Teaching a Circuit to Talk
The most significant and potentially “posthuman”
issue confronted in the designing and teaching of a Writing for Interactive
Media class is determining what actually falls under the definition of
“writing.” Nothing represented the inadequacies of traditional genres
and rhetorical theory more than the encyclopedic, database-oriented software
-- it is a genre that makes up a considerable share of the interactive
software market, and yet the content of these products often look more
like something created by a systems analyst than a writers (as we understand
the term).
For that reason, the decision to include a database assignment was uncertain,
to say the least. Although it doesn’t fit any definition of writing as we
understand it, the final project definitely creates a system of communicating
information, which may very well become a definition of writing (or at least
linguistic engineering) in the future.
For this assignment, students work in teams collecting data, classifying it, and
creating an interface that will allow users to access that information in a
meaningful form.
Most of the work is done with phones, feet, logs, note cards, grids, and
seemingly endless debates about how to classify and tag information so that a
computer can identify each item in the database by a variety of features or
characteristics. The final project itself asks the writing team to construct no more than six to eight
questions to be posed to a users, and these are the only complete sentences the students write. When
programmed as a “live” database, the machine essentially holds a very
primitive conversation.
The computer uses the lexicon and cognitive paths my students have provided and
assembles the data according to the rules of a grammar from a programmer, and
(if all goes as planned) this enables the computer to answer simple requests. For example, a conference
go-er tells the computer that
she wants to know what stores in Crown Center carry Kansas City souvenirs; the
computer complies with a list of stores and refers the user to a map for
directions on how to get there from the exhibit area of the conference.
The exchange can happen in less than a minute, but the groundwork that
enables it is tremendous.
Last semester, my students decided that they
would gather materials and create a database of museums in the Kansas City area
(no small feat). Before beginning the research, they brainstormed on what
information people would most likely want (hours, admission charges, type of
displays, cultures/historical periods included, and amenities like cafeterias,
tours, special access). After collecting pages of material on everything from
the Nelson-Atkins to the Boilermaker’s Archives, the student met again to plan
their interface.
The students were given rudimentary instructions on searching with Boolean
terms, incorporating various web form tools into an interface, and asking a
clear question, then they began work on developing a classification system to be used
with each of the many variable categories. For example, the students elected to
provide the user with a drop-down
menu to select the kinds of cultural or historical exhibits they
wanted to see featured in a museum. To produce this option, they
needed to draft a list of categories (Civil War, Jazz Age, Native American,
European) into which they could sort all of the displays and then assign the
appropriate "tag" or "tags" to each entity covered by their
database. And, of course, they needed to do this type of classification for
every variable.
For
the finished project, they turned in one hypertext screen with their interface
form and pages and pages of data grids -- so, for example, if I found the column
"food services," I would learn which museums had vending machines,
snack bars, cafes, or full-service restaurants (or any combination of that list).
Obviously, this project taught primary research and analytical skills. Clearly,
it focused on structuring information in an organization that can be accessed
directly by a computer program as well as indirectly by a consumer, which sounds
like writing. But when there are no
theses, no sentences, no punctuation, it is difficult to really feel comfortable
calling it writing, yet – as computers become more human – it is not
difficult to imagine that this is what a large part of providing computers with
language capabilities will become.
In the end, what finally tipped the scales in favor of including database
construction in the class was this: if they didn’t learn it in the Writing for
Interactive Media class, we didn’t know where they would learn it. But the
question still remains, could this be a preview of one of the first truly
posthuman genres? Is it writing at all?