Saturday, June 25, 2005

Information as Probability

Information has been the inverse of probability since the 1940s: that's how the telephone people saw it. That's how they needed to see it. Specifically, Shannon's definition is well known:
  • H = -∑pi logepi
  • Thus there's more information in the series "1, 15, 138 ..." than in the series "2, 4, 6, 8 ..." If you go to the movie and everything happens exactly as you would expect, if the blond looks like you would expect, dresses as you would expect, if the fiancé says exactly what you would expect, right down to the expression on his face, then the movie contains no information, it isn't very good, even fans of dullness won't like it.
    Information is the spice: the only spice ever needed.
    But how about considering information as related to probability in the following sense: cultures hear and understand only what they expect to hear and understand.

    I've seen people listen to sermons; and hear only what they wanted to hear: read books; and understand only what they expected ...

    I'm still fussing with the Mi.org Entrance, still don't have it quite right, and that must take priority; but when I can I must develop this wrinkle. Meantime, for example: if you take a people whose idea of god is a patsy for favors, you can then steep them in a literature of martyrdoms and non-communications, non-cooperations, you'll still have a people who believe in god as a lucky charm. Indeed, we do: Noah, Abraham, Moses ... Jesus ... make no impression.

    Wednesday, June 01, 2005

    Information as a Model of Experience

    Different cultures, different disciplines, different agendas ... the same individual running an alternate agenda ... model the world differently.
    Shannon / Wiener defined information in terms of probability. My theory of Macroinformation also utilizes probability. But Gregory Bateson defined information in terms of difference: any difference that makes a difference: and difference is my Macroinformation's main explanatory referent.
    Zoroastrians pictured experience in terms of light and dark: and many a philosophy since has followed suit. Light and dark soon fuzzes with good and evil, God and Devil. Physicists modeled the universe in terms of matter: matter and force. Modern physicists revised the models so that matter and energy are the main players. Energy won: now they model the universe in terms of energy quanta, energy events: the idea of matter evaporated into energy. Anyone should be able to supply additional examples; I segue to my point: Macroinformation models the world, the universe, experience ... sentiens ... in terms of information.
    In a limited sense my theory of Macroinformation is my theory of EVERYTHING.

    First I urge readers to disabuse themselves of the notion that information and data are synonyms. Information cannot exist without data any more than blood can exist without iron atoms, but by the time we're modeling how blood clots, we've left iron atoms far below us. For example: my first sentence: if information is significant difference, and different cultures model the world differently, then new information issues from these differences.
    Not data. Oh, there's data aplenty: already operating in the different cultures and their different models; no, new information emerges from comparing them. New data of course will follow, accompanying the new mapping. I am making more data here: to illustrate what I see. First I see it - new information; then I describe what I see with new data.

    To talk about what I see I invoke a concept of levels of information. To a limited extent, levels are already familiar with regard to information: any HTML document distinguishes between data and metadata. The data is what I'm saying; the metadata holds instructions on how the data is to be displayed. For example, the mechanism on a typewriter that stops the carriage return at a left margin is a form of metadata, and in this case, it's "hardwired" into the typewriter. If at the left margin, I press the letter A, that's data. If instead I press the tab key, that's more metadata.
    Macroinformation requires more than two levels.
    First, a perception of difference must be possible. To see the photon - to notice it, it has to be dark. Foreground must be distinguishable from background (and an artist, like Degas, who will occasionally blur them, is making very complex new information!)
    Second there's data, marking the simplest differences. The character 1 is distinct from the character 0, the character A is distinct from the character B, The number 1 is further distinct from the letter A; the capital A is distinct from the lower-case a ... There's more than one level within a level, there are differences between Kingdom and Phylum, Phylum and Class ... and further differences within Kingdom, Phylum, and Class. Still, so far, it's all data.
    Third, there's metadata, already introduced. Centered text is distinct from text justified left, test justified right is distinct from both; italic text is distinct from Roman ...
    Fourth, there's metainformation. The way that a noun is distinct from a verb is different - at a higher (more complex) level - from the way that the noun "salad" is distinct from the noun "days." Language, grammar ... meaning ... would not be possible without metainformation. Without metainformation all the data in the universe would be so much sand on the beach. There could be no sentience.
    I repeat: sentience is 100% dependent on the possibility of metainformation.
    Fifth, there's macroinformation. Fifth, sixth, and seventh, there's macroinformation. "Fifth" here is more complex "to infinity."
    (Is it actually infinite? It's premature to say.) (For the moment, I use the word infinity not the way a mathematician does, but the way a photographer does: "further away than my present equipment can make meaningful distinctions about": in the distance.

    When, with our eyes (and our visual cortex), we see a tree, we don't (and can't) count the photons. Neither do we count the medium between us and the tree. Neither do we count the carbon atoms, the nitrogen, the hydrogen ... atoms in the tree. No: we "see" the "tree." Neither do we estimate all the quarks, neither do we estimate all the hyperstrings. We see a mental construct - tree - in the macrouniverse: the universe of the sense that we make of common human experience. But the photons, the atmosphere, the carbon, nitrogen, hydrogen ... atoms, quarks, and maybe hyperstrings too, are there; or there wouldn't be any tree; neither would there be any "we."

    Thus, initially, I discuss and analyze macroinformation the way we discuss and analyze the oak: knowing that I'm omitting huge essential bulks of things.

    The modules at Macroinformation.org, more than three hundred of them to date, go into as many of these things - as I've yet dealt with. Some are well composed; some are just sketches, many unfinished. Many repeat the same illustrations similarly. In other words, Macroinformation is a theory in progress of being formed, and written. The need for the theory has been in my head for several decades; the actual work is now entering its eighth year.

    Even from this brief introduction I trust the newcomer can see that I am arguing that new information may issue from cross-cultural examinations. This information will all be macroinformation: quantumly more complex than metainformation. Art has always done that: generated macroinformation. The musicians set up rhythm here, melody there, a bass here, a soprano there, on the beat here, off the beat there ... The musician's hearers will hear in all cases, however differingly from individual to individual, new information, in their own audio-cortex, emerging from differences between the parts: and the parts will sometimes seem to blend, other times seem to produce friction ...

    Above I cited two nouns as distinguishable: salad and days. I pointed out that their distinction is different from, of a lesser order than, the distinction between any noun and any verb. Ah, but the way Shakespeare used them, in the mouth of his Cleopatra, the distinctions become greater. "Salad days" transcends its grammatical classification, leaps beyond a sub-category of metainformation, straight into sublime, unfathomable, macroinformation. The phrase has been analyzed in literature for five centuries, going into a sixth. But to the best of my knowledge what I say about it at Macroinformation.org has never been said before.
    It's a noun cluster. That's been said. Fowler's English Usage discouraged noun clusters. Should we red pencil it? Delete it from the play? Tell everyone that our greatest poet didn't know proper English? Or should we recognize that the solecism is essential to this example of the highest reaches of literature?
    This part I don't believe has been said: the two nouns, grammatically, in usage, fight each other. It's like shoving President Bush up against a Muslim. They function like two magnets forced together, same poles facing. They repel each other. They are a minimalist, euphonious oxymoron.
    But, saying that they're of the same grammatical class, and that English normally likes to distribute classes, put different classes between similar classes, is misleading. (The rest of this I guarantee has never been said or written except by pk.) Existentially, salad and days are very different. Days are a measure of time: a comfy, familiar measure, common to all human experience. Time, however real, is abstract. It measures in only one of the familiar dimensions. It has length, but no width, no depth, no height. Time comes from forever and goes toward forever, but it has no substance. Time merely endures. Time cannot be stopped from enduring. Days follow days follow days.
    Salad contrasts; yet there are similarities. Salad does occupy space, has dimension, has weight. Ah, but very little weight. Pick up a tear of lettuce, and more people will say it weighs "nothing" than will offer some decimal fraction of a gram with a lot of zeros after the zero point. And indeed, for Elizabethans, salad was the nothing dish: insubstantial, not satisfying, of nourishment not worth considering. (We differ: but that's a different macroinformation!)

    That's enough for the moment: except to say that the way salad fights with days is highly complex, as is the way that salad agrees with days.

    Criticism, all the criticisms, have arsenals of tools. English criticism has irony, anomaly, discrepancy ... I say that there isn't a one of those tools that wouldn't benefit from an informational review. And I say the same about music, painting ... and politics, economics, theology ... The entrepreneur sees a difference between price and value: and exploits it. This, all, is analyzable as information.

    Macroinformation is, certainly in potential, a theory of "everything."

    At least in the universe of conscious thought: which I call Sentiens.

    To what extent that universe is exclusively human remains to be known.

    2005 06 02 I have converted this post to HTML and incorporated it at Macroinformation.org as the new Entrance. Any changes will go there, not here.