Information Technology Reference
In-Depth Information
1.
Writing model of representation. By “writing”, I mean that the representa-
tion is something apart from you. It is a resource in situated action (Such-
man 1987). You have to make sense of it in each next situation. There are no
complete, systematic, guaranteed rules for this making-sense. And when
you
do
manage to figure out what in some situation a representation is
talking about, there is no way to finish listing what about the situation
enabled you to do this.
2.
Dependency model of routine evolution. All forms of activity are snap-
shots in the evolution of routines. The routines themselves are intertwined
with the patterns of society and with the layout of particular places. The
model says: when you think a new thought in some situation, you connect
it back to its premises (Stallman & Sussman 1977). When you believe those
same premises again in some future situation, you automatically call up the
conclusion.
I'm trying to relate these two ideas. I think most concrete activity requires you
to interpret representations, that is, “making sense of a representation in each
next situation by figuring out what in the situation it is talking about”. So, for
example, “turn left” will indicate different actions in different situations.
This sounds like a lot of work. If it were really a fresh challenge to make
sense of “turn left” or “open the bottom drawer” on every moment, how would
we ever do anything? But I don't think it's that bad. What you have in practice
is a patchwork of routinized methods. In some past situation someone said
“take the next left”, and you took certain specific concrete actions: you looked
around, you searched for particular shapes and colors, and maybe you walked
around and looked some more. You had your own reasons for doing all these
things, and all of your actions and their reasons got connected, so now they're
ready and waiting to happen again in new situations. In subsequent situations
some of the reasons might not have applied, so instead you took other actions,
and these themselves led to new connections.
Perhaps after enough of this you develop a sufficient repertoire of routines
to apply “turn left” to almost all of the left-turn situations you encounter in
the average day. You've developed habits of interpretation. So whenever you
tell yourself - or someone else tells you - “turn left”, you'll be able to do it right
away, “automatically”, without any hesitation or difficult figuring-out.
These routines, like all routines, will evolve (Agre 1985a, b). Very often the
evolution of a routine involving a representation will permit you to undertake
the activity without the representation being present in physical form. For ex-
ample, using a recipe ten times might let you make the dish without the recipe.
