times? It seems to me a
very boring way to proceed, and hence not the best way to proceed. The
best way to proceed seems to me to choose the option that appeals most
to you in the current context, and stick with that.
How am I wrong?
> Third, because some of the branches may disappear for reasons that
> aren't clear. If "ask Mrs. Warburton why she likes darjeeling so much?"
> is available to me now, as a menu item, but not later, after I've asked
> her about the health of her daughter Imelda, I'm pissed off. I would
> much rather have a basic "ask her about darjeeling" command that will
> remain available later, should I choose to use it.
This is true. It is an annoyance that can be minimised and even
eradicated with a good design of menus. (I think, for instance, that I
managed not to fall into this trap with Fate.)
> If you want to view the conversation topics, type the command that makes
> them visible. That's not difficult. The point, to me (and I'm mainly a
> writer) is to preserve the integrity of the narrative flow. A TADS-style
> "You could ask her about her buttercups, or about World War III" seems
> very intrusive to me, unless I've directed the PC to reflect on what
> sort of conversations might be appropriate.
Hm, okay. I'll think about this, and certainly write it down as
empirical evidence. :)
Regards,
Victor
Victor Gijsbers
> You can't please everyone. I recommend you only count the people who
> *do* like something, expecting any game to displease some amount of the
> audience, but not being nonplussed by that.
Thanks for the advice, Bert--I'm sure my game is going to displease
enough people. ;)
But displeasing people is not a _goal_: it should only happen as the
result of people not sharing my vision. Given that vision, I do need to
find a way to implement it that (a) does it justice, (b) pleases as many
people as possible.
Do you yourself like conversation menus?
Regards,
Victor
Victor Gijsbers
> On 10 mar, 16:29, Victor Gijsbers <vic...@lilith.gotdns.org> wrote:
>> What I'm currently doing is implementing a hybrid ASK/TELL-Menu system.
>> The player can bring up topics, to which the character either responds
>> directly, or which bring up a conversation menu with further options.
>> The player can also use a general "talk to" command to start a
>> conversation menu about a predefined (by the author) topic, so that they
>> won't get stuck in trying to think of the right topic.
>>
>> The idea here is that the main conversation menu is the only thing you
>> need to complete the game, but that there is a lot of (optional)
>> interesting and useful information 'hidden' behind the ASK/TELL system.
>>
>> I am not sure that this is a good idea.
>>
>> If the very appearance of a conversation menu is something bad, then
>> this is bad idea. Also, a hybrid system may be confusing; or it might
>> manage to bring the _bad_ aspects of its components together.
>
> I'm not an IF author, but as a player, I'd like to mention that the
> most satisfying conversation system I've seen in an IF game was the
> hybrid ASK/TELL-Menu system in "City of Secrets". I thought it was the
> best of both worlds, with a great feeling of freedom and richness:
> like games with menus, it had flowing conversations instead of
> successions of unrelated questions and their answers; but unlike games
> with only menus, you could also often find unmentioned topics when you
> wanted, which made the conversation feel less railroaded. Also,
> although it's less important: in this game, the menus were fixed in
> their own window at the bottom of the screen, so they didn't have to
> be re-printed at every turn in the main window; I think it looks nicer
> that way.
I am going to check out this game.
(I always avoided it because from the number of help requests on RGIF I
had concluded that it was a very difficult puzzlefest. Which is a genre
I like to ingest in very, very tiny portion with _huge_ amounts of time
between them.)
Regards,
Victor
Victor Gijsbers
> I'd like to suggest a less artistic objection to menu-based
> conversations. When a menu-based conversation drops in, the game is
> essentially saying, "Warning! Warning! You are at a crossroads!"
> Pretty clearly, your choices will make some difference in the
> development of the game. Therefore, my first reaction is always to
> save the game and then mechanically go about trying each option.
Okay, and you are the _third_ person to bring this up, and the third
person I'm going to ask to explain this to me.
Why on earth is your first reaction to try out all the different options?
Interactive narrative _is all about making choices_ is it? So why would
the reaction to making a choice be a kind of nervous "Oh no, I do not
want to make a choice! Let's try out all the different possibilities!"
Why not just make a choice?
Are people afraid of missing something? Do they think I'm cheating them
by breaking my part of the social contract? (My part of the social
contract being, of course, that if I offer you a choice, all options are
viable--none will exclude you from experience a perfectly satisfying
play-through of the game. I don't offer "fake" choices, where only one
of the options is the "right" one.)
Can anybody explain this to me?
Regards,
Victor
Victor Gijsbers
> One of my favorite techniques in _Torment_ was to have identical
> conversation choices, one of which was followed by "(lie)". I used a
> variant of this in _Chicken and Egg_ but I don't know if anyone a)
> noticed, b) found it amusing, or c) discovered how pointless the actual
> choice was.
I think "[Lie]" was _before_ the choice, actually. ;)
I think I did some of that in "The Baron" as well, but I'd have to check
out the source to see where.
Regards,
Victor
Nikos Chantziaras
This style of conversation is supported by Tads 3 out-of-the-box. You
can also have automatic "hints", like:
>ASK GUARD ABOUT KEY
Shaddup," the guard says. "Nobody's going to criticize me for
talking to a prisoner."
(You could criticize the guard, tell him to go to hell, or try
to kill him.)
At this point, and *only* at this point, the player can respond with any
of those:
>criticize him
>tell him to go to hell
>try to kill him
You can find some useful info about this system here:
http://www.tads.org/howto/t3actor.htm
Conrad
> Are people afraid of missing something? Do they think I'm cheating them
> by breaking my part of the social contract? (My part of the social
> contract being, of course, that if I offer you a choice, all options are
> viable--none will exclude you from experience a perfectly satisfying
> play-through of the game. I don't offer "fake" choices, where only one
> of the options is the "right" one.)
>
> Can anybody explain this to me?
I have that reaction to menus too: maybe not as strong as some, but I
have it.
Speaking only for myself, to the extent I do this it's because I'm
responding the same way I might to a test. And my goal is not exactly
to have a conversation the way I might as a human speaking to a human,
but to understand and appreciate the logic of the game.
So, the ones that really get me are those where I'm offered a choice
like this:
"So what do you think? Do you think the Glactic Neewink will buy our
spaceship?"
> "Yeah, sure." <
"Not a chance."
"It depends... would you come down a little on the price?"
"Depends... can you come down on the price a little?"
-- I don't understand what these choices *mean*. For one thing, why
am I being offered this choice at all? Am I being tested on my
loyalty? My honesty? My insight into the thinkings of the Galactic
Neewink? And what on Earth is the difference between the last two?
If it's a choice between killing someone and letting them live, I can
handle that: but if it's important enough for me to be given a
choice, but yet I don't understand what the significance of the choice
is, then that's a Kafkaesque bind, and it reminds me of the old CYOA-
style puzzle:
You are at a fork in the road. Do you go left or right?
[Go Left] -> You fall into a pit trap and DIE HORRIBLY! Return to
page one and start your adventure again. (?!)
[Go Right] -> You wander into a small villiage where the people
believe their next visitor is fated to become king! You're rich
beyond your wildest dreams! -the end-
--Now you may not be doing that, but if I can't see the reason and
thinking for the choice, the only datum I have is that something is
happening which is important enough for me to be given a choice. So,
for me to understand the nature of the choice I'm making, I have to
try out both.
In your phrasing, it's not exactly that I'm afraid of missing
something, so much as I want to understand the logic of the game and
the nature of the choice I'm being offered.
Those times when I'm offered a choice like:
"Gee, you're such a good friend," Bob says. "I'm so glad I confided
in you."
"No problem," you say: "You can always trust me."
> Betray Bob to the police. <
Betray Bob to the mafia.
Keep it quiet.
-- I have no inclination to lawnmower, at least not immediately.
Maybe later, on replay, I'll want to explore the options. But that's
not lawnmowering, that's fun.
Conrad.
Jeff Nyman
news:47daaacc$0$14359$e4fe...@news.xs4all.nl...
> Smoov...@gmail.com wrote:
>
> We shouldn't be forgetting that Jeff's sample is not just biased in terms
> of all of them being more or less new to interactive fiction; it is also
> biased in that all of them are writers.
Not *all* of them; but that was mostly the makeup of the group. Here's how I
originally described it:
"[The group] included writers (who had at least short stories or one novel
published), writer wanna-be's [meaning, their primary aim was to write
stories, not games] and game writer wanna-be's [meaning, their primary aim
was to write games, not stories]."
So there was a certain percentage that had not written novels (or anything
of the static fiction variety) nor had any plans to, but did want to be game
writers (or already were game writers, but not in the textual IF medium).
This being said, though, you are correct: there is an inherent selection
effect present based on the biases and preconceptions I was hoping to avoid
or at least mitigate and it is definitely important that this be kept in
mind. The "control group" was the "game writer wanna-be's." My only
stipulation was that they had not been exposed to textual IF in terms or
programming it and, ideally, hadn't played it in quite some time (if ever).
This did inherently present a bias in that I didn't have a group that was
already familiar with and comfortable with the mechanics of textual IF. That
probably seems like a glaring omission but it was one I felt somewhat safe
in making, knowing that I had observed this community for quite a few years.
Plus the "control group" in this case was more in the sense of simply
wanting to create a game, rather than necessarily worring about "telling a
story" in the strict sense that a novelist (or novelist wanna-be) would.
Regarding the other game experience, I didn't filter out too much of exactly
what other games they had played but given that they had an active interest
in writing computer games, there is some presumption of experience across
the game formats. We did also consider graphical adventure games as well as
first-person shooters so we did do some comparisons with other games.
Interestingly, CRPGs didn't come up that much because the people in question
hadn't really played them all that much, at least recently. I did have a few
fans of "Starflight" and "Wasteland." The comparisons with other forms of
game was probably not as rigorous as it could have been, to be honest.
- Jeff
Jeff Nyman
> Jeff Nyman wrote:
>
> I don't agree with _all_ the detail of your story; especially not with the
> part where you seem to argue that the aim of fiction is to faithfully
> recreate how certain people act / how human nature works.
I agree and if I said that I was being too simplistic. What I mean is
recreating how "certain *possible* people act." After all, the protagonist
could be a robot or an alien. But usually they have some features that allow
us to relate to their motivations or desires in some way. (Even Greg Egan,
an author who went out of his way to present vastly different species of
evolved human, had his characters act in ways that, while somewhat alien,
were still recognizable.) Further, not just how *human nature* works but how
those *possible* people will react, given the nature we've established in
the story (where we might not even be talking about "people" at all).
Sometimes you read about "alien nature" or "robot nature" or whatever else.
But it usually translates back to human nature ultimately. "Animal Farm" was
effective because while it was animals doing the thinking, talking and
feeling, those were thoughts and conversations and feelings that humans (the
only readership possible for us) could relate to.
If someone shoved me in a corner and demanded I pick an "aim" to fiction, I
would say it's to please the author and to please the reader. Doing the
former will lead to the latter, albeit a certain subset of the reading
public. I've come to believe that most writers do write for themselves, but
they touch upon that storytelling nerve that exists in pretty much every
member of the human species and so they end up writing for others as well.
The point there is simply that what pleases some readers, bores others. For
example, you say this: "...not just any 'could', but a 'could' that grants
me a glimpse--however vague and transitory--of new and unknown heights of
compassion, love, beauty, tenderness, humanity."
I would agree that could be some authors' aims when they write their
fiction. To those readers who that appeals to, they will share the aim of
the writer and probably enjoy the work. To others who don't need or want
that level of "new and unknown heights", that aspect may not matter as much
or at all. Either way, I don't think that element can be the "aim" of
fiction though and the only reason I say that is because part of what you
are describing is inherently subjective and often those elements you mention
can be found even when that was not the authors' intent or aim.
As you say, a subject for a different thread -- but, just to reinforce your
earlier point, I might have been coming off a bit more simplistic than I
intended.
> Yes, but either I'll have to implement a huge number of 'remind'
> responses, or I have to contextually tell the player that she can now use
> the verb 'remind'. Your suggestions as to how to do the latter are
> interesting, but call for a very specific style of writing--almost
> necessarily first person, very self-conscious and reflective; a style of
> writing that I can see working for a short noir detective story, perhaps,
> but not for many other projects.
On the first-person viewpoint, I noticed a preference for this with my
group. It was found to be the "most effective" style for writing textual IF.
Mind, I'm not saying it *is* the most effective style. I'm just talking
about the perceptions that my group had.
A bit of a side-note but kind of on track -- one of the few things that John
Crossan said that I agree with is that "memory is as much or more creative
reconstruction as accurate recollection." This was part of how we modeled
our protagonists who were telling the story in first-person in the past
tense. Thus the aspect of "memory" is almost inherent. As the story was told
by the protagonist, you as the player went along with it, living through the
creative reconsturction, having to figure out not just what was going on but
on how the protagonist conceived of the events and so forth.
We used many first-person novels for some our work. For example, Stephen
King's "From a Buick 8", Richard Morgan's "Altered Carbon" and "Broken
Angels", Jack McDevitt's "A Talent for War", Kage Baker's "In the Garden of
Iden", John Shirley's "Demons", Robert Wilson's "The Chronoliths", Maxine
McArthur's "Time Future" and "Time Past", Michael Laimo's "Deep in the
Darkness" as well as many others.
The point there being whole novels (not all of them in the "noir detective"
style) are written in the first-person viewpoint and are compelling reads
for that, some of the above books being quite popular.
> This latter 'problem' is something that Jim Aikin also brings up. I am not
> 100% sure I understand it, though. There seems to be a tension. On the one
> hand, the player (presumably) wants to steer the story one way rather than
> another. (Otherwise, why play interactive narratives?) But on the other
> hand, if I understand you and Jim right, the player wants to see _all the
> text_, or _everything that happens_, or _all the possibilities of the
> game_.
I can't speak for Jim, but speaking for me: when I read a novel, I want to
read the whole novel and thus I want all the text and all the possibilities
that the author put to form. I want to try to see the author's vision, if
there was such, and "get" what the author wanted me to get, even if that was
only a couple of hours of entertainment. In textual IF, does that mean "all"
the text or "all" the possibilities? It's a good question. I guess I don't
know.
Certainly there's the possibility for allowing a richer experience the next
time through if there is "more" to the story but if there is that much more
and if it is important to the story, why was it not included in the first
place? In other words, if there are other things going on but it wasn't
necessary for the player to get to them or see them to "get" the value of
the story, how valuable were those things?
It's a large question, I think. Certainly novelists leave out tons of
details -- but, to that novelist, those details are not the important ones.
The important ones are the ones they *are* showing you. But that's static
fiction. Textual IF presumably allows for more than that. I just don't know.
Personally I very rarely go back and play textual IF, at least any time
soon, similar to how I'll often wait quite a bit of time before coming back
to a novel. And when I come back to a novel it's the inherent familiarity
that actually helps me look for nuance. If the novel could change the next
time through, I'm not sure I would like that as much.
> But these two are simply incompatible (unless you are willing to do
> replays). So what _does_ the player want? Does he want choices with
> consequences, or does he want to be able to read everything? I cannot give
> him both, as a matter of logic.
I agree. I think the problem is that "the player" is not a monolithic
creature. There are as many types of players as there are types of readers.
That's what I meant earlier about the aim of the novelist: please themselves
and they'll probably please some segment of readers. Who knows? It might
even be a very large segment. If the writer's intent was to provide a
situation wherein the player/reader could see all of the work, then that
might dictate some of the choices made in implementation. If the writer's
intent was to make a dynamic work that would change -- and change
substantively based on actions the player/reader took -- then that too might
dictate some of the choices in implementation. Or perhaps the implementation
choices remain the same, but how the story is presented differs. (For
example, perhaps it's made very clear that the choices you take will lead to
a drastically different story.)
Your question is a good one and worth discussion and I think it goes just
beyond conversation systems with NPCs; it really speaks to the entire
narrative of the story, of which conversation may be just one part.
- Jeff
Jeff Nyman
Sorry to jump in on Jim's response here, but a few things here go with my
points earlier. I'll make this brief, I promise.
> But surely a piece of interactive fiction _is_ software and _is_ a game?
>
> I mean, your works contain a lot of puzzles; thus, they are obviously
> games, and people are constantly being reminded that they are games. I
> even had to get some pencil and paper to solve the secret code in "Mrs.
> Pepper"--that certainly breaks the narrative flow and reminds me that I am
> playing a game.
That's true -- but it's how obvious it does it, too, right? I mean a novel
is just printed words on paper. It's probably not even something that
happened to a real person. But we can still get so entranced with the story
that we forget that reality for a bit. (Which I would argue is hard to do
when you're reading a Choose-Your-Own-Adventure book, for example. The very
mechanics of having to care about the numbers -- *while reading* -- sort of
destroys any illusion.)
> Which, to me, seems to be no problem at all. I _am_ playing a game, so why
> would I want to forget about that?
As you said in response to me, "[something that] grants me a
glimpse--however vague and transitory--of new and unknown heights of
compassion, love, beauty, tenderness, humanity." I would think that when
this happens, you're not necessarily thinking "This is just a book. This
never happened to anyone. It's not even real." For that moment -- however
vague and transitory -- you are swept up in thoughts based on what you read
and it had a major effect on you. It would probably be somewhat upsetting if
that effect was nullified by some mechanic like "If you want heights of
compassion, skip to chapter 27. If you want heights of love and tenderness,
skip to chapter 30. If you want to see how the beauty ties in with humanity,
skip to chapter 31."
That's contrived, I realize, but the point is that you wouldn't necessarily
want the mechanics of the book to intrude.
I agree -- people probably don't really and truly "forget" that they're
playing a game any more than they really and truly "forget" that their
reading a book. But, for a time, that aspect is kept at bay to the extent
that the medium allows.
But --- all of this suggests that people look at textual IF as game first,
story second. Again, I can't speak for Jim, but part of what I've been
looking at is the possibilities of story first, game second. That
expectation may be very different for certain people, so ...
> But do menus somehow do this in a worse way than puzzles or tactical
> challenges?
I think this is where the expectations come in. What audience are you trying
to reach? If someone reading a novel finds halfway in that it becomes a
choose-your-own-adventure, their expectations might be upset enough that
they don't like it. But if your audience was choose-your-own-adventure
readers, then you probably would have structured your story that way to
begin with. And if that wasn't your audience, you probably wouldn't do that.
- Jeff
Emily Short
I saw a lot of this reaction when I first released Galatea: many
people complained that they didn't like not being able to see all the
endings, or not being *sure* they'd seen all the endings, and this got
a fair amount of discussion here and also led to a bunch of email from
people who wanted a list, or a complete set of walkthroughs to every
ending, or something like that.
I think there are a couple of things at work here. One is a puzzle-
focus that has somewhat diminished in the community since that time,
which says that since seeing every possible outcome has the form of a
puzzle, it constitutes a challenge that many players want to resolve,
even though the game was not originally designed to present that
particular challenge and may be very ill-designed to support it.
The other is perhaps a little less at odds with the goal of
interactive narrative, though. The thinking goes a bit like this: the
author has provided a branching story here. Therefore, he had
something he wanted to convey that could not be conveyed by showing me
just one narrative path (either in a short story or with the linear-IF
techniques we see in lots of games). So part of the point of the game,
perhaps, is to explore and compare these different branches, and use
that comparison to elicit a deeper meaning from the game. What do the
choices all have in common? Are there certain kinds of outcome the
author has made inescapable?
I don't think that's entirely misguided.
There's a kind of middle-of-the-road kind of play, which I adopted
with Fate and which I think is probably most productive with games
with lots of endings, namely to play a few times until I felt I had a
feel for what the game was doing and had also obtained the endings
that I specifically was most interested in. That might mean trying for
a "winning" ending -- ie, an outcome I felt was positive enough to
live with -- but it might also involve pushing the bounds of the
narrative in other ways by looking for the worst possible ending, or
trying to see what happens if the PC sticks to some particular
principle (such as never actually killing another living creature),
or... well, you get the idea.
Ron Newcomb
> Jim Aikin wrote:
> > Victor Gijsbers wrote:
> > "THIS IS SOFTWARE. THIS IS A GAME."
> why would I want to forget about that?
I agree with Jim here. When you read a book, you're not constantly
reminded that it's just a book, that the characters in it aren't real
people, that their problems aren't real problems. Immersion requires
that the medium disappear.
> In my work in progress, people will be constantly reminded that they are
> playing a game because they will have to device cunning tactics in order
> to win difficult fights. Is that a bad thing?
Soldiers & generals solve such problems in real life a lot. Just
because the players has to figure something out doesn't mean it
automatically breaks immersion. It depends upon how the author
presents it, of course.
> > That's not difficult. The point, to me (and I'm mainly a
> > writer) is to preserve the integrity of the narrative flow.
I like how Pytho's Mask does menus: in a status bar at the -bottom-
of the screen. The menu appears essentially in the same place as
normally printed menus -- near the command prompt -- but it doesn't
actually insert itself into the narrative's text flow; once you've
chosen, all the choices disappear. If you're going to do menus, I
nominate this as the right way to do it.
-Ron
Jim Aikin
>
> Okay, and you are the _third_ person to bring this up, and the third
> person I'm going to ask to explain this to me.
>
> Why on earth is your first reaction to try out all the different options?
Because one of them may be significant. Conversations (in either IF or
static, conventional F) should not be just window-dressing. They should
move the story forward -- even if it's only by giving the reader some
insight into the character. In IF, that's not their only function. It's
entirely possible that you'll need to ask the archbishop about the
beggar's matchbox in order to get a clue that will allow you to solve
the matchbox puzzle. So it becomes imperative to try all of the options
that the author provides.
> Interactive narrative _is all about making choices_ is it? So why would
> the reaction to making a choice be a kind of nervous "Oh no, I do not
> want to make a choice! Let's try out all the different possibilities!"
Because some choices may be sub-optimal. There are many games in which
you're advised to save the game before trying an action (such as
crossing the rickety bridge) that may lead to a situation that you don't
like. Why should conversation be exempt from that?
--JA
Bert Byfield
>> You can't please everyone. I recommend you only count the people
>> who *do* like something, expecting any game to displease some
>> amount of the audience, but not being nonplussed by that.
> But displeasing people is not a _goal_: it should only happen as
> the result of people not sharing my vision. Given that vision, I
> do need to find a way to implement it that (a) does it justice,
> (b) pleases as many people as possible.
> Do you yourself like conversation menus? Regards, Victor
My WIP has menus. Generally they involve entering an initial letter
rather than a number. I played a lot of *Wasteland* which used a lot of
menus, and I never objected. *Fallout* and *Fallout 2* used
conversation menus: e.g. 1) "Good morning" 2) "Eat lead, sucker!" 3)
"What time have you got?" and so on, and I managed to stand it. I think
a user knows he or she is plaing a game/story on a computer, and that
it isn't too jarring to "remind" him or her of that fact. If they
*don't* know, how can they find the keyboard and spell out stuff to
enter? If I expected to make a lot of money, I would design for the
common denominator user, but as it is, I design for a game/story that
*I* want to play. ;-)
Bert Byfield
Emily Short
> I agree with Jim here. When you read a book, you're not constantly
> reminded that it's just a book, that the characters in it aren't real
> people, that their problems aren't real problems. Immersion requires
> that the medium disappear.
When I read a book, it's still a physical object in my hands; it still
requires that I turn the pages. When I'm at the movies, there is
demonstrably a screen, with edges and curtains and occasionally
flickery specks of dust. At plays or musicals, I can see the sides of
the set, can tell that the backdrop is fake, etc. None of that stuff
bothers me.
The medium disappearing, I think, has most to do with the player's
ability to learn the medium's limitations and incorporate them into
his expectations -- to learn to filter them out, in other words.
That's why it's more problematic when brand new UI features appear in
the middle of a game; the player has his filter in place, but now new
aspects of the medium draw his attention.
I generally find that I don't notice the conversation methodology very
much if (a) it's one I'm already familiar with from a bunch of other
games (for some games I don't even remember well enough to be able to
say now whether they used menus or ASK/TELL); (b) the game is itself
long enough that the mechanism starts to feel natural.
There's obviously something to be lost by presenting new IF players
with UI features they find off-putting, but on the whole I suspect
that the right conversation method for a given game is going to be the
one that most supports the game's other goals (linear, directed
narrative, or exploratory? giving the PC a definite voice, or allowing
him to craft his own?) rather than the one that best imitates the
experience of reading a book.
Jim Aikin
>
> I don't agree with _all_ the detail of your story; especially not with
> the part where you seem to argue that the aim of fiction is to
> faithfully recreate how certain people act / how human nature works. I
> believe that fiction is the realm of possibility, not of necessity; of
> redemption, not of knowledge. The psychologists can tell me how people
> _do_ react in certain situations. I want the novelist to tell me how
> people _could_ react in those situations, and not just any 'could', but
> a 'could' that grants me a glimpse--however vague and transitory--of new
> and unknown heights of compassion, love, beauty, tenderness, humanity.
Unknown to whom? Clearly not to the writer. Unknown to you as the
reader, perhaps.
Forgive me; I'm very tired. I just played (cello) in the pit in the
opening night of 'Carmen' with the local opera company, and I'm
exhausted. So I'm just going to ramble here for a minute and then go to bed.
I don't know what the aim of fiction is (or should be). I write
conventional fiction, and I've sold a few stories here and there, but I
think I'm too old and cynical to have anything resembling an "aim" as an
overarching goal of writing. What I have is much more practical: I want
to write a story that is emotionally meaningful. To me, at least, and
hopefully to a few readers.
Whether any of my stories will ever achieve unknown heights of anything
is (a) highly unlikely, and (b) not for me to say. But I can tell you
that I want to -- no, make that MUST -- understand the emotional impact
of the events in the story on the main character(s). The impact must be
significant to them! I must also try to make that impact clear to the
reader, but without bludgeoning them with it.
I think I could make a case that it's difficult for IF as a medium to
present characters who are in the grip of strong emotion, simply because
of the fact that the player directs, to some extent, the path the
narrative takes as it meanders through "story space."
I could also make a case that if you've got a gripping story to tell,
the mechanics of the conversation system don't matter a whit. Ask/tell
will do you just fine. Conversely, if you _don't_ have a gripping story
to tell, all the software gewgaws on the planet ain't going to cover up
the problem.
Here's another thing to think about: The PC needn't be the person to
whom the emotionally meaningful events or insights occur. Things can,
and perhaps should, happen to minor characters along the way.
Forgive me for quoting one of my own games, but in "Lydia's Heart," you
(Diane Lindsay) can ask Tyrone (a middle-aged, rural Southern black man
who is somewhat sympathetic to your plight) about yourself. Not about
himself ... but here's the result:
It doesn't seem too likely Tyrone will know anything about why you've
been brought here, but you decide to take a chance and ask him anyway.
He wipes his hands on a greasy rag and takes a long time about replying.
"I remember when I was little," he says at last. "Seemed like a single
afternoon would stretch out forever. Seemed like it would never end.
Course, it always did. My daddy'd come home drunk, and he'd whup me 'til
his arm got tired. But before he come home I could pretend I was just
floatin' on this ocean of afternoon that would just go on and on and on,
rockin' on the waves and all. Sometimes I pretended so good I pretty
near believed it."
Tyrone doesn't seem to have answered your question at all. But Tyrone
knows you (Diane) are going to die before the afternoon is over, and
while he feels bad about it, he's not going to save you. So he (and
David too) are caught in a moral dilemma to which they have no answers.
Most of the other characters are just bad guys, but in the course of
conversing with Tyrone and David, you may find some wisp of emotion
sneaking into the frame. It's not the PC's emotion, however. It's
something that is only glimpsed.
As an author, I'm not in a position to force the player to have those
conversations with the characters. So it's entirely possible that most
players will never see what's going on. If I'm going to write
interactive fiction, I need to accept that -- because I feel it would
rip the heart out of the narrative (so to speak) to insert a line that
says, "You could ask David about himself." To do that would be to remind
the player, "Hey, you're using software, and there are options built
into the software, and here's a good one, you really ought to check it
out." The moment I do that, THE PLAYER IS NO LONGER SEEING THE WORLD
THROUGH DIANE'S EYES. The player is pulled out of the fictional world
and is back sitting in a chair staring at a computer screen.
Not gonna go there.
--JA
Blank
Actually I think lack of exposure to current IF mechanics and
conventions /adds/ value to the group's opinions. Working with IF
inevitably means that the author is "trained" to construct IF in a
certain way: not just peer-pressure and the expectations of the likely
audience, but the library itself and the tools we use shape our
thinking. This was brought home to me recently in another thread where
we were talking about navigation systems and comparing IF and MUDs. I
hadn't realised the extent to which compass-navigation had locked us
into thinking in terms of a grid-square world.
Similarly the messages from the library are all in 2nd person present,
and I remember when I first started tinkering with IF how difficult I
found it just to write decent text because all my fiction-writing habits
used 3rd person past. As a result of this thread I started noodling
round with a little test piece of IF in 3rd and I'm finding it very
difficult to "change gear": not only do I keep dropping into the
standard IF tense, but I'm having trouble putting the plot together too.
I've found these threads fascinating and stimulating, and I'd like to
add my voice to the chorus of thanks - you've really done something
useful here, I think. I've certainly found your group of "outsiders" a
very useful mirror.
--jz
Blank
Perhaps the function of the story is in the presence of things, both of
which must be present but cannot be seen at the same time. Imagine a
work which is set up as a good old adventure type, with the player as
hero, saving the world from the forces of darkness. The twist is that
the player gets to choose which character to be for the whole of the
story, and no matter which one he picks, that character is presented as
the hero, and the opponent as the evil villain.
Play the game once, and you've got a standard adventure. Play it twice,
and you've got the point.
--jz
Blank
I think it's just that people hate stuckness, and in order to avoid it,
they're willing to do boring spadework. By showing the exact limits of a
character's knowlege, a conversation tree makes it possible for the
player to /know/ they've exhausted all the possibilities.
Of course if you use a hybrid system, with Ask/Tell responses sometimes
bringing up short menus of extra choices then you no longer have a
single, unified conversation tree, (a conversation shrubbery, perhaps
:o)) and so the player loses the seductive guarantee of a complete
examination of the conversation space.
I don't hate conversation menus provided that (a)they're not modal - I
hate the feeling of being "trapped" and /only/ able to talk, and (b)they
don't completely blank out the whole game - at minimum I still want
access to scrollback, ideally to the room description, inventory etc.
too. This is partly pragmatic: if the npc hands me a ring during the
course of a conversation, I want to be able to look at it and then ask
about the things I see. It's also psychological: I want to still feel
'connected' to the game world, and single-mode full-screen menus make me
feel I've been shut away in some sort of little closet.
Yes, conversation menus do make it obvious that I have only a few,
limited choices instead of the *illusion* of limitless freedom. But I'm
willing to accept that in return for (a)less stuckness: I won't be
floundering around trying to hit teeny little conversation topics in the
dark, and (b)as the player I get to use a richer style of conversation
than the "information token exchange" which reduces npcs to databases.
--jz
Victor Gijsbers
>
> "Gee, you're such a good friend," Bob says. "I'm so glad I confided
> in you."
>
> "No problem," you say: "You can always trust me."
>
>> Betray Bob to the police. <
> Betray Bob to the mafia.
> Keep it quiet.
>
>
> -- I have no inclination to lawnmower, at least not immediately.
> Maybe later, on replay, I'll want to explore the options. But that's
> not lawnmowering, that's fun.
You are very right, and make an excellent point about how important it
is to make clear to your readers/players (I would very much like a
single word that has both these meanings, but neither "pleaders" nor
"rayers" sounds very good) what kind of choices you are offering them.
Are it adventure book style "your life depends on this but you cannot
know what the right choice is, ha ha ha!" choices?
Are it sophisticated but cruel IF "your life depends on this so you'd
better gather all the clues and find out what the right choice is before
you try" choices?
Are it narrativistic "you've got a choice here as to which direction the
story moves in, and all the possible choices are valid and will lead to
something interesting" choices?
I agree with you that the first possibility is very bad indeed; while
the second and the third, though both interesting, should probably not
come together in a single game, and any game which uses one them should
make very, _very_ clear which one. (And of course there are other and
in-between possibilities.)
Now both my published works involve choices of the latter kind; and I
was thinking of that kind of choice when I asked my question. But I
didn't make this clear; and yet, making this clear (not just now, but
also in the game itself) is essential.
So might it be the case that conversation menus (and other types of IF
interface where it is clear that you are making an irrevocable choice)
are less of a problem if it is clear that there is no _right_ choice, no
choice you have to make in order to find an essential clue, or get to
the rest of the story, or understand what is going on?
Regards,
Victor
Victor Gijsbers
This is a very useful post, thank you.
I don't know if I'd ever really thought about the difference between the
following two kinds of design:
1. A piece where the point of the game is only revealed after
exploration of different branches of the narrative.
2. A piece where the point of the game, though tied up with the
possibility or necessity of making choices, is revealed on a single
play-through.
(I'm using the term 'branch' here even though it's a bit unlucky. The
medium of interactive fiction does not actually facilitate writing
treelike games.)
Of course if you want to review the game, understand it, write a
critical essay about it, exploring the branches will always be
necessary--but it is not always necessary if you just want to _play_ the
game.
Thus, I suppose that games like "Galatea" and "Whom the Telling Changed"
belongs to the second class of games: additional play-throughs reveal
more about the game, of course, but they do not reveal a 'point' that
was somehow not yet clear when I played it for the first time.
"Photopia" would perhaps belong to the first class: if the point it
wants to make is about the extent of your power over the narrative,
well, of course that extent is only made (really) clear after you try to
do different things.
I like your middle-of-the-road kind of play. "Fate" and "The Baron" were
designed with that in mind, I think, though I'm not 100% sure I had it
in mind _consciously/in an articulated fashion_.
I do think however that one reason you can adopt middle-of-the-road play
in those pieces, or in "Galatea" or "Whom the Telling Changed", is that
they are quite short. Once you've played through them once, you can
probably play through them again in 10 to 20 minutes.
Longer games, though they _can_ be replayed, cannot assume as easily
that they _will_ be replayed. So I think the author of a longer game
will have to make sure that her game yields its message (and doesn't
leave too much unsaid) on a single play-through. (Though slightly more
divergence at the end of the game might be more useful, because its
easier to replay the end.)
Regards,
Victor
Victor Gijsbers
Okay, so suppose I promise at the outset of the game that NO choices
within conversations (or indeed, no choices that have to do with the
story) are sub-optimal. They are different, yes, and they will lead to a
(slightly) different story--but none is sub-optimal. You can choose
whichever one you like, and you won't be missing a clue, or a central
insight into a character.
Suppose I do that--would that solve the problem?
Because I am entirely willing to make that promise, and, in fact, have
never written a conversation menu without working on the assumption that
this promise was already implicitly in place.
Regards,
Victor
Jim Aikin
> Jim Aikin wrote:
>>
>> you're advised to save the game before trying an action (such as
>> crossing the rickety bridge) that may lead to a situation that you
>> don't like. Why should conversation be exempt from that?
>
> Okay, so suppose I promise at the outset of the game that NO choices
> within conversations (or indeed, no choices that have to do with the
> story) are sub-optimal. They are different, yes, and they will lead to a
> (slightly) different story--but none is sub-optimal. You can choose
> whichever one you like, and you won't be missing a clue, or a central
> insight into a character.
>
> Suppose I do that--would that solve the problem?
Sure, but it would also pretty much guarantee that your story would be
dull and trivial.
If a choice is not MEANINGFUL in terms of the movement of the story,
then the interactivity is just a pointless waste of time. A meaningful
choice doesn't have to mean helping you find a key to a locked door. It
can also include things like revelation of character. But if you're
promising us that none of the conversational branches will be
meaningful, why should the player engage in conversation with your NPCs
at all?
The whole point of conversing with an NPC is to move the story forward
by making meaningful discoveries.
--JA
Jim Aikin
>
> As an author, I'm not in a position to force the player to have those
> conversations with the characters. So it's entirely possible that most
> players will never see what's going on. If I'm going to write
> interactive fiction, I need to accept that -- because I feel it would
> rip the heart out of the narrative (so to speak) to insert a line that
> says, "You could ask David about himself." To do that would be to remind
> the player, "Hey, you're using software, and there are options built
> into the software, and here's a good one, you really ought to check it
> out." The moment I do that, THE PLAYER IS NO LONGER SEEING THE WORLD
> THROUGH DIANE'S EYES. The player is pulled out of the fictional world
> and is back sitting in a chair staring at a computer screen.
>
> Not gonna go there.
As a footnote to the foregoing:
However, a system that implements 'think about X' could provide much or
all of the functionality of the intrusive "(You could ask the waiter if
he has ever tried the soup)" prompts. This would (a) leave the basic
presentation of the text seamless and (b) allow the conversation nudging
to be done **within the POV of the PC** rather than as a paid political
message from the author.
>give soup to waiter
The waiter sneers. "There is nothing wrong with our soup! You ordered
it. You eat it!"
>think about waiter
As you gaze in disbelief at the surly waiter, it occurs to you that you
might ask him whether he has ever tried the soup himself. You might also
ask him if you can speak to the manager. (Dumping the soup on the floor
... also a possibility.)
'think about soup' would, at that moment, produce exactly the same output.
--JA
Adam Thornton
Jim Aikin <midig...@sbcglobal.net> wrote:
>If a choice is not MEANINGFUL in terms of the movement of the story,
>then the interactivity is just a pointless waste of time. A meaningful
>choice doesn't have to mean helping you find a key to a locked door. It
>can also include things like revelation of character. But if you're
>promising us that none of the conversational branches will be
>meaningful, why should the player engage in conversation with your NPCs
>at all?
>
>The whole point of conversing with an NPC is to move the story forward
>by making meaningful discoveries.
I think the disconnection between your position and Victor's is that you
are assuming there is one direction in which a story can proceed
forward, and that he is claiming there can be several. A conversation
menu can send the game down different paths, but those paths are not to
be construed in terms of "winning" or "losing" paths or indeed "better"
or "worse".
Adam
Jeff Nyman
"Adam Thornton" <ad...@fsf.net> wrote in message
news:9hdta5-...@quicksilver.fsf.net...
> I think the disconnection between your position and Victor's is that you
> are assuming there is one direction in which a story can proceed
> forward, and that he is claiming there can be several.
Just to chime in a bit ...
... herein lies the disintinction I was making between "game first, story
second" and "story first, game second" that I had reported with my group.
The issue was seen as being one of how cohesive a narrative can be if there
are several ways in which the story can proceed forward. The contention was:
not very cohesive. (That's just the contention; not to be mistaken for me
claiming it's true.)
** The idea is that a story that tried to allow several ways to proceed --
and still keep a narrative track -- was a game first, story second. Further,
it was argued to be (usually) a poor story but perhaps a good game.
** The idea that a story tried to allow one way to proceed -- keeping the
narrative track -- was a story first, game second. That still didn't make it
a good story, of course, since that depends on the author. Further, the
extent to which it made a good game was by the way in which the
player/reader was allowed to explore the narrative, while still being
constrained by it.
One argument is that this ability to tell several stories (or several
aspects of the same story) is something textual IF has over static fiction.
For what it's worth, I found this argument did very little good with my
group and that included the game writers, at least to an extent. Their
argument was that what textual IF has over static fiction is the ability to
make the one story something that can be explored in more depth while still
keeping it as a cohesive story.
As you can imagine a lot of this also got into the nature of puzzles, which
is why I mentioned earlier about the idea of characters in the game being
puzzles in and of themselves. But the wider conversation was about how to
incorporate the notion of "puzzles" (which the group preferred to call
"complications") that were situationally relevant and useful to keeping the
story going. In other words, you, as the player, could clearly see why
solving this complication was necessary and important -- not to completing
the game, but to unfolding more of the story.
As just one example relevant to this thread, menus were seen as a "puzzle"
that was put in place just to indicate to the player where the road could
diverge. How much it could diverge often wasn't obvious, somtimes even after
taking a certain path. (Only replay would answer that question, which was
found as one of the shortcomings of menus since it presupposed replay.)
Menus were seen as a 'weak' puzzle because, of course, they clearly weren't
intended as a puzzle, per se, but rather as just a way to force the player
to make a choice, little different from the Choose-Your-Own-Adventure "turn
to page X" choices.
So I think this topic gets into wider aspects than just conversation systems
since menus could allow you to provide gating criteria for various actions
the player takes, not just conversations. So the story could take off in
different directions based on criteria other than conversation, which
probably brings up a good question:
Wold people like to use menus in this sense? (Would people mind if the story
presented some 'nodes' where what you, as the player did next, came down to
a menu choice?)
If yes, how much of that would players be willing to tolerate?
If no, why is it tolerated in conversation systems?
The goal for my group was to eventually focus on textual IF mechanisms
(menus being just one of those) to see if those mechanisms were designed ...
(1) ... as a reaction against other mechanisms (like ASK/TELL, in the case
of menus)
(2) ... to provide a way to better tell the story
(3) ... to provide a way to not tell a story (thus more create a game)
That last one isn't meant to imply that textual IF authors are consciously
sitting there trying to *not* tell a story, but rather that the mechanisms
of the format had become so ingrained that the mechanisms were simply used,
*regardless* of whether or not they were effective for telling a story.
Notice that this contrasts with (1) a bit wherein, in that case, it was felt
textual IF authors probably *did* have the story focus first but felt that
existing mechanisms didn't allow it to be done.
To spare anyone suspense about the outcome of this set of discussions,
although I've indicated it elsewhere, the thought was that the narrative
possibilities of including contextually relevant and situationally useful
material *as part of the narrative* was simply not done by the majority of
textual IF authors. Just as there were conventions established around the
mechanisms over the history of IF, so too were conventions established about
how to use the mechanisms or whether to use them at all.
- Jeff
Ron Newcomb
> So might it be the case that conversation menus (and other types of IF
> interface where it is clear that you are making an irrevocable choice)
> are less of a problem if it is clear that there is no _right_ choice, no
> choice you have to make in order to find an essential clue, or get to
> the rest of the story, or understand what is going on?
Gameplay design (according to Chris Crawford) states that, given a few
discrete choices, it -must- be a tradeoff. Reasoning: if only one
option is the "right" one, then there isn't a true decision to be
made; it's a false choice, a no-brainer. (Or a disguised no-brainer
wherein you must lawnmower to find the right option.)
There's a similar rule against presenting trivial choices.
-R
Jim Aikin
>
> I think the disconnection between your position and Victor's is that you
> are assuming there is one direction in which a story can proceed
> forward, and that he is claiming there can be several. A conversation
> menu can send the game down different paths, but those paths are not to
> be construed in terms of "winning" or "losing" paths or indeed "better"
> or "worse".
There are usually several directions that the story can develop,
including (as a general rule), ***You have died***. Most of them would
tend to be sub-optimal at best, ranging downward from there to highly
undesirable (as in, ***You have died***).
In conventional fiction, there is almost always a specific way that the
story needs to end, in order to be emotionally satisfying -- or, if
unsatisfying, for the story to "ring true" in terms of our understanding
that that's what those people would do in that situation. Offering the
reader a branch in which Odysseus arrives home to find that Penelope has
married one of her suitors ... it would be a very sub-optimal ending.
It's hard enough to come up with ONE emotionally satisfying ending.
Trying to figure out three or four that are ALL emotionally satisfying?
I lack the fortitude to attempt it.
--JA
Conrad
> Gameplay design (according to Chris Crawford) states that, given a few
> discrete choices, it -must- be a tradeoff. Reasoning: if only one
> option is the "right" one, then there isn't a true decision to be
> made; it's a false choice, a no-brainer. (Or a disguised no-brainer
> wherein you must lawnmower to find the right option.)
That's true if I'm out to "win" -- to get the most treasure, to solve
the dungeon, etc. -- but what if I'm out to define what "winning"
means?
I'm not BSing here: it's an important distinction. If I'm called on
to make a decision that *shapes* the character, that *defines* my
priorities, then yes it's a trade-off in a sense, since there's an
opportunity cost of the other things that aren't my priorities: but
that's not the same as saying that there is only one objectively right
answer.
> There's a similar rule against presenting trivial choices.
Sure.
Conrad.
Adam Thornton
Jim Aikin <midig...@sbcglobal.net> wrote:
>It's hard enough to come up with ONE emotionally satisfying ending.
>Trying to figure out three or four that are ALL emotionally satisfying?
>I lack the fortitude to attempt it.
I think that's precisely what Victor is attempting.
Adam
Ron Newcomb
> That's true if I'm out to "win" -- to get the most treasure, to solve
> the dungeon, etc. -- but what if I'm out to define what "winning"
> means?
Ah, I was answering Victor's question with a yes, plus a reason. I
think I dropped an important paragraph saying so. *sigh* Sorry.
Oh, and before I forget, I agree with you too Conrad. :)
-R
Jim Aikin
You may be right. I had the vague impression that he was advocating
conversational menus as a generally valid tool or approach to providing
a conversation interface.
If you've got four NPCs and only one conversation menu for each of them,
with four topics to choose from, that's 16 possible branches for the
story. I don't think you're meaning to suggest that Victor thinks it's
practical to provide a story with 16 different emotionally meaningful
endings -- or, more realistically, hundreds.
The point I'm driving at, in my usual all-thumbs way, is that in any
given game MOST of the directions that NPC conversations can go will NOT
be especially meaningful in terms of opening up a different branch of
the story -- but that a few of them SHOULD be.
If none of the conversations leads to a different branch, then they're
just window-dressing. (Not that that's necessarily a bad thing, but
let's call it what it is.) If all of them do, you're deep in a
combinatorial explosion. Therefore, a few of them will, but most won't.
And if you've got a menu-based system, it will be necessary for the
player to try ALL of the branches in order to be certain that nothing
important was missed. Conversely, with an ask/tell system, it's up to
the player's intuition -- you mull over what has happened in the story
so far and try to come up with possibly useful topics of conversation.
When you learn something interesting, you move on with the story by
going to a different room or whatever.
It may be a puzzle to find the right topic -- but it's a far more
interesting puzzle than just choosing item 6b from a list.
I'm coming to be more firmly of the opinion that 'think about X' is the
right way to let the player take a peek at the likely topics of
conversation.
--JA
Ron Newcomb
> > [menu means] "THIS IS SOFTWARE. THIS IS A GAME."
On Mar 14, 2:21 pm, Emily Short <emsh...@mindspring.com> wrote:
> On Mar 14, 3:58 pm, Ron Newcomb <psc...@yahoo.com> wrote:
> > I agree with Jim here.
>
> When I read a book, it's still a physical object in my hands; it still
> requires that I turn the pages. When I'm at the movies, there is
> demonstrably a screen, with edges and curtains and occasionally
> flickery specks of dust. At plays or musicals, I can see the sides of
> the set, can tell that the backdrop is fake, etc. None of that stuff
> bothers me.
I think you're using the wrong metaphor. Using a keyboard in IF
doesn't bother us either. But when the "software" behind the work
shows how it's put together it's bad. The analogous example to a
novel would be seeing that a character did something because the
author needed him to for a plot reason, rather than a character
motivation reason: the "software" behind the work is showing.
(I sound contradictory, agreeing with both Victor & Jim while they
disagree with each other. I agree with Jim in that explicit menus
*are* a neon sign proclaiming software and they break immersion, but I
agree with Victor that there's right ways & wrong ways to do a menu if
you're gonna do one. Personally I'm cool with menus in game, provided
they're done the right way, since technology is such we're forced to
make a sub-optimal choice. Optimal being a Strong AI parser. )
While I'm trying to agree with everybody, I agree with you (Emily)
here:
> it's [...] problematic when brand new UI features appear in
> the middle of a game; the player has his filter in place, but now new
> aspects of the medium draw his attention.
-R
Emily Short
> > Jim Aikin <midigur...@sbcglobal.net> wrote:
> >> It's hard enough to come up with ONE emotionally satisfying ending.
> >> Trying to figure out three or four that are ALL emotionally satisfying?
> >> I lack the fortitude to attempt it.
>
> > I think that's precisely what Victor is attempting.
>
> You may be right. I had the vague impression that he was advocating
> conversational menus as a generally valid tool or approach to providing
> a conversation interface.
>
> If you've got four NPCs and only one conversation menu for each of them,
> with four topics to choose from, that's 16 possible branches for the
> story. I don't think you're meaning to suggest that Victor thinks it's
> practical to provide a story with 16 different emotionally meaningful
> endings
Fate seems to me an attempt to do exactly this. (I don't know the
exact number of endings, but that is probably on the right order,
anyway.)
The options are not arbitrary; the game asks the player to make a
series of decisions that all concern the same themes. What are you
willing to sacrifice in order to achieve safety for your unborn child?
What does "safety" even involve? Once you've made one sacrifice, does
it become easier to make the next? And so on. So the result is that
the endings are all reflections on a single set of moral questions. I
didn't think Fate a perfect success, but it is fairly unified
thematically, and the endings all make sense as outcomes of the
player's choices. (If there's a flaw, I'd say that there's not enough
"down time" in the game -- space when the player is not making
critical decisions but just being with the other characters and
thereby building an emotional attachment to them -- which means that
though the story endings make sense, they are not as resonant as they
might be.)
> The point I'm driving at, in my usual all-thumbs way, is that in any
> given game MOST of the directions that NPC conversations can go will NOT
> be especially meaningful in terms of opening up a different branch of
> the story -- but that a few of them SHOULD be.
>
> If none of the conversations leads to a different branch, then they're
> just window-dressing. (Not that that's necessarily a bad thing, but
> let's call it what it is.) If all of them do, you're deep in a
> combinatorial explosion. Therefore, a few of them will, but most won't.
>
> And if you've got a menu-based system, it will be necessary for the
> player to try ALL of the branches in order to be certain that nothing
> important was missed.
That's only true if the game doesn't communicate what sorts of
decisions are important. This is like the example offered earlier in
the thread, where the player can choose to protect his friend or turn
him over to the police. That's obviously a significant choice, where
"ask Bob about his family life" seems more like an information-
gathering choice.
In Fate, at least, I generally felt that I knew whether a given
conversational gambit was there merely to help the protagonist (and
the player) gain new information, or whether it constituted a
significant choice. (As I recall, in Fate, significant choices were
occasionally followed by some kind of "are you sure?" question, but I
think the writing and the thematic consistency did most of the work
here.)
Anyway, this whole issue that sometimes a menu forces you to make a
choice in the game strikes me as confusing the interface with the
underlying game structure. There are lots of games in which you can
make irrevocable choices through the ordinary command structure.
Sometimes those moves are obvious. Sometimes they're not. I suppose
the menu structure *does* impose on the player a certain clarity (all
options are enumerated, whereas through the parser they may not be)
and urgency (you have to make your choice NOW, not after taking
inventory, looking, and otherwise faffing about for a few more turns),
and that produces a different *feeling* to the choice. But with either
menu or command-line decision-making, the author still has the task of
communicating to the player what are the interesting choices in this
particular work; of revealing or concealing the various available
options; of making it clear when something is just exploration and
when something is going to Matter.
For comparison, I've played a few games where I was disconcerted
because a choice I made through the parser turned out to matter a lot
more than I expected. I seem to recall this coming up in Mind of the
Master, for instance -- commands I thought were merely going to move
me around the game map turned out to have major narrative effects, so
that I quickly got disoriented about what was going on.
Conversely, part of the intention of Floatpoint was to articulate to
the player a significant choice and set of possible options in a way
that would be a) clear enough that the player knew when the moment of
choice had arrived and what all his possible decisions were but b) not
require the use of a menu system.
Jim Aikin
>
> Conversely, part of the intention of Floatpoint was to articulate to
> the player a significant choice and set of possible options in a way
> that would be a) clear enough that the player knew when the moment of
> choice had arrived and what all his possible decisions were but b) not
> require the use of a menu system.
And that was one of its real strengths as a game! Plus the idea of
symbolic actions as a form of speech that will have real consequences. I
liked that a lot.
--JA
Victor Gijsbers
How are optimality and meaningfulness connected?
One way to make sure that none of the options you choose is optimal is
by not providing a metric for success: if the game leaves it up to the
player to decide which of the several endings is the _best_ ending, then
the choices that differentiate between these endings are necessarily not
ranked on a scale of good and bad--at least not by the game.
Would you rather commit suicide or let the Logos take over the world?
Should we fight the other tribe or welcome them? Do we want to convince
these guys to come back to earth, or not, given all the positive and
negative consequences? (To name just a few examples from recent IF.)
Are these choices not meaningful?
(I'm not saying that we have perfected the craft of making them carry as
much meaning as the can--we have not. But they already carry some meaning.)
Here is another example. One conversation branch reveals that the
protagonist is willing to kill even his own friends if the law requires
it. Another conversation branch reveal that the protagonist is willing
to defy the law he has promised to uphold if the life of his friends is
concerned.
This choice is meaningful--or at least can be, when embedded in the
right game. But that doesn't mean I as the author have to make sure that
one of them is _better_ than the other, does it? I don't have to end of
the choices with: *** Your fellow agent hears you, and shoots you in the
brain, screaming: "Traitor! Law is greater than friendship!" ***; while
the other choice leads to: *** You shoot your former friends, get a big
promotion and live happily ever after! ***
In fact, doing so would _rob_ the choice of its meaning rather than
_imbue_ it with meaning.
So I think it is demonstrably false that only choices ranked on a scale
of optimal/suboptimal can be meaningful. And I think this _has_ been
demonstrated.
Maybe we disagree about this:
> The whole point of conversing with an NPC is to move the story forward
> by making meaningful discoveries.
I would say that no, this is not the whole point of conversing with an
NPC. There is something else: the point of a particular conversation (or
of any other action in the game, for that matter, Emily is right when
she points out that we have moved beyond conversation here) can also be
moving the story forward by making a meaningful choice.
That is the essence of what I'm saying. The story need not just be there
to be discovered; it can also be there to be chosen.
And that's what I'm interested in, and my suggestion of using menu-based
conversations should be understood in that context. :)
Regards,
Victor
Tom Hudson
> One way to make sure that none of the options you choose is optimal is
> by not providing a metric for success: if the game leaves it up to the
> player to decide which of the several endings is the _best_ ending, then
> the choices that differentiate between these endings are necessarily not
> ranked on a scale of good and bad--at least not by the game.
>
> Would you rather commit suicide or let the Logos take over the world?
> Should we fight the other tribe or welcome them? Do we want to convince
> these guys to come back to earth, or not, given all the positive and
> negative consequences? (To name just a few examples from recent IF.)
>
> Are these choices not meaningful?
For those of us who also play computer "role playing games", there's a
common practice of saving before critical plot junctures. We want to
know the results of both choices, but either the game-before-the-
choice isn't interesting enough to play through a second time
("grindy"), it would take too long to play through a second time, or
would be exactly-the-same-up-until-the-critical-choice.
Then, in some of the newer crop of RPGs, you also are faced with all
sorts of small things you do during the game affecting the outcome
without making an obvious difference at the time. So you know (from
hearsay) that X is a possible plot branch, and you'd like to explore
X, but you never know just how to get there, or when you've locked
yourself out of that plot branch.
I'm tempted to do the same in IF. Most IF that I've played don't give
me meaningful ways to play differently through the story outside those
critical plot junctures, and the story isn't interesting enough to
play through exactly-the-same-way a second time to get to it.
Tom
Jim Aikin
>
> How are optimality and meaningfulness connected?
I think that's a question that each author would have to answer for
him/herself in the context of each individual work. I'm not sure it
would be possible to generalize about it. (Indeed, I'm not even sure I
could figure it out.)
> Would you rather commit suicide or let the Logos take over the world?
> Should we fight the other tribe or welcome them? Do we want to convince
> these guys to come back to earth, or not, given all the positive and
> negative consequences? (To name just a few examples from recent IF.)
>
> Are these choices not meaningful?
That depends entirely on the quality of the game -- the quality of the
writing, the depth of the characterization, the precise nature of the
plot problem, the culture in which the protagonist lives, and so forth.
I'm going to make up an example here. Let's suppose that the game is set
in a space station, and you can either save your crewmates from the
aliens and die yourself, or join the aliens and let your crewmates die.
Is that a meaningful choice?
Quite possibly not. Quite possibly the game is abysmal, the protagonist
impossible to identify with or relate to, the crewmates cardboard
cutouts, the aliens preposterous ... you see the problem? What creates
meaning is the author's compassion for the character, and his/her
ability to draw forth the same kind of compassion in the reader/player
through detailed vision and fine writing.
Nothing else matters, in the end. At least not in a serious work. In
comedy, of course, the resources and the intended effects are quite
different ... though not as different as one might suppose.
--JA
Victor Gijsbers
> Quite possibly not. Quite possibly the game is abysmal, the protagonist
> impossible to identify with or relate to, the crewmates cardboard
> cutouts, the aliens preposterous ... you see the problem? What creates
> meaning is the author's compassion for the character, and his/her
> ability to draw forth the same kind of compassion in the reader/player
> through detailed vision and fine writing.
Of course.
But it at least _seemed_ as if you were saying that meaning needed
something else as well, namely, an author-provided metric of success, a
ranking of the choices on a scale of optimality.
If I misunderstood you, or if that was not what you intended, then we
simply agree.
(I would stress that there are other things that can create
meaning--such as choices of the reader--but I absolutely agree with you
that meaningful choices can only happen within a framework successfully
set up by the author through... whatever it is that the author uses.
"Compassion" is not a bad choice, though I suspect it might not describe
all successful works. It does to a large extent describe what I am after
myself.)
To come back to the original topic of discussion--would you say that if
I do not provide a metric of success, but I do make the choices
meaningful (through setting up compelling themes and characters and
letting the player influence the character and comment on the themes),
then menus are not as bad as you would usually find them to be?
Regards,
Victor
David Fisher
news:338fb913-2e6d-4c95...@a70g2000hsh.googlegroups.com...
>
> One thing I observed recently which is maybe relevant: games where the
> player knows *more* than the viewpoint character, through knowing more
> about how the world works than the VP character does, have had several
> prominent recent examples: Lost Pig (vp char is not that bright),
> Child's Play (vp char is a baby), and Treasures of a Slaver's Kingdom
> (vp char is a stereotypical Conan-type hero, a little bit dumb).
>
> ToaSK is an especially interesting case because it thrives on making
> the hero comical to the player, through parody.
...
>
> I found that that affected my goals as a player: I was more interested
> in controlling the PC as a puppet in a way that made an entertaining
> story, than in controlling the PC as though I shared all his goals and
> identified with him.
>
> So that's one direction this kind of dissociation can go. I didn't
> feel as though the PC were "me" in any important sense at all.
Do you think this could work in a game which was more "normal" -- as opposed
to parodies (ToaSK) and linear games like Ramases, where the PC is
uncooperative?
I'm thinking of something where the player can both influence the PC
(participating in his character development) and observe him (ie. in spite
of the player's influence, the PC still won't go along with certain
actions -- or maybe needs to grow and change a bit more before he would be
willing to do a certain thing. Teach the PC courage by overcoming a lesser
challenge before he will be willing to take on a greater one; or get the PC
to do something distasteful before he will be willing to do something
*really* icky).
David Fisher
Adam Thornton
David Fisher <david...@australiaonline.net.au> wrote:
>I'm thinking of something where the player can both influence the PC
>(participating in his character development) and observe him (ie. in spite
>of the player's influence, the PC still won't go along with certain
>actions -- or maybe needs to grow and change a bit more before he would be
>willing to do a certain thing. Teach the PC courage by overcoming a lesser
>challenge before he will be willing to take on a greater one; or get the PC
>to do something distasteful before he will be willing to do something
>*really* icky).
Have you played _Gamlet_ all the way through? It's got a lot about
player-character complicity.
Adam
Conrad
>
> Have you played _Gamlet_ all the way through? It's got a lot about
> player-character complicity.
That's an alternative spelling of "Hamlet" -- does it have anything to
do with Shakespeare's play?
Conrad.
jon.i...@gmail.com
> The other is perhaps a little less at odds with the goal of
> interactive narrative, though. The thinking goes a bit like this: the
> author has provided a branching story here. Therefore, he had
> something he wanted to convey that could not be conveyed by showing me
> just one narrative path (either in a short story or with the linear-IF
> techniques we see in lots of games). So part of the point of the game,
> perhaps, is to explore and compare these different branches, and use
> that comparison to elicit a deeper meaning from the game.
I really hate this line of argument, although I appreciate plenty of
players disagree with me! I've never seen the appeal of interactive
experiences as being one of comparing different timelines/outcomes -
as Victor said somewhere, this sounds like a very boring way to
proceed, and the idea of making a game that requires playing more than
once seems like an arrogance on the part of the author. Which is not
to say that multiple readings can't enhance the experience, of course,
just that I hate the sensation of finishing a game and thinking "did I
do that right? Or should I have done it a different way." I remember
after All Roads came out, several reviewers commented that they
intended to "go back and see if there were other endings". Which was
nice because it means they felt engaged and involved with the
development of the narrative; but was also misguided, because there
were no other endings.
I can see that a game with this replay/compare model could be very
effective and powerful so I won't suggest it's categorically flawed.
What I might say is that it's not a very constructive paradigm, in
that it doesn't (in my view) open up the widest and richest seam of
possible experiences.
(I should note that "Dead Cities" sort-of has multiple endings, but
the point was never to replay, more to accomodate what the player
might do smoothly. The multiple endings are lazy - for the most part,
it's the same structure with text swapped in and out - but, that said,
several players still went through more than once looking for an
optimal outcome. )
In all honesty, I think the desire to "see all endings" comes from the
fact that most of us players are also designers, and we want to see
how the game works and what makes it tick. It's like watching a
romantic comedy and trying to spot the details of the formula: a sound
enterprise because there is one, but it rather detracts from the
enjoyment!
jon
David Fisher
> In article <12058775...@kangaroo.ozonline.com.au>,
> David Fisher <david...@australiaonline.net.au> wrote:
>>I'm thinking of something where the player can both influence the PC
>>(participating in his character development) and observe him
> Have you played _Gamlet_ all the way through? It's got a lot about
> player-character complicity.
No, I haven't (not quite my thing!) -- how does complicity work in Gamlet?
David Fisher
Ron Newcomb
> I can see that a game with this replay/compare model could be very
> effective and powerful so I won't suggest it's categorically flawed.
> What I might say is that it's not a very constructive paradigm, in
> that it doesn't (in my view) open up the widest and richest seam of
> possible experiences.
I have a "me too" for jon.
-Ron
jon.i...@gmail.com
> Teach the PC courage by overcoming a lesser
> challenge before he will be willing to take on a greater one; or get the PC
> to do something distasteful before he will be willing to do something
> *really* icky).
I do like this idea, but it strikes me as more of a gameplay mechanic
than a narrative one. I suppose it draws a bit too much attention to
the player/PC split which damages immersion. Strangely, in a game like
Rameses I didn't feel this to be a problem, however.
I think I'm willing to accept a PC who won't do everything I say, but
as soon as I start to think around the split and try to work out how
to cajole my PC, then I'm pointing out the gap rather than accepting
it, and that's a problem.
Of course, different scenarios could handle this better. A "voice over
the radio" game would be an ideal set-up for this sort of thing (and
Fail-Safe kinda does this, a bit, at the end).
jon
Jeff Nyman
news:ffd2aa00-036d-49e7...@e10g2000prf.googlegroups.com...
> I think I'm willing to accept a PC who won't do everything I say, but
> as soon as I start to think around the split and try to work out how
> to cajole my PC, then I'm pointing out the gap rather than accepting
> it, and that's a problem.
That's a good distinction.
I wonder, though: maybe it's not that a PC does not do everything the player
says. But, rather, maybe its *how* the PC does it that can be interesting to
the player. Or maybe what the PC's interpretations of the action or its
consquences are, as told to the player by the PC. Or maybe its the reactions
of other characters to the action taken by the PC that can surprise and
entertain the player. Or the story has been set up sufficiently, the PC
makes it clear what has to happen in some fashion that the player can at
least accept.
The key here is that if the player doesn't necessarily equate themselves
with the PC, they do not necessarily get surprised when the PC does not do
something or, at the very least, has some concerns about doing it. The
assumption is that immersion happens because the PC and the player can
conflate. I've found that when you take that assumption outside of a largely
game-playing crowd (or one that has been mired in textual IF's conventions
for a long time) that idea of immersion is flipped on its head. It's the
very conflation that causes people to not become immersed in the story.
(They may become immersed in the game, but then they're thinking about it
more in terms of a game rather than as a compelling story.)
Anyway, going back to your point, I think it's definitely a valid one but I
would just caution that we can think of a gameplay mechanic *and* a
narrative mechanic. It's not necessarily an either/or. And the two mechanics
together are, to me, where some of the challenge to story-driven textual IF
really come in.
- Jeff
Adam Thornton
I think I can safely say "yes" without too many spoilers.
Adam
Adam Thornton
David Fisher <david...@australiaonline.net.au> wrote:
>"Adam Thornton" <ad...@fsf.net> wrote in message
>news:5ic6b5...@dhcp-va3-30.va3.sinenomine.net...
>> In article <12058775...@kangaroo.ozonline.com.au>,
>> David Fisher <david...@australiaonline.net.au> wrote:
>>>I'm thinking of something where the player can both influence the PC
>>>(participating in his character development) and observe him
>...
>> Have you played _Gamlet_ all the way through? It's got a lot about
>> player-character complicity.
>No, I haven't (not quite my thing!) -- how does complicity work in Gamlet?
OK, spoilers ahead
So, during the course of the game, the PC is revealed to be suffering
from (inherited) syphilis, which has unhinged his mind about the time he
has reached adolescence, which is when the game is set.
In the endgame, the character must perform some disgusting and
contrary-to-his-usual-nature acts in order to advance the story to its
conclusion. There is speculation as to whether these actions are due to
the breakdown of his mind due to his madness or to literal demonic (or
divine) possession by the player. And certainly the things you're
making the poor lad *do* aren't particularly *good* for him. Thus the
question arises why you're torturing him to advance the story, and then,
at the end, the author steps out of the frame to, in effect, berate you
about the head and shoulders with a sheep's bladder on a stick.
The issue of complicity is raised more subtly in _So Far_, of course:
the puzzle with the boy. Also in _Metamorphosis_, come to think of it.
Look at the different ending texts you get based on the different
degrees of violence you apply to solving the various puzzles.
Adam
Blank
> Conrad wrote:
>
>>
>> "Gee, you're such a good friend," Bob says. "I'm so glad I confided
>> in you."
>>
>> "No problem," you say: "You can always trust me."
>>
>>> Betray Bob to the police. <
>> Betray Bob to the mafia.
>> Keep it quiet.
>>
>>
>> -- I have no inclination to lawnmower, at least not immediately.
>> Maybe later, on replay, I'll want to explore the options. But that's
>> not lawnmowering, that's fun.
>
>
> You are very right, and make an excellent point about how important it
> is to make clear to your readers/players (I would very much like a
> single word that has both these meanings, but neither "pleaders" nor
> "rayers" sounds very good) what kind of choices you are offering them.
>
> Are it adventure book style "your life depends on this but you cannot
> know what the right choice is, ha ha ha!" choices?
>
> Are it sophisticated but cruel IF "your life depends on this so you'd
> better gather all the clues and find out what the right choice is before
> you try" choices?
>
> Are it narrativistic "you've got a choice here as to which direction the
> story moves in, and all the possible choices are valid and will lead to
> something interesting" choices?
>
> I agree with you that the first possibility is very bad indeed; while
> the second and the third, though both interesting, should probably not
> come together in a single game, and any game which uses one them should
> make very, _very_ clear which one. (And of course there are other and
> in-between possibilities.)
>
> Now both my published works involve choices of the latter kind; and I
> was thinking of that kind of choice when I asked my question. But I
> didn't make this clear; and yet, making this clear (not just now, but
> also in the game itself) is essential.
>
>
> So might it be the case that conversation menus (and other types of IF
> interface where it is clear that you are making an irrevocable choice)
> are less of a problem if it is clear that there is no _right_ choice, no
> choice you have to make in order to find an essential clue, or get to
> the rest of the story, or understand what is going on?
>
>
> Victor
Thinking it over, it ocurred to me that the times menu choices worked
best for me was not when I was asked to choose an /action/ from a list,
but when I was asked to choose from different /consequences/. Perhaps
that's only a different way of saying "clue important decisions
sufficiently", but it seemed a useful way of thinking about it.
So imagine: You've been sent to find Fluffy, your best friend Karl's pet
poodle. You find her, but she's been trapped in machinery and nastily
mangled but not killed. Through the ordinary IF commands the player can
either mercy-kill her, rescue her (still suffering) or leave her to die
naturally.
Let's imagine now that the player returns without Fluffy and gets into
(partially menu-driven) conversation with Karl. The decisions about what
to tell Karl would be based on his character, and (assuming Karl had
been well enough developed) the player should have a reasonable guess at
how Karl would react to his best friend seeming callous, or lying to
him, or appearing incompetent, or breaking a promise. So the player is
not really thinking about the mechanics of choosing A, B or C, but
rather the implications and consequences further into the game.
Notice that from the way I've listed them it's clear there is no "get
out of jail free" option: the player is just being offered a choice of
later problems.
--jz