The TADS manual suggests some pretty careful planning before you start
coding; draw your map, figure out your main puzzles, anticipate most
The ALAN manual takes this quite a ways, too, suggesting that you write no
text at all (to speak of) until you've actually coded and walked through
your entire map.
In short, most of the authoring systems' authors appear to be in favor of a
_lot_ of planning and layout in advance. (One possible exception is Drew
What's His Name, the creator of Gamescape. He says he just jumps right in.
Gamescape doesn't seem to get any discussion in this newsgroup, probably
because it's an inferior subset of even AGT.)
Now, I understand the need for planning: it avoids holes in the plot and
gameplay, it achieves consistency, and it avoids difficult and time
consuming rework, thereby increasing the odds that the story will actually
But is there an inhibitor to creativity here? Many authors of novels state
that, despite their plans, the characters seem to take the novel places
which they had not anticipated. I too find that, as I begin to write, more
ideas suggest themselves and the stories go their own way.
So, what's the consensus? Plan your work and work your plan? Plan it but
then wing it? Or just wail away at the keyboard and see what happens?
(Side note: computer people, I've observed, tend to want to get the details
in advance- even if that is difficult. After all, a program won't work
unless the details are covered. It's a bit of a mindset, particularly among
>I'd like to ask what various authors' preferences might be; lots of
>planning ahead of time, jump "write" in and start coding; something in
You're probably right that the temptation to plan everything in advance
comes from a programmer's background.
I generally have an idea of what I want to have happen in the game before I
write it, but I don't agonize over specific details. I have a harder time
corralling my ideas into something coherent.
Leary talks about his "end-in" design style: he knows clearly what he wants
for the end and the beginning, and then fills in the middle. To a certain
extent, I do this too. But I do find that the details fill themselves in.
A big difference between static fiction and text adventures is that text
adventures have puzzles. (I still feel like puzzles are important, in that
they make the work enjoyable and accessible.) I've found it incredibly
hard to keep the puzzles from leading the whole story. I tend to have a
couple fully worked out puzzle ideas at the outset -- then I either have to
figure out a way to make the puzzles appropriate for the game, or adapt the
story to the puzzles. This probably weakens the works.
d...@ai.mit.edu MIT AI Lab NEW .sig Lite!
ADVENTIONS: We make Kuul text adventures! Email for a catalog of releases.
Meticulous plotting and planning has its disadvantages:
(a) it is deeply depressing to look at a simply enormous paper plan
which has no electronic substance at all yet, as this offers the
prospect of huge amounts of grinding implementation to do without
the fun of having ideas along the way;
(b) it locks you in to an initial view of how the game will play,
the need to preserve which may make you constrain the plot heavily.
My process goes roughly like this:
1. Basic idea
2. A little research, coding up some simple puzzles which you think
will go somewhere but don't yet know where
3. Division of the game up into a few chapters or stages, and of the
(as yet unwritten) map into broad regions
4. Fleshing out the regions more or less by improvisation, except that
by this stage one usually has library research to help out, and a
pile of odd notes and ideas put aside during the earlier stages
5. Tidying up and heavy redrafting: organising the plot better
7. A redraft (possibly a structural one) based on what the play-testers
My present game is hanging about in stage 4, as it has been for about
a year now, since I'm being a perfectionist for a change.
A game that's been planned ends up with lots of structure that's
perceivable to the player; there are premonitions, foreshadowings and so
on. Events later in the game make sense in terms of events early in the
game. These kinds of things are what turn an adventure game into a work
of interactive fiction.
On the other hand, as Graham Nelson pointed out, It can be quite
intimidating to be presented with a big plan that has to be implemented.
Too much drudgery makes for a dull game -- it's awful to sit down and
think, "I've got to write fifty room descriptions today, and each one of
them has to be clear, crisp and vivid while conveying exactly the
information I want it to convey". In these circumstances bad prose is
what results. A game that's been over-planned can end up too structured
Jumping in and starting to implement puzzles right away generates
valuable spontaneity and the prose that I produce in this circumstance
tends to be of better quality. But without an overall plan all I end up
with is a set of linked puzzles.
Maybe a combination of both approaches -- do a little planning, start
implementing some of the bits you've worked out, use that implementing
experience to modify the plan, and so on -- produces the best results.