Last night I was watching Alias, and realized just how pulpy the show
is. Like old pulp fictions, comics, radio shows, where each plot
builds on the last but doesn't seem truly connected. I thought about
making serial interactive fictions.
Then this morning I was looking at a decision tree and started
thinking about it again. Plotting out a large game in advance, but
releasing it piece by piece. I also thought about how so much artwork
goes into comic books, yet they meet a monthly schedule.
With this in mind, here's my proposal.
I want to bring together a diverse group of people who can bring
different strengths to a project to create a serialized interactive
fiction. While people will specialize in an area (plot, coding,
descriptions, etc.) everyone is welcome to contribute to any part of
Two concepts are in play here.
1. That a team of people will be involved in creating an interactive
2. These interactive fictions will be released serially on a schedule
(quarterly most likely).
Everything else is to be determined by the team. For example, I want
to create an alias type game that builds on a central plot and each
story acknowledges the previous ones and builds around a central
secret to be revealed at a later date. It may be decided, instead, to
create a star trek type game where each game is a discrete unit and
barely acknowleges other episodes other than large trends.
Hopefully we can all gain experience in working on software projects.
I'm not sure it will be resume` worthy, but who knows, if it goes well
maybe we'll incorporate and start working on a comercially viable
Comments, crits, and flames (and volunteers) welcome here on RAIF or
This is something I've been wanting to do for a long time. I never
managed to work out how to make it feasable, but I'm all for trying
it, just the same.
What problems did you run into? I'm all for learning from others.
I'm figuring deadlines and roles are the most important parts. You can
miss a deadline, but at least it's there to say "hey, where's the
product?" So things like 'the plot is due by XX, the room descriptions
are do by YY, and the code is due by ZZ' goes a long way towards
having people focus on what's important rather than spend their time
arguing or worrying about logistics.
The trick is finding people who enjoy doing every piece of the puzzle.
People who enjoy plotting, people who enjoy making puzzles, people who
enjoy coding, etc. And people who are willing to do it for no monetary
Well, firstly, the IF community has always been firmly of the belief
that team-authorship is not well-suited to IF.
Also, I spent a lot of time trying to sort out logical connections; if
one doesn't insist that episode 1 ends in a certain way, episode two
doesn't make sense. I spent some time trying to work out the
boundary. I'll abstract and befuddle one of the series I had
Suppose that episode one of Survivor: Moon Base Alpha has Jimmy, the
plover mechanic from mars, voted out the airlock. Now, obviously, he
can't be in episode two. So, you've got three options: either rig the
game so that jimmy is *always* voted off the island, ignore the issue
of who was voted off last week, or have some way of informing episode
two of what happened in episode one.
The first option is the easiest, though it's liable to result in very
linear games. THe second option will work for small things (To take a
page from Star Trek Voyager, it's always okay to ignore the question
of whether or not we made peace with Alien Race X in the next episode,
bvecause we'll never see or hear from them again), but is infeasable
on the large stuff. Plan three is, obvoiusly, the most interesting,
but it has some insanely major problems.
THe first major problem is: How do you implement this? It could be as
simple as askiong the player, though this would probably be
tedious. ALternatively, you could use a password to configure the
game (ie, when you finish episode one, you're given a password, or the
interpreter creates a file, etc, which is given to episode two). But
this too has issues; obviously we don't want to force the player to
play every episode up to the current one if he's just found the page
and is interested. ANd if the player wants to try to see the season a
different way, should he have to play the whole thing over? (I
imagine some kind of hybrid system would work best, but this si a
programatical nightmare which I didn't have time to solve then.)
Even worse, though, you get a combinatoric explosion. Sure, you might
want to allow everything the player does in episode 1 to influence
thigns later on, but... Let's say that there are 20 castaways on
moonbase alpha. There are 20 ways episode 2 can begin. Another
castaway is voted off in episode two, so there's 20*19 ways episode 3
can begin. This is bad mojo, since you've got an insanely large
number of possibilities.
Also, if you're going to pass information back and forth from one week
to the next, you need to do a lot of planning ahead. If you decide
that episode 61 is going to have elements in it that depend on what
the player did in episode 7, you'd better darned well hope that
episode 7 exported the relevant details. ANd if you've got to plan
that far ahead, you start to lose the advantages of having not written
the whole thing as one epic game.
So, these are hte major issues I came up with. If people are
interested in this sort of project (I, like you, I think, also
envisioned this as something of an episodic TV paradigm), I'm still
massively interested. I've got a few series ideas simmering on the
I find that the more people share responsability for completing a
project, the more likely the project will be delayed.
If multiple people can by their actions delay the product, you are
multiplying the riskes, and if one becomes late, the others will have
less pressure and become late also.
Also, there is all the blame thing, where one has to wait for
something the other one is supposed to do to be able to move forward.
I can see multiple people working on a serie of IF games, but not by
sharing the coding of every game.
Most efficient, I think would be to have each person responsible for
an episode. Keep the episodes short and plan them in advance. Of
course, every person would have to be quite versaile.
Even this way, you better have some filling eposodes if an episode is
It will be important to have someone dedicated that manages the
general plot line and the episode completions.
I'm interested in doing this. I seem not to be the only one. So, how
I propose the formation of an "Interactive Fiction Network" -- an
organization which exists to release serial IF.
Some (not paerticularly well organized) thoughts:
The IFN will, at regular intervals, release episodes belonging to a
number of IF series.
The organization of members: Each series will have an executive
producer, who is generally responsible for the series. It is proposed,
though not required, that each episode in a series is written by a
single primary author, with a number of other consulting authors.
Three styles of series occur immediately (these are not precise
categories; any series is liable to draw on elements from more than
- World-linked stories
This is a series of stories which share a common model world --
like the standard sets of, say, a situation comedy.
Implementationally, this may take the form of a shared library
implementing the world model.
- Loosely-linked stories
This is a series of stories which share some logical element, but
do not necesarily share objects directly between episodes. Such
stories would probably be character-driven, and would follow the
adventures of character or characters.
This is a series of stories which are linked only thematically,
such as 'The Twilight Zone'
As you can tell, there's a lot of details to hammer out.
And since this isn't a formal proposal, and since you can formally
propose allyou like and nothing gets done unless you do something
about it, I'm going to grab the reighns here and say that, barring
anyone else stepping up and saying that they want to, I'm willing to
act as organizer for the thing. Mail interests to me, and I'll set up
some kind of mailing-list deal.
In fact, even better, don't just send interests. If you've got a
series proposal, send it in. If you have a firm idea as to what sort
of thing you'd like to do, send it in.
There's , I'm sure, room for all things here -- writers, code
specialists, technobabble translators, graphics, whatever.
How about it?
L. Ross> Also, I spent a lot of time trying to sort out logical
L. Ross> connections; if one doesn't insist that episode 1 ends in
L. Ross> a certain way, episode two doesn't make sense. I spent
L. Ross> some time trying to work out the boundary.
This may be too hokey for you, but the first solution that occurs to
me is to have the series run backwards in time. (i.e., episode 2 is a
prequel to episode 1.) That device has worked well enough in a lot of
joh> This may be too hokey for you, but the first solution that
joh> occurs to me is to have the series run backwards in
joh> time. (i.e., episode 2 is a prequel to episode 1.) That
joh> device has worked well enough in a lot of fiction series.
Oops -- caffeine turning on brain now-- that doesn't solve anything,
does it? It just means your episode 2 ending has to be constrained to
make sense with episode 1. Never mind.
So the title of this thread should actually be...
I go now. *runs*
One possibility, if this doesn't throw the discussion off in a wrong
tangent, is something akin to the _Sanctuary_ series of books, aka
"Thieves' World." (I remember the details of the experiment more than
the books themselves -- bear with me.) This ties in with your first and
The idea was, a group of SF and fantasy writers hammered out a
fantasy-based city and world, complete with maps and government and
cultures and so forth. They left the details ragged enough so they
could go off on their own and craft short stories that fit this new
world without constraining their imaginations. As the series
progressed, authors would reference or utilize characters and incidents
from previous works, even from stories they didn't pen. There was no
tight-knit sequence; reading a collection was more like a series of
contemporaneous visits with interesting people in the city. Of course,
if a character dies in one story in one book, it doesn't make sense for
him/her to pop up in the next, so temporality was an issue. (Although,
I do recall one character dying and showing up in another story in the
*same* book ... this wasn't a problem for me, though.)
It wasn't a group-writing project; each story was penned by one author
then collected by an editor. So, my suggesting is, treat each IF game
as a short story rather than a novel chapter. My gut feeling is, this
approach would be a lot easier than an episodic series of games, one
hinging on the previous' ending. I also think it'll give each author
more freedom, and in the end, better games.
I was only proposing one series that could do some genre hopping.
Something along the lines of "You're a modern day Joe Guy (or Jane
Girl) when you get sucked into this vast international conspiracy
(a-la the Matrix)." The main plot would follow along these lines, but
then there's the one-off episode where you discover a time machine and
have to go back in time (to say medieval times) to stop the murder of
someone. Or the one off where you have to travel to a secret base on
Mars. Satisfying a desire to do a fantasy or sci fi, but not related
to the main plot. Like the musicals on Xena - fun, but let's get back
to the main plot.
The series as a whole would have an "executive producer" but anyone is
welcome to submit ideas, and if approved, carry them out themselves or
with a/the team, kind of like having a guest director. This executive
producer is where the "buck stops" and s/he is responsible,
ultimately, for getting things done, making decisions, and recognizing
and ending unproductive activity (like too much deliberation over the
plot or platform). I always thought I would be this person. ;)
To address some of the logistics mentioned earlier, I think your
"Survivor" example is too open ended. By it's nature, it would be
episodic, and each episode would be self-contained, but elements would
carry from game to game, and yes we would plan far in advance. For
example, in episode 1 you hack into a computer and get a password. In
game 5 you're asked for that password. Of course, events in game one
determine which password you get, and which password you enter
determines what happens next.
The external trappings remain the same from game to game, but certain
key scenes change depending on your actions. If you don't get the
password in episode 1, or never played it, no harm, no foul, you just
don't learn another secret, and in episode 10 you won't know that the
closet is really an entrance to an underground datacenter, and in
episode 15 (season finale) you won't be told certain aspects of the
In other words, you cleverly hide the things that are important, the
things that carry and the things that don't. As far as episode
planning, there are one of two possible approaches that I think should
be considered. One is "Planning Planning Planning" where you know in
advance what will happen in nearly every episode. The other is "By the
Seat of Your Pants" where you do things on the fly with very little
planning. I think a balance between them has to be struck.
I imagine a small tightly knit development team. They would all know
the end state, the secret that's revealed, and can work towards it,
drop hints, and drop false hints. In between, we might not know how to
get there, but we would have fun along the way. Planning will be part
of the fun, and we will begin to weave threads in that will have
lasting effects. Subplots that get revealed in later episodes, things
Think of how well Back to the Future II integrated with the first
movie, though I don't think they knew what they would do in the second
movie. Somehow those episodes fit together nicely without advance
I also believe it would be an enriching experience for all involved.
We would all learn about programming, about plot, about characters,
about teamwork, about managing projects, about customer interaction,
etc. Who knows, this experience may lead to making each of us more
financially viable in terms of saleable skills. Pie in the sky, we
could get funding and start our own game company.
The idea of having a person responsible for each episode may work. I
don't really know. I tend to believe that a small team might be better
(3-5 people) than an individual person for this particular purpose.
Maybe we could set it up as a competition with two or three small
teams, each working on episodes to see who comes up with the best
episode. Have a voting feature where fans of the series can vote for
their favorite episode. Have three seperate but inter-ralated plots
(one for each team), have a team for the good guys and a team for the
bad guys, and plays would have the opportunity to play both. The ways
of deciding on the teams will be determined in large part by who
volunteers. If we have 50 people we can do a lot, but the logistics
might get out of hand. If we have 5 then we can do less, but perhaps
To address the "mupltiple people slow things down" comment, I'm
familiar with the ideas in the Mythical Man Month (qv), and work as
a non-programming IT manager. I have to balance the enjoyment and
involvement of everyone on the team and get things done, I hope to
carry that over to this project as well. I want to bring the best out
in people, identify their strengths and build on them. That's the
skill set I bring to the table (as well as my knowledge of plot, my
creativity, my minimal knowledge of programming, etc.). This
experiment is as much about distance teamwork and the people involved
as it is about the games that would come out of it.
My plan was to give it enough time to get the feedback of everyone on
this newsgroup who cares to contribute (I assumed one week for
discussion to ramp up, get good feedback and find people who are
interested), and then decide either to set it up or scrap it.
I'm familiar with the gung-ho excitement of a new idea, and I'm also
familiar with how often that peters out once the real work has to get
done. I'm trying right now to balance the managing of expectations and
the creation of excitement about a new project.
Is anyone here familiar with the now defunct (?) beehive development
group? It was a collaborative group of programmers (for mac) who came
together to propose projects, develop products, and market them. This
sounds very much like your idea, L Ross.
 specifically, the idea that the more people who are involved in a
project the longer it will take, and that the more complex the project
the more time it will take exponentially from a smaller project.
>The series as a whole would have an "executive producer" but anyone is
>welcome to submit ideas, and if approved, carry them out themselves or
>with a/the team, kind of like having a guest director. This executive
>producer is where the "buck stops" and s/he is responsible,
>ultimately, for getting things done, making decisions, and recognizing
>and ending unproductive activity (like too much deliberation over the
>plot or platform). I always thought I would be this person. ;)
Right, and that's perfectly amenable to the structure I'm prosposing
here -- I didn't mean to steal your thunder or anything; as I said, I
only offered myself as organizer to make sure there was one -- and
because I imagined you would want an executive producer role, as you
said. It looks a lot like you're interested in exploring serials
which are more loosely knit than the ones I'm interested in.
>To address some of the logistics mentioned earlier, I think your
>"Survivor" example is too open ended. By it's nature, it would be
>episodic, and each episode would be self-contained, but elements would
Yeah. It was just the easiest way to get a combinatorial explosion I
could think of.
>In other words, you cleverly hide the things that are important, the
>things that carry and the things that don't. As far as episode
>planning, there are one of two possible approaches that I think should
>be considered. One is "Planning Planning Planning" where you know in
>advance what will happen in nearly every episode. The other is "By the
>Seat of Your Pants" where you do things on the fly with very little
>planning. I think a balance between them has to be struck.
Certainly. The 'planning planning planning' approach is a lot more
difficult in a multi-author environment -- if you look at TV shows,
the ones which best manage a consistent storyline across long
stretches of episodes are often controlled very tightly by a single
>I imagine a small tightly knit development team. They would all know
>the end state, the secret that's revealed, and can work towards it,
>drop hints, and drop false hints. In between, we might not know how to
>get there, but we would have fun along the way. Planning will be part
>of the fun, and we will begin to weave threads in that will have
>lasting effects. Subplots that get revealed in later episodes, things
The tighthly knit development team is very tricky to implement.
>Think of how well Back to the Future II integrated with the first
>movie, though I don't think they knew what they would do in the second
>movie. Somehow those episodes fit together nicely without advance
But, curiously, Back to the Future 2 and 3 have almost nothing to do
with each other, despite the fact that they were produced
>The idea of having a person responsible for each episode may work. I
>don't really know. I tend to believe that a small team might be better
>(3-5 people) than an individual person for this particular purpose.
Well, you run into trouble; aside from the fact that organizing a team
over the internet is hard, there is also, I think, a longstanding
belief around here that collaboration is Lots Of Trouble.
>Maybe we could set it up as a competition with two or three small
>teams, each working on episodes to see who comes up with the best
Do not say the 'C' word.
This certainly does sound like a feasable idea, but it seems to be
along the lines of a fan-fiction. The only difference being who
establishes the world. Personally, I would tend to believe that unless
we could make a world that was really really attractive, it would be
difficult to attract writers until it was sufficiently fleshed out. So
if, for example, two or three people would agree to this concept,
write games in this shared world, and then let people know they can
also make games in this world, then it might seem attractive. Though
I'm still tempted to think that unless this world is Final Fantasy or
Metal Gear people would still turn it down. You would really have to
capture people's imaginations.
I remember the Robot City series of books which were Isaac Asimov
sanctioned, but written by different authors. One writer would
complete the book and hand it off to the next author. The concept was
novel enough to make it interesting: robots who were activated in a
world with no humans. Introduce the two human main characters and
voila, chemistry. A chemistry that each author could run with.
This seems more like an idea to throw out into the ether. Get together
with some buddies and make up a world. Ask if anyone would be
interested in creating an IF in this world and just see what happens.
Or get together with the same buddies and individually write IFs in
this world, and in each one have the appropriate disclaimer about the
legalities of creating IF in this world.
My 2 cents.
> Last night I was watching Alias, and realized just how pulpy the show
> is. Like old pulp fictions, comics, radio shows, where each plot
> builds on the last but doesn't seem truly connected. I thought about
> making serial interactive fictions.
This is sort of what I'm shooting for with Earth & Sky. In fact, release 2
of E+S is subtitled "Serial Interactive Fiction," in hopes that players
won't feel quite so caught short by the ending. The feel I was going for
was of issue #1 of a new comic, introducing the characters, involving them
in a mini-story, and introducing an overall plot arc.
E+S is a solo project, and so doesn't really fit into your concept
otherwise, but I'm hoping to prove the viability of at least a "limited
series" of serial IF pieces.
Paul O'Brian obr...@colorado.edu http://ucsu.colorado.edu/~obrian
SPAG #27 is here with interviews, articles, and reviews about the 2001
IF competition! Let joy be unconfined. http://www.sparkynet.com/spag
1. To pull together a group of people to see whether or not they can
create an interactive fiction collaboratively. To have fun. To learn.
2. To create serialized interactive fiction people will get excited
about, talk about, and even gossip about.
My vision (yours may vary and different visions are welcome) goes
something like this.
A team of people, somewhere between 3 and 10 will come together and
work on a plot, a world, characters, etc. Enough to start writing
games. The games at first would be small, but would grow as we got
more comfortable. Some games would be one-off's in the world, others
would be directly related to a central plot. The same way season
finales in many TV shows have to do with the central plot, and while
many episodes in between do not.
We would struggle with and overcome working as a team, plotting,
characterization, architecting, mapping, coding, etc. and try to
maintain some sembelance of a release schedule. (I look to Tom Biskup,
Joel Spolsky, Frederick Brooks, and my own experience for the
realities of development, creativity, and schedules.)
If the world itself gets exciting enough, other people may want to
write games in the world, and we would let them as long as the events
didn't clash with ours. I.e. if we wrote a Lord of the Rings someone
who wanted to write a Hobbit would need to include a sequence about
finding a ring.
I want to experiment with a few ideas. Collaborative development.
Serial Interactive Fiction. Hype & Marketing. Plot in these unique
constraints among other things. Mostly it probably stems from a desire
to "be part of something bigger than myself," to experience the rush
of accomplishing something through hard work. To create something
artistic without expectation of monetary rewards. To find, however odd
and indirect, a way to communicate with people and entertain them.
I have a firm belief that this could work, and that we all could
benefit from the experience. Working on teams at work and in (musical)
bands, I know that a group of people can come together for a task and
create something larger than the sum of it's parts, and that the
biggest obstacle is time. Setting aside the time to do the things.
1. Begin work on a viable world and start spreading the idea around.
2. Attract people to the project & lay ground rules for behaviour,
3. Begin work on plots, both long and short term.
5. Solve logistical problems that are immediately related to
6. Write a quick one-off game to see if this distance-collaboration
thing can work.
7. If we actually get this far, we'll address this step then.
We'll have to think of a medium for communication that is private so
the world doesn't know exactly what to expect. (mailing lists, chat
Major plot points, a world, themes, etc.
How we will go about the actual coding (CVS? One person ownership?
What sized team(s).
Arbitration (who has final word? one person? vote? what about personal
Marketing (websites, what do we say about the game, etc.)
Assuming I meet my goals as discussed above, I don't want to get
caught up in deliberation about minor details, that will stall the
I don't want to create a world that is flat or one dimensional or a
plot that cannot be exploted.
While I want everyone to care about the project, I don't want it to
become a burden or obligation that anyone for any reason dislike being
I don't want the "too many cooks spoil the soup" syndrome where
democracy leads to something that's more lowest-common-denominator
rather than the best of what each of us can bring out.
I don't want to over-plan, no do I want to leave it so open that the
games wander aimlessly about.
Organizing any group is difficult, so we'll be bound by whatever tools
are available free or donated. A CVS system will be important.
The nature of 'serial' games will impose it's own constraint in terms
of plotting and timely releases.
Communication - I'm assuming e-mail, and occasional ICQ/AIM/IRC type
chats will occur when communication needs to be rapid.
Ego is a constraint. Neither egomania nor an overly sensetive person
(flip sides of the same coin) will be healthy. I would hope to find a
way to keep people and finding positive ways for them to contribute.
Divvying up the work. I'm sure we'll come up with tasks that become
like "chores" and while I would do my best work with anyone who asks
for it, sometimes a nose needs to be put to the grind. Otherwise it's
all just talk.
This is by no means a complete document, it just represents my
thoughts at the time. Again, I view this as an experiment, an
opportunity, and a fun, challenging and rewarding way to spend time.
In terms of next steps, a "show of hands" of people who would be
interested, and then we can decide on setting up some sort of mailing
PS, L. Ross, no hard feelings. Sorry I got a little offended. I'm glad
someone else out there shares my enthusiasm for this type of thing.
> One possibility, if this doesn't throw the discussion off in a wrong
> tangent, is something akin to the _Sanctuary_ series of books, aka
> "Thieves' World." (I remember the details of the experiment more than
> the books themselves -- bear with me.) This ties in with your first and
> second categories.
> The idea was, a group of SF and fantasy writers hammered out a
> fantasy-based city and world, complete with maps and government and
> cultures and so forth. They left the details ragged enough so they
> could go off on their own and craft short stories that fit this new
> world without constraining their imaginations.
And at first it did make for a good series. In the end, however, they
got more and more constrained by the world they had created for one
another, and felt the need to go off into new, uncharted territory,
using a war campaign as the excuse. IMO it all went downhill from there.
I'm actually toying with a similar idea: at first, I was planning a
really large adventure game divided into chapters, but then I thought,
why not make each chapter (or couple fo chapters) a separate game?
That would make the resulting games much easier to handle both for the
writer (me), the testers and the audience.
> It's worth remembering that this kind of thing *has* been done in IF, in a
> sense; Sierra had the Kings Quest, Space Quest and Police Quest series, and
> LucasArts had the Monkey Island series. Of course they all used the first
> I suppose it's different in the sense that these were all stand-alone,
> full-size games, but the series worked because while the games weren't
> necessarily linear in themselves, different players would have achieved
> essentially the same things at the end of each episode, regardless of the
> route taken. So the essential story is fixed, and can continue happily into
> the next part.
Why is everyone so obsessed with alternate endings anyway?
Let's examine Metal Gear. This is an interesting example. I don't
think I'll be too spoily right now, but stop reading HERE if you don't
want to know something about Sons of Liberty or Solid. I've never
played the Sons of Liberty so my information comes from an interview
with the author anyway. In Solid there were two possible alternate
endings, either meryl dies or she doesn't. In Sons of Liberty she
simply doesn't come up. Doesn't make an appearance, isn't even
mentioned. If another sequel is made, that might be addressed.
Subplots and threads would appear and disappear in different games.
Sort of like this:
Game 1: a, b, c
Game 2: b, d, e
Game 3: c, d, f
Game 4: a, b, f
etc. This way you would just write alternate versions of one scene.
Even the puzzles can remain the same. Or perhaps knowledge from game 1
can lead to a small subplot, enrichment, or shortcut in game 4. Like
the "Hello Sailor" in the Zork series.
In a collaborative environment, people could work on sections and
release them when they're ready. Let's say we have quarterly releases.
If team/person "A" isn't done with their piece yet, we'll release in a
different quarter. Of course, this complicates the job of whoever ties
things together, but... I guess those are the tradeoffs that need to
> I think it's going to be very difficult to get away from this. Otherwise, if
> you let there be two materially different ways of ending the first episode,
> you're going to end up having to write two different stories for episode two,
> each for only a half of your likely audience. And why would you want to do
> that when you could write two new games that could be played by everyone?
I don't want to give away all my ideas here, but there are alternate
ways to ensure plot consistency and offer variety.
I've received some interest in this idea, so I've set up a Yahoo!
Group for this (sorry, no Majordomos at my disposal).
The URL for this is: http://groups.yahoo.com/group/tsif
or sign up by e-mailing: tsif-su...@yahoogroups.com
Interested parties welcome. Kibbitzers will be fed to Grues.
I only read the first book, and this was many years ago, so I'll take
your word for it.
However, if such an undertaking was undertaken in the IF world and it
resulted in so many games that the series wound up cratering, well, I
don't count that as a failure.
: Well, firstly, the IF community has always been firmly of the belief
: that team-authorship is not well-suited to IF.
: Also, I spent a lot of time trying to sort out logical connections; if
: one doesn't insist that episode 1 ends in a certain way, episode two
: doesn't make sense. I spent some time trying to work out the
: boundary. I'll abstract and befuddle one of the series I had
: Suppose that episode one of Survivor: Moon Base Alpha has Jimmy, the
: plover mechanic from mars, voted out the airlock. Now, obviously, he
: can't be in episode two. So, you've got three options: either rig the
: game so that jimmy is *always* voted off the island, ignore the issue
: of who was voted off last week, or have some way of informing episode
: two of what happened in episode one.
: The first option is the easiest, though it's liable to result in very
: linear games. THe second option will work for small things (To take a
Not necessarily. Suppose you make sure that Jimmy is always voted off first.
How other characters see you might change because you voted off Bob, or
tried to influence others to vote off Bob, or whatever. This could
cause for some linearity by setting up the "alliances" you will need later,
and perhaps later on, you could influence the vote results, since there are
less people and your vote matters more. In this way, you would still
have to carry over data, but you wouldn't have to encode extra
characters that aren't used.
: page from Star Trek Voyager, it's always okay to ignore the question
: of whether or not we made peace with Alien Race X in the next episode,
: bvecause we'll never see or hear from them again), but is infeasable
: on the large stuff. Plan three is, obvoiusly, the most interesting,
: but it has some insanely major problems.
: THe first major problem is: How do you implement this? It could be as
: simple as askiong the player, though this would probably be
: tedious. ALternatively, you could use a password to configure the
: game (ie, when you finish episode one, you're given a password, or the
: interpreter creates a file, etc, which is given to episode two). But
: this too has issues; obviously we don't want to force the player to
: play every episode up to the current one if he's just found the page
: and is interested. ANd if the player wants to try to see the season a
: different way, should he have to play the whole thing over? (I
: imagine some kind of hybrid system would work best, but this si a
: programatical nightmare which I didn't have time to solve then.)
This seems kind of a minor issue, imho, although it would be nice to
have some way of setting up the variables of the game so that
you wouldn't have to replay the season just to find out what would have
happened in episode 7 if you solved a puzzle in episode 2 a different way
or made some other influencing decision.
: Even worse, though, you get a combinatoric explosion. Sure, you might
: want to allow everything the player does in episode 1 to influence
: thigns later on, but... Let's say that there are 20 castaways on
: moonbase alpha. There are 20 ways episode 2 can begin. Another
: castaway is voted off in episode two, so there's 20*19 ways episode 3
: can begin. This is bad mojo, since you've got an insanely large
: number of possibilities.
Not really. Simply encode 20 characters for each episode X and remove
X-1 characters from the game. Make the characters flexable enough that they
interact with any random person (ie, not hard coded to certain characters).
For instance, IE, A hates B, Likes C and D, wants to be more friendly with
E and is envious of F (Sorta like the Sims, I guess). That way, if
D is out of the game, A can still react normally and attempt to spend the
episode trying to get E into the alliance of C and D while trying to
influence others to vote against E (or whatever).
Of course, Towards the "end" of the season you would be encoding 20 characters
when only a couple are used in the game, so you might want to
end the season halfway, or something.
: Also, if you're going to pass information back and forth from one week
: to the next, you need to do a lot of planning ahead. If you decide
: that episode 61 is going to have elements in it that depend on what
: the player did in episode 7, you'd better darned well hope that
: episode 7 exported the relevant details. ANd if you've got to plan
: that far ahead, you start to lose the advantages of having not written
: the whole thing as one epic game.
Just wondering, but has anyone suggested 61 episodes? That seems excessive.
When I think of serial, I think 6-20 episodes, maybe.
Also, how hard would it be to export all the variables within
each game? It can't be that hard, surely. Of course, you might
not be able to export whether I solved puzzle fizzbang before puzzle
wallawalla, but it would be easy to set up to save how I solved
each puzzle, what choices I made, what happened to each object, etc.
: So, these are hte major issues I came up with. If people are
: interested in this sort of project (I, like you, I think, also
: envisioned this as something of an episodic TV paradigm), I'm still
: massively interested. I've got a few series ideas simmering on the
: back burner.
Some other ideas:
How about a series located in one setting, with different (or the same)
characters. For instance, a city is designed by the team, and rules are
set before hand, various objects (people?) are discussed and defined, etc.,
and then everyone designs their own story.
Another idea is that you store a few important details based on the small
details. For instance, if certain circumstances happen, a friendly
NPC will become resentful of you and will act different in future
stories, but the details/cause of the resentment will be lost (except to the
player, who will know. And if the details aren't needed, there's no point
for storing the data.) (and then 3 episodes later, maybe you have
an opportunity to appologize and repair that friendship).
-- Edan Harel
: This certainly does sound like a feasable idea, but it seems to be
: along the lines of a fan-fiction. The only difference being who
: establishes the world. Personally, I would tend to believe that unless
: we could make a world that was really really attractive, it would be
: difficult to attract writers until it was sufficiently fleshed out. So
: if, for example, two or three people would agree to this concept,
: write games in this shared world, and then let people know they can
: also make games in this world, then it might seem attractive. Though
: I'm still tempted to think that unless this world is Final Fantasy or
: Metal Gear people would still turn it down. You would really have to
: capture people's imaginations.
One plus this has is that once the encoding for the setting is done(once),
authors onlly have to concentrate on the characters and objects (and
maybe a few changes to the setting, like a secret passage, or the
inside of a house that wouldn't play a part in any other story), or
removing sections of the city that have no purpose in the game.
-- Edan Harel
: It's worth remembering that this kind of thing *has* been done in IF, in a
: sense; Sierra had the Kings Quest, Space Quest and Police Quest series, and
: LucasArts had the Monkey Island series. Of course they all used the first
Actually, many of them fell in the second group (ie, the games weren't
linear, but didn't carry over information to the next adventure, or
that non-linear stuff wasn't relevent later. I'm also fairly sure
there was some game which alllowed options at the end, but assumed
one of the ends for the following game.
The Hero's Quest/Quest for Glory series used 3. It would carry
over stuff like inventory and what kind of class you were in
and how much experience you had, etc. If you played through several
of the games, you could become a paladin. Plus, there were different
endings for different characters. And, of course, a number of RPG games
allow you to carry games over from one game to the next.
: I suppose it's different in the sense that these were all stand-alone,
: full-size games, but the series worked because while the games weren't
: necessarily linear in themselves, different players would have achieved
: essentially the same things at the end of each episode, regardless of the
: route taken. So the essential story is fixed, and can continue happily into
: the next part.
Just like tv shows. Some are fanatacal about continuity, and others can
be seen in any order.
: I think it's going to be very difficult to get away from this. Otherwise, if
: you let there be two materially different ways of ending the first episode,
: you're going to end up having to write two different stories for episode two,
: each for only a half of your likely audience. And why would you want to do
: that when you could write two new games that could be played by everyone?
Or the stories can branch back together again. Or maybe you're playing the
game from a different perspective. ie, in one perspective, you're working for
something and in another you're working against it, or as a double agent.
-- Edan Harel
[ . . . ]
> The Hero's Quest/Quest for Glory series used 3. It would carry
> over stuff like inventory and what kind of class you were in
> and how much experience you had, etc. If you played through several
> of the games, you could become a paladin. Plus, there were different
> endings for different characters. And, of course, a number of RPG games
> allow you to carry games over from one game to the next.
What I find interesting about QFG is that the final game in the series,
#5, *did* allow you to create a end-of-game character save file, just like
all the other games did. And the end-of-game character save file for #5
is *larger* than any of the other ones.
Each game saved, at least, which game it was from, and the character's
basic statistics. Some of them saved how much money you had. #3 saved
your inventory list, but that information wasn't used in #4, which started
you off with no inventory, anyway. (Presumably, the save file format for
3 was designed before they decided how to start #4. Which is something to
keep in mind: should they have plotted out that opening *before* they
designed #4? It wouldn't make sense to do so in a multi-author system.)
The end-of-game character save file for #5, though, saves information
about the game *world*, unlike all the other ones, which save information
only about the character himself. It saves the character's name, stats,
spells, and inventory, but it also saves the state of the world, based on
choices you've made. You had the option to ressurect one of two NPCs who
died in previous games, although you couldn't ressurect both without a
save-game editor. You could have gotten engaged to one of four NPCs (and
a bug in the game made it possible to get engaged to up to three of them
without resorting to saved-game editors -- you couldn't get engaged to all
four, because you could only ressurect one of them.) Several NPCs could
die in the final battle, including someone you were engaged to, but, if
you won quickly enough, it was just barely possible to keep everybody
alive. When you were offered the crown to rule the city, you could accept
or decline. And you could, possibly, become the Chief Thief of the
Now, they could have this much possible change to the world, because they
were never planning on making another Quest For Glory game. It *might* be
possible to write a game where all those differences showed up in the
introductory text, but didn't make a difference in gameplay. But it would
be tricky. (Why do I know this stuff off the top of my head? Because I
really like the world they've created in the Quest for Glory series, and
think that *somebody* should go ahead and write intfic continuing the
story. Even if it *is* a style of game that is out-of-fashon for intfic
these days. So I am. Don't tell Sierra. (although they seem to be fine
with fanfic written about the world, and this is just the same thing)).
But, I think that pretty much sums up one way of dealing with continuing
storylines: Quest for Glory is episodic. You're never in the same part of
the world twice in a row. You start out in a fantasy version of a
Barvarian village; the next game's in an Arabian Nights-type setting,
followed by an African savanah and jungle, then an Eastern European
village, and finally, a fantasy version of the Greek islands. The major
"quest" in each game is important enough to have ramifications in other
games, but none of the "optional quests" are things that anybody in other
parts of the world would have heard about.
I suppose it would be possible to write serial intfic on the same model.
And it would be pretty in genre for, say, pulp fiction.
"We could watch THE PRISONER and then watch TELETUBBIES!" -- my mother
These decisions, preferences and notions are based on a variety of
things such as personal taste (I.e. you can't have multiple endings
and then assume one of them is the 'true' one), percieved technical
limitations (do i have to code future games for all possible endings
of the previous games?), etc. It would be up to the developer(s) to
decide how to handle these things and address these challenges, but it
can certainly be done.
Okay, so you're playing Zork I, you have items in the trophy case,
you've killed monsters. You want to be able to save this state and
bring it into the 2nd game, so in the 2nd game you start out with
items in the trophy case, and the creatures that are dead remain dead.
I don't see why this couldn't be accomplished by writing one giant IF
and splitting it up into two or three parts. You'd have to 'trick' the
standard Z-Machine (if this what you're using) save game feature or
write your own, but in essence, I don't see why this can't be done.
Let's take an extreme example, you write something approaching an RPG
with character classes and random treasures that help you along. You
want to be able to save your class, any stats or skill levels, and
your treasure. You can do this easily in any standard game.
Now, let's say this same single game is divided into three parts. One
part is a quest into a forest, another through a series of caves, and
another involves a dragon and a heap of gold (sound familiar?). Why
not let the player save their state after the forest, and let them
take their character into the caves.
I have another idea along these lines, but I'd rather do it than talk
about it right now. Perhaps by saying the following without actually
stating what I'm thinking I'll inspire someone else to come up with
something equally interesting.
You're used to things that appear in serial format being linear
(books, movies, even games). You're used to IF that's non-linear.
There is an obvious, simple way to combine the two. I've been saying
it all along, but nobody's seeing it yet.
> I've received some interest in this idea, so I've set up a Yahoo!
> Group for this (sorry, no Majordomos at my disposal).
> The URL for this is: http://groups.yahoo.com/group/tsif
> or sign up by e-mailing: tsif-su...@yahoogroups.com
> Interested parties welcome. Kibbitzers will be fed to Grues.
I'm afraid I may have offended some people. While my goal for the list
was for conversation aimed specifically at this project, I think I may
have scared off some people who were unsure as to whether or not they
could find a place. I'd like to clarify the purpose of the e-mail
The e-mail list is for the discussion of ideas directly related to (a)
team developing, and (b) serial IF. I have a sense that there may be
team development on non-serial IF and single-developed serial IF
within this project as well.
I am looking for people who are interested in either serial IF _or_
team development of IF. This is akin to a research project. I want to
control the flow of information out of the team so we release our
findings formally rather than in a haphazard manner. Because of this,
I want to gather interested parties as formal members of the team,
rather than letting people join and drop out, join and lurk (and then
work on their own projects on the side), etc.
In poker a kibbitzer is an onlooker who comments without
participating. I'm looking for people who are willing to participate
both in the discussion and development of serial IF and team games.
Why both team and serial at the same time? Well, put simply, life is
short and serial game are long. I'd also like to see what team
development does to the cycle time and quality of IF. Also, it would
be interesting to see if traditionally lone coders could come together
in a manner approaching that of the development of a commercial
So, anyone intersted in being a formal member of the team, whether
your contribution is in ideas, code, or devil's advocacy, and is
willing to commit to the tenets of the team, you're welcome to join.
I was following up your argument that it would be difficult to write a
sequal game to which the first game had multiple endings.
I was (poorly) making the point that at any stage in a single IF there
could be multiple paths the player could take. I'll take Silent Hill
as an example. I played it recently so it's fresh in my mind
There are 5 different endings to Silent Hill. In fact, if you plotted
out the game you would find that there are even different versions of
cut scenes, and different ways of handling puzzles and certain fights
depending on things that happened earlier.
Let's use numbers to represent chapters and letters to represent
variations. Silent Hill might look something like this:
Chapter I (a)
Chapter II (a or b)
Chapter III (a or (b or c))
Chapter IV (a or e or (b or (c or d)))
Largely the gameplay is the same, after all, you're in the same town,
facing the same problems. However, certain aspects of puzzles or
certain cutscenes change. Perhaps there are even sections that are
very much different (I only played through to one of the multiple
endings, so I don't know.) If you had item X you could use it in
chapter IV to avoid the C situation, but you would have to get it in
chapter III. Metal Gear was this way too.
You could, in theory, have chapters 1 and 2 as a single game, and then
chapters III and IV as another game. If you'd plotted it out early
enough in advance, you could write game one before game two. The
advantage is that each game would be a full sized IF which:
isn't too big to frustrate most players
can get your full attention for the period of time you're writing it
can be released more quickly
can work as a stand alone (qv Star Wars)
You can also get feedback on Game 1 that can help you write Game 2.
Coding for game II doesn't become massively complex because you're
coding different versions of certain scenes only. If a puzzle can have
two different outputs, why can't it have two different inputs. I doubt
the makers of Silent Hill bothered to code very complicated variations
on the same events, just simple ones with new cut-scenes.
For example, here are two outputs:
- a locked door. you can remove the hinges or find the key
And here are two inputs:
- a locked door. Player has to find the key or choose the key from a
ring she found.
Trying to head you off at the pass, the keyring is a puzzle in itself,
and was also useful in game one, making it not seem arbitrary, simply
useful twice. To avoid extra coding, the keyring puzzle is very
similar to the missing key puzzle.
In either scenario the programmer has to come up with two solutions to
the puzzle. (I'm not asking you to consider the "two outputs" section,
just the "two inputs.")
Another alternative to the exploding plot idea is the imploding plot.
It goes something like this:
Chapter I (a)
Chapter II (a or b)
Chapter III (a or b or c)
Chapter IV (a or c)
Chapter V (a)
What happened here? Well, there is only one ending, or perhaps two or
three, but the player was allowed to have several paths to get there.
Let's take an X-files type game. There either are aliens or there
aren't. During the course of the game, however, you're allowed to
discover either a or b but not both. One of which may turn out to be a
The secret base is in Alaska
True, you're just delayed in going there.
The secret base is in Roswell.
False, Roswell is just a coverup for a secret base in Alaska, go there
Is either one disappointing? Perhaps. I would say it's all a matter of
execution. If done well, I would accept either one, as a long as I had
the feeling of forward motion the whole time. I guess there is the
chance that upon learning the plot strucure I would feel railroaded,
I'm simply offering an alternative to the "Big Bang" plot.
PS. The Team Serial IF experiment is going to get down to serious work
on Wednesday, after which point I'm going to close off the open
subscriptions to the group and allow subscription by request only.
Why? Well, individual developers have the benefit of working alone.
I'd like for us to be able to work undisturbed and without fear of the
plot 'leaking' out through some unscrupulous noseybody.
After this time, I also largely expect to concern myself with the
games we're writing and discuss the technical and theoretical merits
and pitfalls of serial development less and less here. (To some
people's chagrin and others' joy, I suspect.)
If you're interested in joining do so now, or contact me via e-mail