autumn in Santa Monica
Friday, December 1 2000
I am sort of puzzled that no one took me to task yesterday when I equated information to entropy. I'm surprised a paunchy gentleman with a button-festooned headband didn't pipe up with an, "Excuse me but 'entropy' is disorder and chaos while 'information' is highly-organized and thus of a low entropic state. Is anyone here good at giving backrubs?" There may be some validity in this argument, but what I'm noticing is that an interesting thing happens to both information and mechanisms as they shrink away into microscopic patinas: they become increasingly difficult to apply to new uses. This means that they gradually lose status as interchangeable resources (commodities) and come to take on the characteristics of one-purpose applications, things that quickly become useless as the environment changes or as newer, better technology is introduced. As the interfaces between the components shrink with the components themselves, they become more and more proprietary until they only have one application: the connection of cache RAM to the microprocessor on one specific model and make of Socket-7 motherboard for example. Organic systems are exactly the same way. Think for a moment about the optic nerve of the vertebrate eye. It's a fairly high-bandwidth parallel interface between the eye and the brain. But it is so small and so intimately connected between a particular eye and a particular brain that it isn't really an interface at all, since it could never be unplugged and replaced with another. It is absolutely proprietary to the specific application: half the vision system for one particular vertebrate. This means that once its owner is done using it, a vertebrate vision system is good only for scrapping into its smallest organic components: amino acids, vitamins, nucleotides, etc. None of the complex structures capable of transmitting video to a processing system are of any use to a predator or scavenger eating the corpse of a dead vertebrate. Indeed, there is so little nutrient value in a vertebrate eye that it usually isn't even eaten except by insects and microscopic opportunists. This implies that in relative proportion to other things, a vertebrate eye is highly entropic.
The same is true of electronics. Back in the 1980s I could easily recognize and interface to the functional parts of a computer just by looking at the motherboard. There would be a 74LS245 bidirectional bus driver and it probably was sitting on the processor's data or address bus. Now if I could just decode the address bus and find an unused block of addresses in the address space, maybe I could interface some more memory or a little voice synthesizer chip to my VIC-20. The interfaces were macroscopic and could be manipulated with a soldering iron if you were careful. I could stack static memory chips one on top of the other and bend out the chip select pins and run those to a 74LS138 decoder chip with short pieces of wire. None of these things can be done with the motherboard of a modern computer. The soldered connections are nearly microscopic and large specialized chips contain most of the functionality. If something very small breaks, the whole thing is rendered useless. All that complexity turns immediately into useless entropy. And, as a source of raw materials, motherboards aren't especially valuable. They contain some gold and other precious metals mixed in with toxins like arsenic, but getting them out takes considerable skill. The "digestion" of a broken motherboard is about as complex as the digestion of a sirloin.
All this thinking about microscopic hyper-proprietary interfaces got me to thinking about the possibility of making a standard interface for things that never have had standard interfaces before. Imagine, if you will, being able to attach any sort of video equipment to your head and use it in place of your eyes? Impossible? Perhaps, but not entirely so, especially for someone who has not yet learned to use their eyes: a baby or a blind person. I've read somewhere that the human brain, when first encountering the barrage of data from the human eye upon birth, has no way of making sense of any of it. A period of adjustment time is necessary as the neural circuits necessary to interpret this data form and branch off to interface with other circuits. If a baby had no eyes at all, perhaps a neural interface could be wired in some way to a specially-coded version of a standard electronic video signal. On the brain side of the interface, this would require many electrodes acting in parallel, because unlike electronic equipment, no single point of contact in the human brain can make sense of the high-bandwidth data barrage of, say, a composite video signal. But I think that a brain of proper tenderness could adapt to a well-considered interface, one consisting of several hundred electrodes containing low-bandwidth fractions of a video signal. A brain could eventually wire itself in such a way that the person using this interface would actually be able to "see." Of course, I'm not the only one thinking this way.
It wasn't an especially happy day at work. It's pretty clear now that I need to find a new job the moment I get my yearly bonus in January, whatever that turns out to be. I feel marginalized and under-appreciated, but it's not nearly as humiliating as it was at CollegeClub.com. Indeed, the UK crowd seems to like me well enough, even if I only mustered the necessary motivation to work on their site in the past week or so. That site is actually looking pretty good now, but (in order to have something presentable for Monday), we've all agreed to come in and work out some small remaining issues on Sunday. Yes, Sunday. (I didn't think Brits could be motivated to do anything but participate in a football riot on a Sunday, but I guess I was wrong.)
The day was grey and bleak but I took pictures anyway just because my camera finally has working batteries again. I went out for lunch a little after four and bought my usual poverty fast food: two Carl's Jr. Famous Stars. I try not to watch as the Mexican ladies in the back make them. I also try not to look at my burger too carefully as I eat it. You never know what might be in there.
After work, the woman who is the CTO of the UK site, took the whole team to a place in Venice called "The Brigg." Having a pool table, a funky odor and a good anti-pop jukebox, it's the closest thing you can find to a dive bar on Abbott-Kinney. We drank beers and ate pizza from a pizza place up the street. I tried to pay for a round of beers but the CTO quickly corrected my generous impulses by suggesting that we let the company pay for it. As I chatted about subjects ranging from my frugal environmental lifestyle to the antics of Bathtubgirl, I found myself unconsciously pealing the label from my Sierra Nevada beer. Simon the graphics guy chuckled and said that this meant that I was sexually frustrated. Fair enough, I agreed, but then I asked "Isn't everyone sexually frustrated?" and searched from eye to eye and saw that everyone, even the CTO, was nodding their heads in agreement.
The Brigg is so close to Bathtubgirl Central that when our party finally split up I impulsively decided to just drop in.
Bathtubgirl seemed to be genuinely touched by my unannounced visit. But I wasn't especially entertaining, choosing to pass out in her bed for the first several hours I was there while she did her webcast.
The view from my workplace: the set of Buffy the Vampire Slayer. Click for a bigger picture.
The garish primary colors of Carl's Jr, where I bought my $2 lunch today at 4:20pm
Trees don't get colorful in the autumn of southern California.
Their leaves turn brown, shrivel,
and after much resistance, eventually fall.
These are some exotic gingkos, perhaps the most deciduous trees in the entire world.
The setting here is the company playfield of the MGM headquarters in Santa Monica
(where I ate my lunch of Carl's Jr.)
Palms and gingkos.
Dirty autumnal sand and the fallen leaves of gingkos.
For linking purposes this article's URL is:feedback
previous | next