<movie review: memento, implications for IF?>
David Samuel Myers
to think it has some interesting implications on some aspects of IF game
design, which is why I'm bothering with this. However, I might spoil a
little bit of the structuring of the movie, so if you are going to be
very unhappy about that, don't read the footnotes, which will contain the
really spoilery stuff, and will be stuck way way way down in this post.
What follows is rambly rumination. I don't claim for it to make much
sense, nor for it to be very profound. It will make a heckuva lot more
sense if you go watch the movie, though.
This is a great little movie about memory tricks, and a great big riff on
time-reordering in storytelling. [1] Much tighter than the time-jiggering
in Pulp Fiction, and so much more elegant than the time-repeating type,
like Groundhog Day or Run Lola Run. And different still than a p.o.v. riff
like in Kurosawa's stuff in the 50s. Really, it is more like something out
of Philip K. Dick.
You'll have to pay pretty close attention to get the most out of a single
viewing of the film. Moreso than Usual Suspects (a movie that some people
are confused about), this is going to have a lot of people really
confused. Of course, this is the point.
At the heart of it is a character that has trouble making new memories.
He's not amnesiac, nor an Alzheimer's patient. [2] A made-for-movie
concoction of another sort. In any case, the craziness he has to go
through in order to get through a single scene, unable to keep any of his
short term memory down, is delightful to watch. The experience of having a
PC that faces similar predicaments should be imaginable to many readers.
It's tantalizing enough so that...
I think if someone could actually write the IF game version of this movie,
it would be a mindblowing experience. This is not something I'd try at
home. It would be very very hard to replicate the experience of peeling
back the layers of the onion. There are many other reasons why this is a
very bad idea (people don't like feelings/memories forced upon them, it's
pretty hard to tell the player to start unremembering the clues they've
collected, you don't even have a simple memory model for putting certain
facts in a structure/representation in the gamespace even if you wanted
to, etc).
Nonetheless, there appear to be some interesting questions to be asked
about many things in IF, based on watching this movie. [3] Just how deep
our representation of thoughts, memories, and "things we know for certain
to be true" should be programmed into an IF world come to mind. It's
losing battle, of course, and even some kind of fancy rule based AI wilts
at the prospect of being able to capture the details of what an NPC knows
to be true. Wood is hard? Ok, well that's an attribute we have to add.
Water is wet? Ok. I like ice cream? Hmm.. ok. I'm currently on a mission
to kill Joe? Uh, ok. There's just no way to deal with storing this or
making use of it in a game situation unless we know in advance exactly how
we need to use that information down the line.
That's not a new thought.
From watching this movie, what is new (to me -- but potentially painfully
obvious to others already?) is that the opposite is not true either. That
is to say, it doesn't seem sufficient to encode all of the responses to
logical stimuli in text responses in the game. To elaborate: a logical
stimulus is what I am going to call anything in the game which should
provoke an "A-Ha!" in the PC. That's not an "A-Ha!" in the player... but
in the PC. So touching the hot stove and getting burned didn't really make
the character learn anything new. No new logical input. Just a physical
one. Entering a new room and taking in the scenery is the same. Unless
maybe you just happen to witness the last part of a murder taking place,
or the room description happens to give away a clue that would make the
PC's logical wheels start turning vigorously.
Ah, you say- this isn't fair. Virtually everything could be considered at
least a little bit of a logical input, even the physical inputs will make
the PC think about SOMETHING, and therefore aren't divorced from the
logical stimulus. Sure. But probably 90% of those types of actions don't
have a lot of bearing on understanding the underlying plot of the story.
Which is where things get dicey for me.
Say you read a tidbit in a room description. Or you pick up a note and
read it and it tells you who the killer is. Now while it isn't true that
every logical input to the PC has to have a representation in gamespace,
it seems that some really should. Whether the PC "knows" about the dungeon
yet makes a difference in whether he gets the monologue from NPC1 first or
the monologue cum suicide sequence from NPC2. And now comes the part that
stinks. While not everything exists solely as cued responses and if-then
statements to see what bit of text is launched at the player based on the
state of knowledge the PC has amassed (according to the prevailing
representation comprmise being used), some manner of internal rule basis
may be needed for cases that are harder to resolve than an if-then
sequence keyed to ten different PC_knows_about_the_dragon tags can
provide.
Case in point, the situation such as in Memento [4].
In general, of course, it's wise of a designer to side step all of this.
Most of the time, information is either true or it is not. And most of the
time, there is not much point (as noted earlier) worrying THAT much about
the representation of what the PC knows versus what the player might know.
But in gameplays about memory itself, it's going to be tricky business.
Last chance... don't read the footnotes if you don't want to have things
really ruined for you movie experience in case you haven't seen it yet.
Ok, this REALLY IS your last chance.
[1] The principal trick is that each scene is told for about five minutes
of screen time (covering an unspecified and differing amount of movie
time), and then the next chunk of movie we get to see is the chunk that
comes before, in terms of the underlying linear plot. You can think of it
as being piecewise backwards.
[2] But post traumatic stress disorder is clearly not ruled out, and is
implicated as the real cause for his troubles at more than one juncture in
the film
[3] Such as (1) memory in IF, (2) NPCs that lie, and what to do about
them, (3) how much our trust/distrust of perception should influence game
responses.
[4] In which the PC himself is very very confused about the facts. Not the
player. The PC. The PC has been fed conflicting information from a variety
of sources. In a traditional representation, the guy knows things like
"John G killed my wife" and "Teddy is really a John G." and "Natalie is
using me". What does this sum up to? Who knows... the PC can't resolve it.
However, once you have more than two or three facts placed next to one
another, you are not going to be able to resolve any truth about what the
PC does and doesn't "know for sure" unless you can COUNT ON all of your
pieces of information as being rock solid. In this movie, you clearly
can't. What's a PC (or game designer) to do? (answer, in practice, most
likely try to put up with the combinatorics until they get very very
tired)
Dennis G. Jerz
> Last chance... don't read the footnotes if you don't want to have things
> really ruined for you movie experience in case you haven't seen it yet.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> Ok, this REALLY IS your last chance.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> [1] The principal trick is that each scene is told for about five minutes
> of screen time (covering an unspecified and differing amount of movie
> time), and then the next chunk of movie we get to see is the chunk that
> comes before, in terms of the underlying linear plot. You can think of it
> as being piecewise backwards.
The Harold Pinter play _Betrayal_ has a similar structre. Time moves forward
as usual during each of about nine or ten scenes; but after the lights go
out and come up again, we realize that each successive scene takes place
before the previous one. Not immediately before -- sometimes days or months
before. So the play opens with what looks to be like a somewhat awkward
meeting between casual friends, but as the play goes on, we explore the past
that these two people share.
And I can't for the life of me think of the name of the movie... a 1940s
British film that starts with a woman getting off a train, meeting a causal
accquaintance, and then having a few abrupt words with some man. Then we
see a flashback, which leads up to the moment the protagonist gets off the
train. This causal acquaintance turns out to have shown up just at the
climax -- her interruption, when seen for the first time, is comic; when
seen the second time, her cluelessness and lack of timing make us want to
slap her.
Jason Love
> At the heart of it is a character that has trouble making new memories.
> He's not amnesiac, nor an Alzheimer's patient. [2] A made-for-movie
> concoction of another sort.
>
> [2] But post traumatic stress disorder is clearly not ruled out, and is
> implicated as the real cause for his troubles at more than one juncture in
> the film
Actually, it's my understanding that he's a sufferer of anterograde amnesia,
which is a real condition. Anterograde amnesiacs have functioning
short-term memory and functioning long-term memory. However, nothing new
can be stored in long-term memory - it's effectively locked. Nothing new
can be committed to long-term memory. In other words, they can only recall
their lives before they entered the condition and the most recent of events.
I don't think there's a cure, yet, either.
Abundantly obsequious,
Jason Love
"..."
Gunther Schmidl
amnesia,
> which is a real condition.
I think that in this case it amounts to hippocampal damage, and there
definitely isn't a cure for that yet.
-- Gunther