What makes a game linear?

5 views
Skip to first unread message

Russell Glasser

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

I notice there are a lot of "meta" game discussions going on lately
in rec.arts.int-fiction -- about one authoring system vs. another, about
the organization of this year's competition, about Mark Blank's
anticipated return to authoring, etc. But it seems like it's been a
while since I've seen a really solid conversation about real game
issues, so this I think this is a good time to start a thread about
something that's bugging me.
Namely, how important is it to avoid linearity in writing a game;
and how do you know when you've succeeded in being nonlinear?
I started thinking about this topic just now when I looked up the
review of my "first stupid game" from last year, _Reverberations_, in
Baf's guide. Among other things it said the game had a "linear plot".
That worried me. While I was writing Reverberations, I tried very
hard to make it at least SEEM nonlinear by allowing the three middle
puzzles (bomb, department store, Mayor's file) to be encountered and
solved nonsequentially. As a result, the game was much less linear
than, say, "Toonesia" (which I enjoyed a lot, but it was conspicuously
placed in one-puzzle-at-a-time format). But "Reverb" was very short,
and plot considerations obligated me to split events a distinct
beginning, middle and end sequence. The branching paths really lead
back to the same conclusion in a very short time, so in a sense the
nonlinearity is a sham.
Furthermore, my new competition entry (which I shall refrain from
naming or describing, in accordance with the rules) is shaping up to be
similar in structure. It has a completely linear beginning, a slightly
complex middle with some unordered puzzles, and a final inescapable
showdown at the end.
The question is, how linear is linear? And, how is it possible to
break down the order of game elements and still retain a coherent plot?
Storyline is very important to me in a game, so how can I approach a
game with a general sequence of events and still have some disorder to
it?

--
"The reasonable man adapts himself to the world; the unreasonable one
persists in trying to adapt the world to himself. Therefore all
progress depends on the unreasonable man."
-- George Bernard Shaw

Russell can be heckled at
http://www.willynet.com/rglasser

Neil Brown

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

** Warning: Christminster spoilers at the bottom **

In article <33CC7858...@penning.lanl.gov>, Russell Glasser


<rgla...@penning.lanl.gov> wrote:
> I notice there are a lot of "meta" game discussions going on lately
>in rec.arts.int-fiction -- about one authoring system vs. another, about
>the organization of this year's competition, about Mark Blank's
>anticipated return to authoring, etc. But it seems like it's been a
>while since I've seen a really solid conversation about real game
>issues, so this I think this is a good time to start a thread about
>something that's bugging me.
> Namely, how important is it to avoid linearity in writing a game;
>and how do you know when you've succeeded in being nonlinear?

If a game was totally non-linear, it would have no specific plot at all,
and it would be the job of the player to go around and find interesting
things to do. In my opinion, its best to allow a certain amount of
freedom in the order the puzzles are tackled, and different ways of
solving the same puzzle if possible. It's frustrating when a game forces
a player to solve ABCDEFG in that order. The good thing about the start
of Curses is that if you get stuck on one puzzle, there are still
several others you could be getting on with. If an order is forced, then
the moment a player gets stuck on a puzzle, s/he either loses interest
or reaches for the walkthrough.

> I started thinking about this topic just now when I looked up the
>review of my "first stupid game" from last year, _Reverberations_, in
>Baf's guide. Among other things it said the game had a "linear plot".
> That worried me. While I was writing Reverberations, I tried very
>hard to make it at least SEEM nonlinear by allowing the three middle
>puzzles (bomb, department store, Mayor's file) to be encountered and
>solved nonsequentially. As a result, the game was much less linear
>than, say, "Toonesia" (which I enjoyed a lot, but it was conspicuously
>placed in one-puzzle-at-a-time format). But "Reverb" was very short,
>and plot considerations obligated me to split events a distinct
>beginning, middle and end sequence. The branching paths really lead
>back to the same conclusion in a very short time, so in a sense the
>nonlinearity is a sham.
> Furthermore, my new competition entry (which I shall refrain from
>naming or describing, in accordance with the rules) is shaping up to be
>similar in structure. It has a completely linear beginning, a slightly
>complex middle with some unordered puzzles, and a final inescapable
>showdown at the end.
> The question is, how linear is linear? And, how is it possible to
>break down the order of game elements and still retain a coherent plot?
>Storyline is very important to me in a game, so how can I approach a
>game with a general sequence of events and still have some disorder to
>it?

I've been struggling with this problem as well. I'm working on something
(very slowly) which is based on a story I wrote some time ago.
Obviously, the story relies on certain things happening in a set order,
which isn't really ideal for IF. Christminster deals with this by
waiting for the player solve a certain puzzle or visit a certain
location before advancing the plot (and game time). For example, when
the player finds Malcolm's lab book, we are suddenly introduced to the
villains of the piece, and learn a little of their plans.

A way of approaching this would be to make a list of the most important
events in the plot - those that MUST happen, no matter what. Then, try
to allow the player a little freedom in reaching these.

For example, to get from A to D, you could go via B or via C. Or maybe C
followed by B, or vice versa. Perhaps choosing pathway C would limit
your choice of pathways when travelling from D to G.
______________

Neil James Brown
ne...@highmount.demon.co.uk
http://www.highmount.demon.co.uk

Lucian Paul Smith

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

Russell Glasser (rgla...@penning.lanl.gov) wrote:
: Namely, how important is it to avoid linearity in writing a game;

: and how do you know when you've succeeded in being nonlinear?

Now sounds like a good time to hype the IF Patterns pages again; I've
started a page devoted to just this concept there, at:

http://c2.com/cgi/wiki?IntFicLinearityVsBreadth

Some discussion there is also based on a discussion I had with Neil
DeMause, Dan Shiovitz, and Liza Daly on the ifMUD
(http://fovea.retina.net:4001/).

What we ended up talking a lot about was not how to make games less
linear, but how to _disguise_ a game's linearity. In order to tell a
convincing story, certain things probably have to happen before certain
other things. The trick is to let the player discover these events
happening in order _without_ letting them realize that there's not
actually another order for the events to happen.

One way to have this happen is to have a few 'paths' that don't pan out.
The player may see five puzzles before them, solve one, and continue down
the plotline that that puzzle opens, and never realize (until afterwards)
that the other four puzzles are not actually solvable. I-0 was held as
an example of one game which, (to Neil, at least) although it contained
many different paths, also contained the _illusion_ of having more paths
than it did. 'So Far' worked for me in this regard, as well, with its
locked doors and glimpses of areas beyond the Mall area, among other things.

So, in the end, the player should feel like they are choosing their own
path through a wide field, not realizing that they are actually walking
down an alley, with a field painted on the walls.

Tricky to do right, (and now would be a good time to throw out
suggestions--anyone?), but very satisfying when done well.

-Lucian "Lucian" Smith

Matthew T. Russotto

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

In article <erkyrathE...@netcom.com>,
Andrew Plotkin <erky...@netcom.com> wrote:
}
}(There may also be multiple endings, of course. Or even multiple
}beginnings. (Anyone ever try that approach? For one example, a RPG-style
}game where you can start out a warrior, a wizard, or a thief, in three
}different introductions, and then then play through the same storyline
}with your particular skills.))

Legend pretended to do that, but then blew it, with "Companions of
Xanth". You could supposedly choose one of four companions. While
there was one obvious bad choice, the rest seemed reasonable enough --
and if you've read some of the books, even the bad choice seemed
plausible. But in fact, only one of the choices got you through the
first puzzle. Very disappoining, IMO.
--
Matthew T. Russotto russ...@pond.com
"Extremism in defense of liberty is no vice, and moderation in pursuit
of justice is no virtue."

Paul O'Brian

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

Well, one factor to consider is that (in my opinion) it's a lot
more difficult to make short games non-linear. Something like Zork or
Trinity can open up many different pathways because of its sheer size,
while a game that wants to be solvable in two hours is a little more
constrained.

I struggled with this problem a lot in "Wearing The Claw", and after an
informal raif poll decided to go ahead and leave the linearity (since
fixing it would have required basically restarting from scratch) and
compensating for it by providing a comprehensive hint system and lots of
detail in the individual scenes.

Of course, if I enter the competition again I'll make it a goal to have a
much less linear design. (That and choose a different genre! I got a *lot*
more complaints about that than I did about linearity!)

Did others who entered the competition feel that it was more difficult to
avoid a linear plotline in the short form?

Paul O'Brian obr...@ucsu.colorado.edu
"This is the kind of spontaneous publicity -- your name in print -- that
makes people! Things are going to start happening to me now!"
-- Steve Martin as Navin R. Johnson (The Jerk)

Carl Muckenhoupt

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

Matthew T. Russotto wrote:
>
> In article <erkyrathE...@netcom.com>,
> Andrew Plotkin <erky...@netcom.com> wrote:
> }
> }(There may also be multiple endings, of course. Or even multiple
> }beginnings. (Anyone ever try that approach? For one example, a RPG-style
> }game where you can start out a warrior, a wizard, or a thief, in three
> }different introductions, and then then play through the same storyline
> }with your particular skills.))
>
> Legend pretended to do that, but then blew it, with "Companions of
> Xanth". You could supposedly choose one of four companions. While
> there was one obvious bad choice, the rest seemed reasonable enough --
> and if you've read some of the books, even the bad choice seemed
> plausible. But in fact, only one of the choices got you through the
> first puzzle. Very disappoining, IMO.

Agreed. Piers Anthony wrote a book about a computer game, and Legend
made the poor choice of making a game based on the book, rather than a
game based on the game described in the book.

Sierra, on the other hand, did it perfectly in Quest for Glory 1. The
choice is exactly what Mr. Plotkin suggests (warrior, wizard, or thief),
and affects just about every puzzle in the game. Later games in the QfG
series made the choice less relevant - you wound up doing most things
the same way regardless.

--
Carl Muckenhoupt ca...@earthweb.com
EarthWeb http://www.earthweb.com/

Carl Muckenhoupt

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

Neil Brown wrote:
>
> ** Warning: Christminster spoilers at the bottom **

And now some Reverberations spoilers as well, to keep them company.

>
> In article <33CC7858...@penning.lanl.gov>, Russell Glasser
> <rgla...@penning.lanl.gov> wrote:
>
> > I started thinking about this topic just now when I looked up the
> >review of my "first stupid game" from last year, _Reverberations_, in
> >Baf's guide. Among other things it said the game had a "linear plot".
> > That worried me. While I was writing Reverberations, I tried very
> >hard to make it at least SEEM nonlinear by allowing the three middle
> >puzzles (bomb, department store, Mayor's file) to be encountered and
> >solved nonsequentially.

Really? Hmm, I'm going to have to play it again and rewrite my review.

> > As a result, the game was much less linear
> >than, say, "Toonesia" (which I enjoyed a lot, but it was conspicuously
> >placed in one-puzzle-at-a-time format). But "Reverb" was very short,
> >and plot considerations obligated me to split events a distinct
> >beginning, middle and end sequence. The branching paths really lead
> >back to the same conclusion in a very short time, so in a sense the
> >nonlinearity is a sham.

Interesting that you should bring up Toonesia. I think of Toonesia as
something that presents an illusion of nonlinearity, despite being
entirely linear in terms of its puzzles. It accomplishes this by
allowing you access to puzzles you can't solve yet. You find the mine
entrance fairly early in the game, even though you can't solve the
puzzle of entering it until near the end. This means that, through most
of the game, you know that you can always go and try to deal with the
mine instead. (You won't be able to solve it, of course, but you don't
know that yet.) This illusion of choice decreases the perceived
linearity of the game.

Reverberations is the exact opposite. When I played it, I don't think I
knew that I had a choice of three subquests. (Hence the review.)
Nonlinearity with an illusion of linearity, perhaps? The impression of
nonlinearity is given by the knowledge that there's something else you
could be doing. Two of the three midgame subquests in Reverberations
are surprises that, once you find them, you must deal with immediately
or perish. Only the Mayor's files gives you any advance warning, and
even there, it's solve or die - you can't walk away from the situation
half-finished and explore somewhere else. This is not linearity, but,
lacking a sense of the big picture, it can seem like it.

In addition, there's the matter of adding things to the map.
Reverberations only cares about the ordering of two plot elements: the
beginning must come first, and the ending must come last. But, since the
entire game has the same geography, the only way to enforce this is to
make changes to what is present. The bomb, the hitman, and the DA in her
office are not there in the beginning, and are added when the plot
requires that they be there. This is a technique commonly used to
artificially enforce a linear plot - perhaps used in its most extreme
form by Sierra in such games as Quest for Glory 2 and Gabriel Knight. If
the bomb only appeared after you raided the Mayor's office and the
hitman only appeared after the bomb failed (or something like that), the
game would be linear indeed (and a lot more Sierra-like.) This isn't the
case, but does the player know?

The thing is, Reverberations works. I had assumed that this was because
of its size; small games can generally get away with things that would
become tedious or annoying if carried out over a greater length of time.
The amazing thing to me is that it *seemed* like the three subquests
formed an ordered progression, in terms of escalation of personal
danger. Perhaps I'm a bit too quick to see patterns where there are
none?

> > Furthermore, my new competition entry (which I shall refrain from
> >naming or describing, in accordance with the rules) is shaping up to be
> >similar in structure. It has a completely linear beginning, a slightly
> >complex middle with some unordered puzzles, and a final inescapable
> >showdown at the end.

The classical style, as described by Mr. Nelson in "The Craft of
Adventure". Nothing wrong with that at all.

> > The question is, how linear is linear? And, how is it possible to
> >break down the order of game elements and still retain a coherent plot?
> >Storyline is very important to me in a game, so how can I approach a
> >game with a general sequence of events and still have some disorder to
> >it?
>
> I've been struggling with this problem as well. I'm working on something
> (very slowly) which is based on a story I wrote some time ago.
> Obviously, the story relies on certain things happening in a set order,
> which isn't really ideal for IF. Christminster deals with this by
> waiting for the player solve a certain puzzle or visit a certain
> location before advancing the plot (and game time). For example, when
> the player finds Malcolm's lab book, we are suddenly introduced to the
> villains of the piece, and learn a little of their plans.

Christminster is an good example of a game that has a linear plot but a
highly nonlinear game structure. While some events can only happen in a
set sequence and are linked to times of day, others (chiefly those
leading up to the creation of the elixir) are available to be solved
throughout much of the game. This suggests a formula for introducing
nonlinearity while retaining plot: put your plot on two or more parallel
tracks. For example, perhaps the player is trying to solve the
mysteries of an alien planet while also trying to repair the spaceship
that is the only way home. This gives you two linear plot threads to
work with, while at the same time retaining a certain amount of freedom.
At crucial points, the plot threads can cross.

Andrew Plotkin

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

Russell Glasser (rgla...@penning.lanl.gov) wrote:

> Namely, how important is it to avoid linearity in writing a game;
> and how do you know when you've succeeded in being nonlinear?

> I started thinking about this topic just now when I looked up the
> review of my "first stupid game" from last year, _Reverberations_, in
> Baf's guide. Among other things it said the game had a "linear plot".
> That worried me. While I was writing Reverberations, I tried very
> hard to make it at least SEEM nonlinear by allowing the three middle
> puzzles (bomb, department store, Mayor's file) to be encountered and

> solved nonsequentially. As a result, the game was much less linear


> than, say, "Toonesia" (which I enjoyed a lot, but it was conspicuously
> placed in one-puzzle-at-a-time format). But "Reverb" was very short,
> and plot considerations obligated me to split events a distinct
> beginning, middle and end sequence. The branching paths really lead
> back to the same conclusion in a very short time, so in a sense the
> nonlinearity is a sham.

> Furthermore, my new competition entry (which I shall refrain from
> naming or describing, in accordance with the rules) is shaping up to be
> similar in structure. It has a completely linear beginning, a slightly
> complex middle with some unordered puzzles, and a final inescapable
> showdown at the end.

This is a very common structure. In fact I usually think of just about
all games with this structure; the difference is how long the beginning
and ending segments are, and how complex the middle is. (In extreme
cases, the beginning (ending) is linear only for the length of the
introductory (closing) text. But this is still a fixed starting/ending
point. In other extreme cases, the middle is not very unordered at all,
or even pretty much linear. But still trivially nonlinear in the sense
that you can walk around however you like and pick up objects in
different orders.)

(There may also be multiple endings, of course. Or even multiple
beginnings. (Anyone ever try that approach? For one example, a RPG-style
game where you can start out a warrior, a wizard, or a thief, in three
different introductions, and then then play through the same storyline
with your particular skills.))

> The question is, how linear is linear? And, how is it possible to


> break down the order of game elements and still retain a coherent plot?
> Storyline is very important to me in a game, so how can I approach a
> game with a general sequence of events and still have some disorder to
> it?

Obvious possibilities:

* Have story elements that don't care what order they're in. If the quest
of the hero to slaughter whatsisname includes finding the holy potato,
the magic watermelon, and the almighty avocado, you could probably write
those three scenes so that order is irrelevant. This is less applicable
to character development stories :-) but maybe not entirely.

* Have story elements that aren't closely coupled to the game elements.
After each puzzle, the mad scientist shows up and yells at you for a
while (and maybe you yell back), but the order of the conversations is
the same, regardless of what order you solve the puzzles in. Drawbacks:
you can't converse about what you saw in a particular puzzle.

* Fake non-linearity, where you can explore a large fraction of the game
(working "around" puzzles in some sense) before going back and solving
things. This is more important than it sounds, I just bet. Often when
people say "linear", they mean "Every time I solve a puzzle, I gain
access to a new set of rooms containing a new puzzle, and the old rooms
become irrelevant (or inaccessible.)" If you let the players at all the
rooms right at the beginning, it reduces this sense of linearity, even if
the puzzles must be solved in a fixed order.

* Fake solutions, where you solve things in several different orders, but
the only way that actually lets you win has a fixed order. (This requires
a slightly weird attitude, in which you-the-author accept that the player
*can* see plot events out of order, but you rely on the fact that it will
all come together correctly when he wins.)

--Z


--

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the
borogoves..."

Russell Glasser

unread,
Jul 16, 1997, 3:00:00 AM7/16/97
to

Andrew Plotkin wrote:

> (There may also be multiple endings, of course. Or even multiple
> beginnings. (Anyone ever try that approach? For one example, a
> RPG-style
> game where you can start out a warrior, a wizard, or a thief, in three

>
> different introductions, and then then play through the same storyline

>
> with your particular skills.))
>

Actually, yes: this is basically the direction of "Hot Cross
Doubles", my first big game. Actually you might say that the game has
multiple middles followed by a choice of endings, as well as multiple
player characters. But I don't want to give anything away yet.

> * Have story elements that don't care what order they're in. If the
> quest
> of the hero to slaughter whatsisname includes finding the holy potato,

>
> the magic watermelon, and the almighty avocado, you could probably
> write
> those three scenes so that order is irrelevant. This is less
> applicable
> to character development stories :-) but maybe not entirely.
>

I thought about this after I posted, and realized that old games
like Zork can get away with almost total nonlinearity because they have
neither plot nor character interaction. (Okay, they do have SOME
character interaction... "You dodge as the troll comes in low!" But
whacking a character upside the head with a sword isn't exactly the kind
of interaction I meant.) But the thing is, Zorks (and Enchanters, in
a way) are unstructured easter egg hunts. That doesn't work as a
storytelling device.

> * Have story elements that aren't closely coupled to the game
> elements.
> After each puzzle, the mad scientist shows up and yells at you for a
> while (and maybe you yell back), but the order of the conversations is

>
> the same, regardless of what order you solve the puzzles in.
> Drawbacks:
> you can't converse about what you saw in a particular puzzle.
>

I never thought of that approach... I kind of like it.

> * Fake non-linearity, where you can explore a large fraction of the
> game
> (working "around" puzzles in some sense) before going back and solving

>
> things. This is more important than it sounds, I just bet. Often when
> people say "linear", they mean "Every time I solve a puzzle, I gain
> access to a new set of rooms containing a new puzzle, and the old
> rooms
> become irrelevant (or inaccessible.)" If you let the players at all
> the
> rooms right at the beginning, it reduces this sense of linearity, even

> if
> the puzzles must be solved in a fixed order.
>

Path to Fortune, anyone?

> * Fake solutions, where you solve things in several different orders,
> but
> the only way that actually lets you win has a fixed order. (This
> requires
> a slightly weird attitude, in which you-the-author accept that the
> player
> *can* see plot events out of order, but you rely on the fact that it
> will
> all come together correctly when he wins.)
>

Change in the Weather, anyone?

> --Z

Russell Glasser

unread,
Jul 17, 1997, 3:00:00 AM7/17/97
to

Lucian Paul Smith wrote:

> Andrew Plotkin (erky...@netcom.com) wrote:
> : Russell Glasser (rgla...@penning.lanl.gov) wrote:
> :>> Andrew Plotkin (erky...@netcom.com) wrote:
>
> :>> * Fake solutions, where you solve things in several different


> orders, but
> :>> the only way that actually lets you win has a fixed order. (This
> requires
> :>> a slightly weird attitude, in which you-the-author accept that the
> player
> :>> *can* see plot events out of order, but you rely on the fact that
> it
> :>> will all come together correctly when he wins.)
>
> :> Change in the Weather, anyone?
>

> : Actually, no. Same technique used on different materials, for a
> different
> : purpose. In "Weather", you can take certain *actions* in several
> : different orders, but it's all in a single plot event. Your actions
> : change the course of that event. It has nothing to do with arranging
> a
> : set of plot events in several different orders.
>
> Actually, yes--depending on how you define 'plot event'. Personally,
> I'd
> define a 'plot event' quite loosely as 'something that you do which
> has a
> palpable effect on the game world'. (Or, I suppose, something which
> happens at a set time regardless of what you do which has an effect on
>
> the game world--like the earthquake in Zork III.)
>
> **** (small) SPOILERS**** for Weather and So Far below,...
>

Larger spoilers for the same have been added.

> Under this definition, there isn't a whole heckuva big difference
> between
> crossing the river in So Far and pushing the boulder in Weather. Both
>
> must be done at a certain time and in a certain way or else the game
> devolves into an unwinable state. However, this doesn't prevent the
> player from encountering and performing other 'plot
> events'--navigating
> the darkness in So Far; getting the sandbags in Weather. They'll just
>
> have to go back and perform them in the correct order before the
> entire
> plot is laid bare (so to speak).
>
> -Lucian "Lucian" Smith

Right, my point exactly. Although you CAN push the boulder at any
moment in the game, if you do it at any time out of sequence then you've
effectively lost. Similarly, there are three things you have to do to
save the bridge; you could do them in different orders, but the timing
of the game is such that you really only have one possible order for
doing that... if I remember correctly, it's boulder, sandbags, branch in
that order. If you do it any other way, then you don't get a long
enough delay to walk everywhere and do everything.
So this adds a new twist to the discussion. "Weather" is mostly
linear because you have to do everything in order, but it doesn't seem
linear. "Reverberations" is nonlinear because there are several things
you can do out of order, but it seems linear.
Which brings up the new question, is it more important to achieve
ACTUAL nonlinearity, or the SEMBLANCE of nonlinearity?

Lucian Paul Smith

unread,
Jul 17, 1997, 3:00:00 AM7/17/97
to

Andrew Plotkin (erky...@netcom.com) wrote:
: Russell Glasser (rgla...@penning.lanl.gov) wrote:
:>> Andrew Plotkin (erky...@netcom.com) wrote:

:>> * Fake solutions, where you solve things in several different orders, but
:>> the only way that actually lets you win has a fixed order. (This requires
:>> a slightly weird attitude, in which you-the-author accept that the player
:>> *can* see plot events out of order, but you rely on the fact that it
:>> will all come together correctly when he wins.)

:> Change in the Weather, anyone?

: Actually, no. Same technique used on different materials, for a different
: purpose. In "Weather", you can take certain *actions* in several
: different orders, but it's all in a single plot event. Your actions
: change the course of that event. It has nothing to do with arranging a
: set of plot events in several different orders.

Actually, yes--depending on how you define 'plot event'. Personally, I'd
define a 'plot event' quite loosely as 'something that you do which has a
palpable effect on the game world'. (Or, I suppose, something which
happens at a set time regardless of what you do which has an effect on
the game world--like the earthquake in Zork III.)

**** (small) SPOILERS**** for Weather and So Far below,...

Under this definition, there isn't a whole heckuva big difference between

Branko Collin

unread,
Jul 17, 1997, 3:00:00 AM7/17/97
to

On Wed, 16 Jul 1997 15:17:32 -0600, Paul O'Brian <obr...@ucsu.Colorado.EDU> wrote

>Well, one factor to consider is that (in my opinion) it's a lot
>more difficult to make short games non-linear. Something like Zork or
>Trinity can open up many different pathways because of its sheer size,
>while a game that wants to be solvable in two hours is a little more
>constrained.

Isn't Zork basically 'hunt the treasure'? In that case you can make it as
non-linear as you want. What does it matter in what order you get the
gold? Wearing the Claw is actually much suited to be a linear game,
because it is a quest. The hero has to solve a number of set problems. It
wouldn't be heroic if there were short cuts.

>I struggled with this problem a lot in "Wearing The Claw", and after an
>informal raif poll decided to go ahead and leave the linearity (since
>fixing it would have required basically restarting from scratch) and
>compensating for it by providing a comprehensive hint system and lots of
>detail in the individual scenes.
>
>Of course, if I enter the competition again I'll make it a goal to have a
>much less linear design. (That and choose a different genre! I got a *lot*
>more complaints about that than I did about linearity!)

:-). I remember writing in my jury 'diary' about Stargazer: "Despite it
being a fantasy game..."

--
Branko Collin - col...@xs4all.nl
"Ik weet niet wat u aan het doen bent, maar wilt u
het alstublieft buiten de vaargeul doen?"
uit: Sylvain.


Giles Boutel

unread,
Jul 18, 1997, 3:00:00 AM7/18/97
to


Paul O'Brian <obr...@ucsu.Colorado.EDU> wrote in article
<Pine.GSO.3.96.97071...@ucsu.Colorado.EDU>...

>
> Of course, if I enter the competition again I'll make it a goal to have a
> much less linear design. (That and choose a different genre! I got a
*lot*
> more complaints about that than I did about linearity!)

Generic Fantasy Mileu, right? I have to admit I have a theory on how to
beat that kind of prejudice...


> Did others who entered the competition feel that it was more difficult to
> avoid a linear plotline in the short form?

start - middle - endgame for me, but with a few twists.

I used a completely un-interactive start-up (it looked interactive, but I
lied). After that my game was divided into conceptual sections (sounds more
impressive than it ended up, he said with an appealing air of
self-deprecation) with objects from some sections being useful in others.
Some sections were better hidden than others, though they were all
accessible more or less from the start (except for the endgame, obviously).
Some puzzles could be solved with what you had in the first room, or indeed
with nothing, others required a bit of prior puzzle solving before the
means became available - but this was necessary to prevent to endgame
becoming accessible at the wrong time.

This assortment of predicaments was probably caused by the fact that I made
most of it up as I went along (over about 6 months). The only thing I found
hard to avoid was new ways of procrastinating doing any work on it. In a
way, I quite like that 'system' - some puzzles unrelated to others which
could be solved at any time, and a narrative puzzle thread to ensure the
plot keeps moving.

And while we're not talking about our upcoming games, the next one will be
completely different. And spellchecked. Honest.

-Giles

JC

unread,
Jul 18, 1997, 3:00:00 AM7/18/97
to

On Thu, 17 Jul 1997 12:46:19 -0700, Russell Glasser
<rgla...@penning.lanl.gov> wrote:

[...]


>Which brings up the new question, is it more important to achieve
>ACTUAL nonlinearity, or the SEMBLANCE of nonlinearity?

(Note: how linear something is, or how linear something appears to be,
are not boolean values, but span across a broad spectrum.)

You say that as if they are somehow mutually exclusive. How non-linear the
game *seems* to the player is most important. But, typically, there is a
close relationship between how non-linear a game seems and how non-linear
it actually is.

Techniques can be used which will enhance the sense of non-linearity. And
of course, there can be linear games which use these techniques that can
appear non-linear, but only to an extent.

In summary, the most important thing is the sense of non-linearity the
player gets, which can be created through a mechanism of the game being
non-linear along with a variety of techniques to make it seem moreso.

FemaleDeer

unread,
Jul 19, 1997, 3:00:00 AM7/19/97
to

Appearance of non-linearity is all one can aim for. A totally non-linear
game wouldn't have a plot.

On the other hand, there is really nothing wrong with a linear game.

Think of Bureaucracy, one of my favorite Infocom games (partly because it
took me so long to finish it, lacking the documentation). Despite the fact
you can do several things on the street in different order, it is
essentially a linear game. So It only has the "appearance" of being
non-linear on the street (and only for awhile, you do have to end up at
the paranoid's house).

And the llama food delivery man, if you don't get the llama food from him
at the beginning of the game, you cannot win. But there is no warning of
this, something many consider unfair. In fact, Bureaucracy broke several
"taboos", but I still found it a fun (annoying, but fun) game.

All rules are made to be broken. <Crack> (the sound of a "rule" breaking)

FD
------------------------------------------------------------------------------
Femal...@aol.com The Tame Computer
"Loyalty to petrified opinion never yet broke a chain or
freed a human soul." Mark Twain (or won a game)

C.E. Forman

unread,
Jul 20, 1997, 3:00:00 AM7/20/97
to

"Giles Boutel" <bout...@wcc.govt.nz> wrote:
>Generic Fantasy Mileu, right? I have to admit I have a theory on how to
>beat that kind of prejudice...

Enter it in the I-F competition under a pseudonym, posing as a
previously unpublished female author?

--
C.E. Forman cefo...@worldnet.att.net
Author of "Delusions", the 3rd place winner in the 1996 I-F Competition!!
Release 3 is now at: ftp://ftp.gmd.de/if-archive/games/infocom/delusns.z5
Ye Olde Infocomme Shoppe http://netnow.micron.net/~jgoemmer/infoshop.html

Kenneth Albanowski

unread,
Jul 22, 1997, 3:00:00 AM7/22/97
to

In article <33CD4D...@earthweb.com>,

Carl Muckenhoupt <ca...@earthweb.com> wrote:
>Sierra, on the other hand, did it perfectly in Quest for Glory 1.

Yes, there's another one of those classic games that haven't been repeated
(even, or indeed especially, by Sierra) The Mage practice "arena", and the
magic endgame are unique, as far as I know. Oh, and I seem to remember a
music box that plays one of the Turkish March pieces, which I personally
think is the mark of a truly great game. ;-)

>The choice is exactly what Mr. Plotkin suggests (warrior, wizard, or
>thief), and affects just about every puzzle in the game. Later games in the
>QfG series made the choice less relevant - you wound up doing most things
>the same way regardless.

Actually, there was a fourth in each (?) of the games: Paladin. You could
only get to be a paladin by choosing to be a fighter, and then always
biasing your choices (what was it, ethics over all? No, I think it was law
over all.) If you constrained yourself perfectly, at the end you got told
you were a paladin.

This has the interesting result of being a choice that _doesn't_ constrain
the player (as they are free to make the wrong choice at any time, and end
up only a fighter), but does provide yet another avenue for replay.

--
Kenneth Albanowski (kja...@kjahds.com)

James B. D'Arcy

unread,
Jul 23, 1997, 3:00:00 AM7/23/97
to

In article <5r1civ$f...@kjahds.com>,

Heh heh heh....

I got to be a paladin in Quest for Glory II by playing the thief character.
Each class had a different ethical code for racking up "honor points". With
the thief I think saying "thank you" a lot and not doing any housebreaking
was enough to win your paladinhood.

In fact, there was nothing one basic class could do that another class
couldn't copy, save the fighter's shield and sword and one or two magic
spells. The game played differently if you used, say, a magic-using thief
vs. a non-magic user, which added even more variance to the plot.

--
Ethan d'Arcy (eda...@hotmail.com)

J. Kerr

unread,
Jul 28, 1997, 3:00:00 AM7/28/97
to

James B. D'Arcy <jde...@ix.netcom.com> wrote:
>I got to be a paladin in Quest for Glory II by playing the thief character.
>Each class had a different ethical code for racking up "honor points". With
>the thief I think saying "thank you" a lot and not doing any housebreaking
>was enough to win your paladinhood.

I tried to be as ethical and virtuous as possible, and *never*
attained paladinhood. Never mind, I preferred housebreaking anyway.
--------------------
byw...@zetnet.co.uk
--------------------


Nick

unread,
Jul 29, 1997, 3:00:00 AM7/29/97
to

On 19 Jul 1997, FemaleDeer wrote:
> And the llama food delivery man, if you don't get the llama food from him
> at the beginning of the game, you cannot win. But there is no warning of
> this, something many consider unfair. In fact, Bureaucracy broke several
> "taboos", but I still found it a fun (annoying, but fun) game.
<snip>
I felt that one of the purposes of Beurocracy was to vent DAs frustration
at the fact that beurocracy is unfair, pointless and frustrating. I think
he felt that it was acceptable to include this sort of thing to make the
point...
Nick


Reply all
Reply to author
Forward
0 new messages