Knowing and Doing: The Metaphor Metaphor
Knowing and Doing: A Blog Entry from Before I Had a Blog #2. Eugene Wallingford has just posted his notes from the Kent Beck's keynote at the 2002 OOPSLA Conference. I was at that presentation, and Eugene's notes provide some key, deep thinking about metaphors and then metaphor as a metaphor. Kent was also unabashed that choice of metaphor has non-trivial consequences (e.g., "War on Terror" and "War on Drugs"), including disfunctional results.
I don't remember Kent giving this illustration exactly (and I may have missed the slide while note-taking):
Metaphors run deep in computing. An example: "This is a memory cell containing a 1 or a 0." All four underlined phrases are metaphorical!
But it is brilliant. It is also an anchor for discussion of the manifestation of abstractions and how that happens with digital systems (itself a metaphor).
I have some additional notes (notebook #35 pp. 93-96) that I will add to the account. They need to be read in conjunction with Wallingford's account (which doesn't seem to allow comments and I gave up looking for his e-mail address, so I am putting them here):
- At some level, there has to be some metaphorical thinking going on. You have to name the objects something.
- Kent thinks that metaphor is more fundamental than he expected. We should leverage it by using it consciously. There is unconscious use of metaphor anyhow.
- Metaphor is a kind of trap? Oh, I misread myself. It is about trope.
- We spend a lot of time not saying what we really mean.
- On Lakoff and conceptual metaphor. All abstract thinking is metaphoric. (One reference is to Lakoff and Núñez, Where Mathematics Comes From: How the Embodied Mind Brings Mathematics into Being.)
- I see a note to point out Wheelright's Metaphor and Reality to Beck, but I never did it. (I did mention it to Lakoff once while aboard an airport shuttle from a cognitive science conference. He didn't seem that interested.)
- There are blends of several different metaphors. This is where the business of memory cells comes up and observation that "container" is a ground metaphor. There are layers of metaphor going all the way down.
- Grounded metaphors have a basis in physical experience. Motion along a path is a ground metaphor for arithmetic. One can physically act out ground metaphors. This was behind the Lakoff and Núñez work in how arithmetic seems innate in children (and for small numbers).
- Hypothesis: All metaphors are based on ground metaphors. [Me: Everything that we do has a physical basis.] An extrapolation that doesn't land for me is that computers have equivalents of physical awareness.
- Kent reflected on his work with Ward Cunningham on HotDraw (a real project as far as I can tell). He uses this to illustrate that metaphors can take a long time to gel, take experience, and there is no shortcut. Use productive waiting.
- Beck is excited about Eclipse as it was emerging in 2002. It excites him and frightens him. The plug-in mechanism used so heavily has to be well-refined and easy to use. He offered some considerations about plug-in ground metaphors, such as power strip. Today's users can reflect on whether it is still exciting or still frightening. (At this point, I had already given up on Eclipse as becoming way too heavyweight and complicated, based on the down-in-the-bowels discussions on the Eclipse lists. That doesn't mean I have a better solution. I simply chose to do without.)
- Beck charged us with the following:
- Become aware, play with your own metaphors
- Unpeel layers of metaphors in informatics
- If the grounding hypothesis is valid, more direct physical activity in metaphors will make us better programmers.
- At the end, Beck asked for a vote on whether he should give up this pursuit of metaphor. There were very few (but some) "give it up" votes.
posted by orcmid
at 6/04/2007 12:32:00 PM
