@stephtwang@ctbeiser@nwilliams030 my interpretation is something like: shenzhen i/o is fun/challenging/funny because you're solving trivial tasks using inadequate representations (eg asm). For work that "actually matters", better strategy may be to seek better representations rather than cope with puzzle-solving
@stephtwang@ctbeiser@nwilliams030 at least, hutchins is claiming that the cognitive evolution of culture proceeds in this way (by re-encoding tasks in better representations rather than continuing to puzzle through them). Evolution is a slow gradual process, of course.
@wcrichton@stevekrouse@rsnous (btw, thank you for provoking this -- it's helpful as I'm preparing to write up Realtalk and need to understand the misconceptions that need to be addressed)
At my first job, I remember protesting that a certain fashionable industrial design decision
would look ridiculous in ten years. My boss responded, "We're not selling this product in
ten years. We're selling it now." I learned something about myself that day.
@tophtucker By the way, have you read much along this thread (that I just made up)? I think you might be into it. I'm thinking of Ong in particular (or at least as a starting point).
(FWIW, after Rich concedes that most data requires "sufficient formatting and metadata" and "some human involvement for interpretation", Alan could have added, "For a scalable (intergalactic) system, that "metadata" must be a *process*, because human involvement doesn't scale.")