stimulus junkies ¶
Walter Kirn's Autumn of the Multitaskers was talked to death several places yesterday, so I'll only touch a part related to what I was thinking recently. From the second page:
Multitasking messes with the brain in several ways. At the most basic level, the mental balancing acts that it requires—the constant switching and pivoting—energize regions of the brain that specialize in visual processing and physical coordination and simultaneously appear to shortchange some of the higher areas related to memory and learning. We concentrate on the act of concentration at the expense of whatever it is that we're supposed to be concentrating on.
I've never seen evidence multi-tasking makes a person more productive beyond gaining extra energy from stimulus if more was needed to keep engaged. If you're already fully engaged, extra input's a distraction and almost certainly makes you do worse than you would on something that needs you to imagine as much as possible to see ahead in the dark.
Kirn's writing slightly overplays criticism of multi-tasking, but you need to take that with a grain of salt. Maybe whining geeks actually do see this, who position his piece as anti technology. Or perhaps they merely love a stimulus fix, and are happy to confuse higher consumption of input with useful production. (Or maybe they see their roles as professional consumers who needn't make anything.)
When I was a teenager I experimented with how many things I could do at the same time, and studied what it did to my memory and attention span, both of which were quite good and could stand a lot of abuse before I reached a limit. One of my experiments involved reciting Lewis Carrol's Jabberwocky (see also wikipedia) to myself — in my mind, not aloud — while doing other things to see what effect it had. I can still do this today pretty much any time I want, since it's hard to forget anything you've rehearsed thousands of times. (On long walks I watched how silent recital changed my thought process.)
I was able to read at the same time, but at rather reduced speed. (When I try it again today, I notice I can also "hear" some of the words I'm reading at the same time, overlaid like a radio and television playing at once.) I could also juggle at the same time, which isn't surprising, since the activity is very different. But I could also juggle and read at the same time, while reciting to myself — after mounting something to read on a wall. I had to watch juggling balls by peripheral vision.
Just reciting a poem subvocally, in an automatic loop, all by itself doesn't cause much impairment, and causes an interesting change in perspective at the same time. I found the change in perspective fascinating, and that's why, periodically, I kept trying it again. As you'd expect, silent recital mostly shuts off internal dialog in the mind, so the constant chatter of commentary you run in the background stops — a blessing if one day you need to stop talking to yourself about a maddening idea.
But an unexpected side effect was a change in my perception of time, and my ability to visualize things far away in time and space in a manner that would normally require a little dedicated concentration. It seemed my internal dialog distracted me, and I could see with my mind's eye much better with it turned off. The effect on my time sense was twofold: 1) my sense of the now was much greater and continuous, and 2) I could recall my past with great clarity because I could just see what I remembered again without editorial commentary. In particular, it was easier to review long term todo lists and priorities, in a manner somewhat like what I often do at the edge of sleep and waking thought. It was easy to ask "What am I missing?" and review issues with a minimum of internal chatter.
When I did this as a teenager, the effect was profound. But when I do it again today, the effect is only slight, for apparently two reasons. First, I'm better at multi-tasking and I don't succeed in shutting off all internal chatter (I can still do it). And second, I seem to be closer to the desired perspective all the time already: I'm almost always seeing what matters to me even if not right in front of me. I'm always addressing a list of problems, and I'm almost always designing, no matter what I'm doing. Which of course means I'm absent minded, explaining why I'm never bored anymore: my thought process is quite entertaining. (Boredom was quite painful to me as a youngster, but I seldom if ever feel it today.)
I read email periodically every couple hours, when I need a change of scenary for contrast, since your mind will silt up from repetitious activity, even if you enjoy it. But in general, I prefer single tasking as much as possible, because I need to devote my extra background capacity to solving problems — I always have a list of not-quite-finished items in queue. If I wallowed in multiple inputs at once, I wouldn't be getting work done. When I go to meetings at work, unless I'm speaking or working out answers to questions, what I'm doing is the same thing I'd be doing if looking at code on my monitor. I don't need to be at my monitor to make progress. I'm in the zone much of the time.
Therefore, when I see someone pay half attention to multiple things at once, I assume they have nothing better to do with their time. No one's at home and the lights are off upstairs. Zombies with tech gadgets wander our streets in search of more brains, always hungry for more.
code calendar ¶
I think I'm on track to finish enough hobby coding for my sons to do something resembling game creation next school year. Now that my day job no longer eats weekends, I'm back to coding for fun weekends. This and last weekend I've been designing a C++ api for æ, coming out a lot different from the earlier C api for several reasons:
- cleaner results on second try
- C is verbose as hell; C++ shows intent better
- slough off weaker notions others liked
- radical name abbreviations in þ style
But I start to think I've little prayer my code will read well to others when I open source it later, unless I can teach you to chunk ideas how I do. I enjoy how I'm coding, but it just doesn't resemble code I write in conventional day job style. (If I'm grumpy maybe I'll use a too damn bad license hinting your recourse on style issues.:-)
email gem ¶
I recently exchanged email with a guy I worked for many years ago. When he asked what I was up to, part of my reply included this ordinary remark:
I work for a networking company most folks haven't heard of, whose name I reflexively avoid telling. I try to keep a low profile. I have a website that doesn't use my name, where I blog a little and slowly document bits of something I'm trying to make.
I'd thought this very bland; but my correspondent had quite another view, making me take another look:
There is something poetic and mysterious and intriguing about the previous paragraph, as if it could be the start of a novel.
How did he think to apply fiction standards to email? (It's a creative view he brings to many things.) When I said I'd quote it, he didn't recall his idea. Maybe it struck me so strongly because I've been thinking about fiction, and it's been very hard to resist.
Last week a background thread in my mind kept fleshing out a funny piece on simulation and artificial intelligence, opening with a wolf named Sly who gets a telegram saying: pig found in briar patch, stop. (He's a retro wolf in a suit, kind of mafia with shiny shoes and a nine: Matrix meets 1940's animation.) I'm a hair from adding a fiction section since mingling it with my blog — the way I used to do — scans a little more creative than I care to look today.
As fun as fiction is to dream up and to write, it's not as valuable to me as serious efforts in code. Documenting ideas related to the code is more a chore than a joy, but gives me better long term return.
laundry ¶
I was going to hold this update another day to go with more engaging pieces about multi tasking and effects on your mind. But instead I'll just push this out now and add more tomorrow night. I just finished six weeks worth of laundry backlog, which ate a lot of time.
simulations and metaphors ¶
Okay, I'll pick up where I left off yesterday: heading back to Hofstadter and category theory. Let's prime the pump a bit with a question or two. What if your mind's a simulation engine? What if a role of cognition is modeling things (around you) to guess what happens next so you can plan ahead. Pattern recognition (brain features) would help match things you recall from life as well as things you simulated in dream or waking fancy. Imagination might be constructive anticipation of things you might see.
Of course, "what if" questions mean very little. I asked so you might form a template of an interesting idea. Then you might see parts of it in what comes next. There's nothing scientific about my conjecture at all, since testing it seems infeasible. But to come up with hypotheses to test, you might as well start guessing. This guess doesn't suck very much, and fits data I've seen, while motivating a mental theory with a darwinian function.
The trick is seeing things as other things: metaphors we impose on complex processes, so we can map things we already know on similar things we see now, or anticipate later. Maybe your mind does time-shifting, mapping, and creative modeling. That would cover a lot of ground with a common purpose obviously useful to survival: predicting and registering.
(Note I'm making this up as I go along, just to introduce quotes below. If it doesn't sound polished, it's because I just ad libbed it now on demand. That doesn't make it false — just rash. If it's drivel, at least it's fresh.)
category theory metaphors ¶
Check out the recent category theory thread over at Lambda the Ultimate. I'm particularly interested in the early lucid post by Tim May where he says:
Well, ironically, metaphors are themselves excellent places to start. Metaphors are mappings between words, concepts, stories. The metaphor of one's life as a journey down a road, a path. (Technically, one might look at one's life and its events and compare it to a journey and its events and see a "mapping between these mappings," a functor.)
If parts of this resemble my last section, then it's because I cheated — duh — by setting it up. However, the similarities are those related to category theory itself, and my material yesterday on spatial metaphors. I might be quoting a lot of Tim May here, so I'll get back to it:
For a really simple, real world example, consider a lawnmower. When out of gas/petrol, it needs to be refueled. The category of empty lawnmowers has an arrow/mapping/morphism to the category of full lawnmowers. Here the focus is more on the process of mapping one category into another, or mapping one collection of things into another collection.
Here I want you to think of the word simulation as meaning "process of mapping", then consider the role of arrows associating source and destination, and the spatial metaphor involved in comparing two things with a view to imagining one being changed into the other, with the arrow being the (um) sign representing the difference, with change modeled by motion.
(I happen to think this categorical way of thinking is immensely important for how we understand the world, how we might program artificial intelligences, and so on. Even more than "design patterns," I think we and other creatures chunk the world into pieces we can understand by finding the mappings and functors and so on which allow us to grok the world. We are, I think, category theory engines.)
Yes, I agreed with this the instant I saw it, and I translated "category theory engines" into similar sounding "simulation engines." Mappings and simulation are more native to my way of thinking than words. A great deal of my thought — especially when working — involves visualizing spatial transformations in computing systems, with heisenburgish fuzzy time effects to model race conditions under concurrency.
In an important sense, compilers for various languages map categories (source code, for example) into other categories (executable code). The way C++ does this mapping may be different from the way Haskell does it, so there are functorial relations between the mappings. (And so on, for many levels of computer programming. Category theory played a crucial role in the 60s and 70s in the work of Peter Landin, Dana Scott, and others.)
This part of May's remarks is related to the purpose of this web site, which is to work on programming language tools that focus on transforming code and data from one form to another, as happens to be useful at the moment, without any special preference for one language over another, beyond practical constraints of time and interest. Tools for self referential generation of code (and data) on demand would be useful, especially if the namespace modeling where code and data live was also managed by such mappings.
seeing is believing ¶
I'm too tired to do justice to this material now, but here is where I should write a lot about Douglas Hofstadter's On seeing A's and seeing As, about the nature of modeling machine intelligence. For example, early he mentions a drawback of logic based approaches, with which I agree:
What seems to be wrong with it? In a word, logic is brittle, in diametric opposition with the human mind, which is best described as "flexible" or "fluid" in its capabilities of dealing with completely new and unanticipated types of situations.
One of the reasons I always hated RDF as knowledge representation was it's use of graph-based first order predicate logic as the end-all be-all of world modeling, when basically it's crippled by static ontologies. They act to stop you from seeing things that are like each other, since that would be a blurring of categories based on partial similarity in disregard of approved hierarchy. Okay, now check out this next paragraph, which is chock full of notions matching my usual tropes here:
When presented this way, visual perception takes on a very different light. Its core seems to be analogy-making--that is, the activity of abstracting out important features of complex situations (thus filtering out what one takes to be superficial aspects) and finding resemblances and differences between situations at that high level of description. Thus the "annoying obstacle" that AI researchers often took perception to be becomes, in this light, a highly abstract act--one might even say a highly abstract art--in which intuitive guesswork and subtle judgments play the starring roles.
Also, compare that to Tim May's parenthetical comment last section programming artificial intelligences: that mappings are how we understand the world. Hofstadter's next paragraph summarizes the point very nicely:
It is clear that in the solution of Bongard problems, perception is pervaded by intelligence, and intelligence by perception; they intermingle in such a profound way that one could not hope to tease them apart. In fact, this phenomenon had already been recognized by some psychologists, and even celebrated in a rather catchy little slogan: "Cognition equals perception."
(Compare this to a quote from Charness I posted recently on chess-perception, which notes good play is more a matter of perception than deep thinking.)
Since it's now way past my bedtime, I'll jump to the end of Hofstadter's piece, to his quote of mathematician Stanislaw Ulam's comments on perception, intelligence, seeing as, and functional roles in contexts:
Convinced that perception is the key to intelligence, Ulam was trying to explain the subtlety of human perception by showing how subjective it is, how influenced by context. He said to Rota, "When you perceive intelligently, you always perceive a function, never an object in the physical sense. Cameras always register objects, but human perception is always the perception of functional roles. The two processes could not be more different.... Your friends in AI are now beginning to trumpet the role of contexts, but they are not practicing their lesson. They still want to build machines that see by imitating cameras, perhaps with some feedback thrown in. Such an approach is bound to fail..."
Then Ulam claimed the abilty to see as (cf earlier metaphors and category theory) was key to understanding intelligence. Such was Hofstadter's respect for Ulam he quelled objections.
context frames value ¶
For contrast with the comments about "context" in the last section, you ought to read my context-frames-value post (from LtU last August), in which I said:
In the 90's for years I ended online messages with this signature: "All values have meaning only against the context of set of relationships." (That's what I'd finally concluded after more than a decade of reasoning about knowledge representation, leading down so many avenues of gnarly logical regress I hate to talk about it.)
This was the message I'd taken from my research (described yesterday) in primitive language spatial metaphors: that meaning was more inherent in context of relationships than in actual objective values. It's only in how a value differs from others — in a way that matters in a given context — that establishes what a value means, there.
evolving ideas ¶
Your theory about how my mind works is likely wrong. I'm used to being attacked as a person I'm not, but folks think I am. (Even my ex wasn't good at knowing my mind any better than necessary to manipulate me, which she was quite good at.) I no longer try to make myself understood, which doesn't improve things, so I compensate by being unobtrusive: I avoid attention.
Anyway, tonight I'm only going to discuss topics related to category theory, and to the writings of Douglas Hofstadter, since I ran into both online this last week in ways bearing upon recent topics here. First I'll mention a bit about my background, to explain why I'd spend so much time in the library as a young man. (I didn't learn as much as the fictional Will Hunting, but I'm more an autodidact than not.)
late bloomer ¶
I had my first migraine with blindspot aura at age 9 in the fourth grade, around the first time I got high school level scores on standard tests (in science). I gather teachers thought I was slightly backward at first, because I wasn't good socially, then changed their minds and gave me puzzles to solve to see if I'd do something interesting. I was tested for eidetic memory several times because I seemed to have photographic recall. But I didn't remember details in complex photographs once taken from my sight, unless I'd consciously registered them. So not eidetic — I couldn't count the Cheshire Cat's tail stripes from memory.
I was verbally backward because my parents hadn't spoken to me much as a child. I was second of five born in a seven year period, so by the time I was four I had three younger siblings, and my older brother got a lion's share of adult attention. I think I heard more spoken words from television than from human beings. My mother raised us all alone because my father was usually out of town on business, playing traveling salesman. Several aspects of my acoustic hardware may have suffered as a result. I have little short term memory for sound, and I have trouble parsing words in noisy places. (This upsets people with accents, when we talk in a restaurant.)
math vs golf ¶
My father was (and is) a golf nut, and his preferred way to interact with me involved carrying his golf clubs for 50 cents per nine holes. One day he played golf with one of my math teachers, then told me about it afterward. It seems my teacher had accosted him on some moral principle, but my dad mainly wanted to get my reaction to one thing: my teacher said he'd be proud to have me as his own son (implying my dad was a clod).
Was I busting his chops too? That's all he really wanted to know; it was a hoot the math dweeb liked me.
library ¶
My home town in Iowa had a public library donated by Andrew Carnegie, and I spent a lot of time there. I don't know any other kids who went into the stacks and systematically browsed entire sections of the Dewey decimal system. During one phase I went through virtually all the books in sections about philosophy, weird science, and miscellaneous exciting drivel on topics like you might find in the National Enquirer. Like Fox Mulder I wanted to believe, but it sounded suspiciously like total nonsense. So I honed my skills to see evidence about things that are known and how people go about knowing things. I was a budding little epistemology wonk. I wanted to know how we knew anything.
I liked math a lot, so my mother bought me Martin Gardner books on recreational math. My math teachers treated me like a prodigy, since I always had their material down cold before they were even done — I anticipated what came next. At 13 I was about as bright as an average adult. By 16 I was often correcting my teachers when they went astray, and made them fix wrongly graded tests when it happened. For some reason my physics teacher had no intuition at all when it came to systems.
In college I wondered if there existed topics I'd somehow missed. How would you go about testing that idea? I tried the Encyclopedia Brittanica — it took me several days to flip through every page of each volume, looking for topics I wanted to know more about, taking notes on things I meant to pursue later. For comparison, I also sampled an edition dating from the 1890's — it had very interestingly different emphases on what was considered important. For example, most industry and technology related to mining and production made a big splash.
hofstadter ¶
One day a new book appeared on the new book shelf in the main library: Douglas Hofstadter's 1979 Godel, Escher, Bach (aka GEB). You could not check out books from this shelf until two weeks had gone by, so I came in every day and read it standing there for an hour at a time.
It's hard to describe what I found so interesting about it. It was fun! And all the writing involved a form of thought resembling what I wanted to pursue: how to reason about what you know. It was one of the few discussions about thought and reasoning I'd seen that was low on pure tripe. I think I'd already changed my major from EE to computer science by then, because I'd found programming was fun while circuits were boring.
My advisor was the computer science department head, and he liked to talk to me about ideas I was having. He didn't advise any other students, and his secretary kept giving me grief until he made it clear I was always welcome to barge into his office. I didn't know why he made an exception for me.
A professor in the architecture department told me she believed it was because I reminded him of himself as a young man. I think she enjoyed the startled look on my face. (This professor was the mother of a girl I'd started hanging out with — a juggler I'd met the year before who was a member of the unicycling club. The daughter was one of those odd young dyslexic women good at spatial things like juggling, and less good at the usual things like faces. She recognized me at a party because of the way I did a particular juggling trick.)
category theory ¶
A couple years after that I was trying to find patterns in lots of different systems, and the way category theory was described caught my attention. I was looking for a common underlying spatial metaphor. I was seeking a new set of primitives for encoding ideas, preferrably using another mostly fallow mental faculty in normal communication systems. I wondered whether getting some of our logical thought process into a spatial medium might use more of our native bandwidth. (I selfishly wanted to teach other folks to stop being quite so dumb, and it seemed more a problem in software than hardware.)
However, when I read books on category theory, I was stumped by the abstract algebra — it didn't connect with my spatial intuition at all. I'm spatially oriented like a geometer and not word/logic oriented like an algebraist. (Presumably this has something to do with the way I missed verbal experience as a little brat. Maybe chatty mothers rear algebraists.) I wish I'd given it more time than I did, because years later I ran across a pattern related to category theory. I found the presence of arrows as a primitive metaphor in Latin word roots.
arrows ¶
I had an idea to find a bridge between verbal and spatial reasoning by looking for a spatial metaphor in words. For no very good reason, I suspected it would be interesting to look at word etymologies. I wondered if a kind of intellectual darwinian evolution of ideas might have selected for effective and orthogonal conceptual primitives in common word roots. I was vaguely aware Latin word roots in particular seemed somewhat analytical.
Anyway, to make a long story short, around 1991 I took a two volume etymology of English words and I went through all of them with a highlighter, picking out etymologies that were suggestive, when I could see a spatial metaphor involved through some stretch of the imagination. My goal was to see if I could pick a set of useful word roots giving me a way to name spatial primitives with terms not totally alien to English speakers.
Then I gathered all the data I selected and massaged it, trying to fit patterns and reduce the parts that had inspired me to some rational system. I ended up with a nice pile of juicy word roots I liked, then I started comparing and constrasting them to find divisions and categories.
But then a disturbing pattern emerged: all of the primitives were part of just one single central idea. Ideas that had seemed quite different to me could be seen as slightly different shades of perspective and context in a common metaphor. The variety of ideas appeared to result from merely combinatorial explosion of options in where you decided to situate your perspective.
The Eskimos proverbially have many words for snow (yes, I know this is more folklore than not). I was seeing that we effectively have dozens (perhaps hundreds) of words for arrow, or parts of arrow, or nuances about the possibility of an aspect of arrow applying to a context. All of the metaphors were grounded in the idea of movement from a source to destination.
Or not moving because a wall or impediment blocks the path. Or only moving according to some frequency, depending on the nature of openings in time or space. You can focus on being at the destination that receives a thing that moves, or you can focus on being at the source that sends. Or you can focus on being the thing that moves, seeing the source and destination from that perspective. You can have shades of nuance in adjective, verb, and nouns that focus on whether you are before, during, or after the movement; are interested in the possibility, impossibility, or accomplished fact of the movement; are concerned about your active or passive agency in any of the positions; are scheduling the combinations of all these factors in the future, or seeing them in the present or past.
By having lots of words that almost mean the same thing, but have subtly different nuances, our vocabulary makes it easier to see nearly identical ideas as semantically different enough to reason about easily without any special effort. Saying every spatial metaphor can be boiled down to some nuance in parts of an arrow prototype is about as interesting as saying you can encode all data in ones and zeroes. While that's true, it doesn't help human beings at all to see data encoded in ones and zeroes.
I'd been hoping for an idea that would improve cognition when you applied it analytically. But I was afraid I'd discovered the opposite: one unifying idea collapses things so you have more trouble distinguishing detail. If you repeat one word over and over to yourself for a while, the word becomes nonsense as your feel for it dissolves into noise. It's the opposition of ideas to others that are different which allows us to see things in more categories. A unifying idea has an effect of dimming perception.
It was depressing for a while, but I found knowing this greatly improved my writing later, so within two or three years I was able to write with great clarity and speed. I can write almost as fast as I can type, and I don't edit what I write much, except to go back and remove extra words. (Only words that change the meaning need to be there, so I can drop connectives, articles, and whatnot when present by implication.) I use words of one syllable when possible when a longer word adds no new meaning.
Hmm, I didn't actually manage to quote Hofstadter tonight. Nor did I get to the category theory material either. However, you'll find they both pertain to what I wrote about above, when I finally get to it soon this week.
priorities ¶
This site's low priority for me lately. I'm only updating tonight after 1am to satisfy a minimal level of posting consistency. It's normal for me to write here only after midnight, when my mind isn't as creative, because I have better things to do with my time. When I'm sharp I write code. Mostly this is because I find what I'm doing interesting, and I want to see it get done.
I've been working freakishly long hours, getting into territory I never believed when folks claimed they did such hours in the past. It's only possible to work 70 hours when you do little else. Today I'll tell you a bit about my week.
seattle trip ¶
Yesterday I flew up to Seattle for a meeting so I could talk with folks about de-dup. Can't tell you much about that, though. It's kind of weird I ended up the de-dup guy, answering questions about what I can get the code to do: how far code can be changed this way or that way to fit one need or another. I often feel more like a Shaftoe in Stephenson novel than a Waterhouse.
I don't fly much, and it's pretty weird I needed to go. I gather management could no longer play middle man well enough, so I had to be on hand, so I could be asked directly without a filter. Or something. I dunno, but a lot of it had a feel of breaking ranks. Folks back in San Jose today found my description pretty interesting. This might have something to do with it: folks believe it when I say it, because I don't spin things.
(I'm really not sure what my role is. I'm just the guy who does things, and I see the whole picture involved.)
I got up at 4:30am for a 5am airport arrival for a 6:30 departure. Then I came back in the evening at 11pm. The day before I wrote code from 6am in the morning until 7pm at night, minus breaks, writing stuff we discussed changing once I got to Seattle. I think I'm taking this weekend off. It'll be the first time I didn't work every day since Last November.
refracted light ¶
I saw an interesting bulls-eye rainbow heading down toward Seattle. This is not a metaphor. I'm just telling you what I saw. I haven't heard this described before, so I thought I'd tell you about it. I think it must be hard to get just the right circumstances to see the visual effect.
I'd been studying the patterns of melted snow south of Seattle, and I saw a hard line of cloud cover approaching, which looked fascinating because it stopped with a knife edge between 100% clear sky and 100% total cotton fabric clouds. We we still very high up, so the clouds looked a thin layer of cotton fabric over the land far below, with cool wrinkle patterns going for miles.
I was staring down at the clouds, which were all white, and saw snow covered mountains sticking up above the clouds. The sun was exactly behind me, as I was looking westward while flying north. The sun was just high enough at 8:15am, so the point directly opposite in the sky was several degrees lower than the horizon — I was looking directly into this point.
Against this bright, all white background I saw a pale bulls-eye shaped rainbow. Not a circle — a bulls-eye around the point opposite the sun, but low intensity. I wondered if my window had a discoloration, but it didn't move as I changed my vantage point, and I saw it stayed directly opposite the sun. Then as we dropped lower on the approach to Seattle, it got more interesting.
I wondered: where's the moisture in the air to refract this light? As the plane dropped, the answer became clear. As the white cloud cover got closer, it assumed more texture and three dimensional relief — causing the bulls-eye to become distributed over the 3D texture of the clouds, so that's where the refraction was occurring: in the clouds.
As the plane dropped further, I was trying to decide if the plane itself was somehow participating in the effect, or whether it was just the clouds refracting light. So I was staring in the middle of the bulls-eye, looking for any effect from the plane, and I saw the plane's shadow in the bulls-eye center on the cloud tops. Which looked really cool, but only lasted maybe twenty or thirty seconds before we got so close to the clouds the rainbow faded.
I wanted a motion camera of course, but didn't have one.
caveat ¶
Please don't get an idea I plan special innovation here in months or years to come. I only aim to use known good technique in a very clear and direct way — just more thoroughly than usual. None of it should seem revolutionary. Reporting radical college musings from 25 plus years ago might accidentally imply I hope old ideas will come to fruition. Well, no, I don't.
I plan no syntax innovation in programming languages. Syntax won't look new, beyond gratuitous renaming. What code means might be interesting; how it looks will be ordinary. Odd bits might occur in async concurrency, and support for multiple syntaxes. I'll start with old and simple syntaxes first.
You should treat all my quotes from old notebooks as game play without any specific goal. This is in keeping with the content involved: playing with ideas is training to recognize when to apply them should the need arise. Compare the last sentence with the story about rat dreams as threat simulation.
advantageous ¶
Let's complete a thought from last weekend's piece on Alison Gupnik's Imagination is Real. Some of you might have read Robert Pirsig's Zen and the Art of Motorcycle Maintenance. If you haven't read it yet, you should. (A full copy is online at design.caltech.edu, but it's white text on black background, sigh). Because it's popular, you might ignore freely offered good bits inside — you do look gift horses in the mouth.
I only quote the part I wrote in a notebook on 21apr1982; I'm trying to show you how busily I looked to confirm an idea we conceive things in ways we find convenient and advantageous in preference to true, then we ignore our complicity.
Poincaré concluded that the axioms of geometry are conventions, our choice among all possible conventions is guided by experimental facts, but it remains free and is limited only by the necessity of avoiding all contradiction.
I cut the part where Pirsig cites Poincaré's idea the axioms of geometry are disguised definitions. (You can find this on most web pages about Poincaré.) The first part above is just context for the part I wanted:
As well ask whether the metric system is true and the avoirdupois system is false; whether Cartesian coordinates are true and polar coordinates are false. One geometry can not be more true than another; it can only be more convenient. Geometry is not true, it is advantageous.
A corollary of interest to me was this: one true way to describe things does not prevent another true way from existing — perhaps a more useful one — that we might not see until we try to imagine it first. It's a sin of omission that worries me.
parts vs concepts ¶
Two days before that, on 19apr1982, I put the fragment below in my notebook. (The way I avoid having too much to say is by only reporting what I put in my notebook in early 1982. I can't undertake to summarize related ideas.) I'll reproduce my original break in the middle, so no single block is too large.
Note all italic emphasis here was added by me in reproduction of the way I copied Pirsig's Zen passage in 1982.
Precision instruments are designed to achieve an idea, dimensional precision, whose perfection is impossible.
The next paragraph's a succinct version of Gupnik's idea:
It's the understanding of this rational intellectual idea that's fundamental. John looks at the motorcycle and he sees steel in various shapes and has negative feelings about these steel shapes and turns off the whole thing. I look at the shapes of the steel now and I see ideas. He thinks I'm working on parts. I'm working on concepts.
chess and architecture ¶
In early 1982 I read a lot of books on chess and architecture because both had plenty of literature on the use of imaginative insight and planning in advance of executing high performance skills. There was quite a bit of research on what expert chess players were doing in their minds, since it seemed a tidy case study in how the human mind works. And plenty of literature in architecture concerned problems in getting visions executed, and having good visions of design in the first place.
This is your warning more than one quote will appear from chess books and architecture books. I had a good reason to make repeat trips to the same well. And you already know what I was looking for — so I was likely to find it. I was trying to find references to ideas saying that imagination of a thing before it appears will influence whether you see it. Trying to expect the novel in advance, and being open to it, primes innovation.
(It's obviously an evolutionary advantage to imagine a tiger eating you before it does. And perhaps it's good to imagine others eating your lunch before that happens, too.)
On 22feb1982 I copied the following from Neil Charness's Human Chess Skill, found on page 39 of Chess Skill in Man and Machine.
'He saw everything!' is invariably the complaint of the chess player who loses a game. Other variants of the lament are: 'I completely missed (seeing) his move' or 'How could I overlook that move?' It is no accident that the operation 'seeing' is an element in all those statements. In the final analysis, perception seems to be the key to skill in chess.
There's quite a bit of Charness in that old notebook. But I'll constrain myself to another quote from page 43 of the same work. The final italics emphasis is mine:
Parenthetically, it also becomes reasonable to speculate about questions like: why does a 'highly intelligent' individual when playing chess, miss obvious moves? Moves are only 'obvious' when the patterns they spring from are recognized.
Since Gupnik mentioned the similarity in style of thinking when scientists imagine and children play, the following fragment from Norbert Wiener's Human Use might fit here. I copied this into my notebook later the same day on 22feb1982:
The distinction between passive resistance of nature and the active resistance of an opponent suggests a distinction between the research scientist and the warrior or game player. The research scientist has all the time in the world to carry out his experiments, and he need not fear that nature will in time discover his tricks and method and change her policy. Therefore, his work is governed by his best moments, whereas a chess player cannot make one mistake without finding an alert adversary ready to take advantage of it and to defeat him.
building models ¶
I was obsessed with the idea of models for a long time, since they appealed to a notion of anticipating future possibility in a consistent system, where all options cleave to a model.
Quotes in this section are from J. J. Coulton's Ancient Greek Architects at Work; problems of structure and design (1977 NA270.C65), which I copied the next day, on 23feb1982.
In discussing the formal and technical difficulties faced by the early Greek architects, one problem was omitted whiich falls rather between those categories — the problem of design.
Which of course resembles problems in software design.
A man modelling a clay figure can start from any point he chooses, can to a certain extent modify the past formed first in the light of what he does next, and can even reject the whole form and start again without much loss; a painter is in much the same position. The sculptor working in marble is rather more constrained, for a serious mistake will be irremediable and costly...
Does this sound like constraints in software, or what?
... but he can work gradually into the stone over the whole of his figure, so that the relation of the parts to each other can be clearly visualized at a stage when minor changes to any of them are still possible.
The next part sounds a bit like foundations of software platforms:
The architect on the other hand must always start his buildings at the bottom, and can not modify at all what he has built first in the light of what follows. Mistakes made at the start can therefore not be corrected, and they will be ruinously expensive, for a monumental building will occupy many men for many years.
And this last part was copied in my notebook in red ink, so I suppose I thought it most important. It should be all italics here, but you get the idea:
For these reasons an architect more than any other artist needs a technique of design, a technique which will allow him to visualize the finished building beforehand with sufficient accuracy to ensure that the lower parts of the building will suit the parts which are to be put upon them, and that the whole building is satisfactory in form, function, and structure.
Since I mentioned software several times, you might keep applying that perspective more than really necessary here. My main idea was importance of visualization ahead of time to efficiently reach goals. Even in software, where there's theoretically time and opportunity to change code, we often can't find time in practice under pressure of competition. To fail in software, you need only go too slowly to stay relevant, so a fixable error in theory might not be fixable in practice.
Software is less like research science, and more like a game with an adversary. (And they wear sheep skins.)
Are we having fun yet? What are chances I've more from the next day in 1982? Yes: back to Charness on chess.
chess perception ¶
The following is from Charness, Human Chess Skill again, page 51 (copied on 24feb1982):
The process of choosing a move seems to involve perception as a primary component, and in particular, the recognition of thousands of stored patterns. It may seem surprising, but chess is not so much a game of deep thinkers as one of skilled perceivers.
recognizing solutions ¶
Because it relates to yesterday's piece on Alison Gupnik's Imagination is Real, I trawled through my 1982 notebook, looking for Robert Pirsig's quote of Poincaré on the idea that geometry was advantageous as opposed to true. And I ran across a few things I'd written myself that looked worth citing here. Each is very short, but still related to the same ideas I think about still.
I wrote this one on 11mar1982 while reading Gombrich's Sense of Order (1979 NK1520.G65). It related to my idea of situational calculus.
Problem solvers find solutions by recognizing them, during a conscious or subconcious search procedure, whereby plausible answers are generated by mental assemblages and juxtapositions of relevant items that are manipulated actively, until an adequate solution is generated and recognized as such.
Note how much this resembles things I wrote here in recent weeks, despite not seeing this notebook in over ten years. The next day (12mar1982) I wrote the fragment below. Notice how it relates to modern ideas of attention economy though this idea was far in the future back in 1982. I was interested in economy of my attention.
Cognition is a screening of real and imaginary events through our perceptual/conceptual complex. I'm currently picking apart cognition and the per/concept complex. The emphasis on analysis of cognition lays on the double role the word plays for both discrimination and thinking. In both cases the critical necessity is adequate amounts of attention. Making discriminations in the environment is a matter of controlled direction of attention.
third world ¶
I wrote the fragment below on 01apr1982 while reading Karl Popper's Objective Knowledge. I was thinking about Popper's notion of third world knowledge: the manifestation of parts of our minds and information in our environment, after we make changes there.
Just as human ability has limitations in a physical sense, so it also has limitations in a mental sense. And just as devices and physical situations may be contrived to enhance scope of human ability to perform actions, so can devices (and third world configurations) be created to enhance scope of human ability to perform cognitive operations.
(There's much more of course, though it's abysmal writing. It was just a note to myself. It's not at all a coincidence I ended up working on internet software years later. I got here on purpose.)
Is that strange thinking for a 23 year old? I was trying to motivate design of tools to improve thought. I was out of school and out of work, and one month shy of moving to Berkeley, where I hoped to study architecture.
real vs virtual ¶
After I comment on recent work status, I'll start the first of probably several days of blogging devoted to topics touching on role of thought as virtual reality theater, whether in dreams or another kind of imagining. Somehow I queued up a long series of related items.
caught up ¶
This weekend I caught up with my day job's schedule, so I can resume a more normal home life soon. I've been making up techniques I've never seen before to get effects I need. While I don't hate shared memory yet, the shine is off and the honeymoon is over. Ultimately, it's hard to write a lot of new software with many degrees of difficulty.
threat simulation ¶
Last week Psychology Today had a story about research that robbed rats of dreams and then studied rat behavior. Ideas were debated on Slashdot under headline Dreams Actually Virtual Reality Threat Simulation? (because the original article focused on poor rat response to actual threat conditions). Jay Dixit's main paragraph says:
Finnish psychologist Antti Revonsuo believes the marooned rats lost their ability to defend themselves not because they were exhausted but because they were robbed of their dreams. Dreams, he contends, are a training ground in which animals and people alike go over the behaviors that are key to their survival. Prevented from dreaming, the rats were unable to rehearse their survival behaviors. In other words, they were defenseless because they were out of practice.
The basic idea is that mental rehearsal improves performance, even in dreams. I cite it here because it relates to other pieces coming soon about the reality of mental models, and how thoughts affect the real world since what we think guides our behavior. And our creative ideas are the source of things we manifest in the world. It's nicely described by Alison Gopnik below as: ideas are real.
changing minds ¶
A wonderful year end series of stories and opinions by many illuminaries appeared last week on The Edge in response to a question: What have you changed your mind about? There's so many good entries I want to quote from later. But today I'll only quote Alson Gopnik on Imagination is Real, which I enjoyed so much I want to quote its entirety here (local copy: gopnik-edge).
real imagination ¶
Alison Gopnik is a psychologist at UC Berkeley who says she changed her mind about epistemology:
Recently, I've had to change my mind about the very nature of knowledge because of an obvious, but extremely weird fact about children - they pretend all the time.
Gopnik wonders how living in an imaginary world is advantageous from an evolutionary point of view. Why do we all love fiction? How can fiction be as good as truth in a scientifically accurate worldview? Can lies be beneficial? How is pretend play good for children?
I still think that we're designed to find out about the world, but that's not our most important gift. For human beings the really important evolutionary advantage is our ability to create new worlds.
Gopnik notes if you look around any room — and at everything in our environment — all the objects around us are constructed, having once only been the imaginings of someone. First we think of things, then we make them. So large parts of our reality start in the mind.
But that's just what our human minds do best - take the imaginary and make it real. I think now that cognition is also a way we impose our minds on the world.
This entire thesis matches the world view I've had for many years now. It's simply how I see things: much of our environment is just ideas made real, when we customize the world to suit ourselves, adapting the world to us instead ourselves to the world.
(When I was 17, one day I noticed everything I learned in school was a construction of human minds. Academic material we learn is digested theory and practice created by someone to explain things. Even history is a story. I started wondering if we went about it as well as we can.)
In fact, I think now that the two abilities - finding the truth about the world and creating new worlds-are two sides of the same coins. Theories, in science or childhood, don't just tell us what's true - they tell us what's possible, and they tell us how to get to those possibilities from where we are now.
A great way to put it: imagination conceives possibilities before we explore usefulness and applicability. Since college I wanted to improve our ability to see possibility, because it seemed a constraint in thinking creatively and perceiving evidence. (I imagined language tools.)
notebook ¶
By chance I ran across one of the notebooks I kept in 1981 and 1982, in which I gathered material about the use of language and notation systems generally to influence how we think, in order to improve our grasp of (eg) situational logic. This notebook is from 1982, and has a lot more entertaining material in it than I remembered. It's too late to cite something now. Well, how about a random taste?
Page 205 from 07-april-1982 quotes several parts of some book — so far I only see this attribution: Forms of Intellectual and Ethical Development in the College Years by William Perry Jr. This is from page 92 of that work:
The assumption is made that, for example, what a student thinks knowledge to be like strongly influences, limits or even dictates his perception of what his instructor is up to.
A quick search on Google turns up the book. Here's a blurb from Jossey-Bass on this ISBN: 978-0-7879-4118-5 title. You can easily find others.
Throughout this journey of cognitive development, Perry reveals that the most significant changes occur in forms in which people perceive their world rather than in the particulars of their attitudes and concerns. He shows ultimately that the nature of intellectual development is such that we should pay as much attention to the processes we use as to the content.
I chose this page to quote instead of others because of the processes versus content remark in the last sentence. I'm not pushing the book; it was just interesting.
