feedback loops and Saturninus

I keep thinking “feedback loops are a good model for considering or defining things.” Why? Is there something universally solvent about it?

Anyway, the useful consideration here (when aiming to define the nature of the feedback loop holding a thing more or less taut in reality): What is it about the natural behavior of the thing that elicits a sustaining response from the environment?

The severe responses of Saturninus breed in Rome an unwillingness to challenge his madly egotistical will (we take it for granted that all things have their Tamara).

Bochner and Completeness

Frustration has become a key response to certain recent art. Frustration because the viewer is looking for a complete idea and is foiled. The notion of completion (i.e., self-containment) is at fault. What is thought and what is experienced continually replace each other.

Mel Bochner, taken from a catalog statement for Art in Series at Finch College, 1967

And in response I wonder: what then of the preceding generation, who strove for completeness and autonomy of the object through that completeness?

We should acknowledge the impossibility of this notion, perhaps scoff a bit at its hubris, but quite impressive is just how complete many of the works were in comparison to just about anything else – what an achievement!

Naturally this sort of completeness (of information, logical support, or definition, etc.) is not attainable in every domain, and certainly not in language. Bochner’s work quite obviously makes its money by exploiting that. There are precise rhetoricians, to be sure, of startling (if not discomforting) objectivity and singularity of interpretation. Granted, language is a relational system to the core, so it is certain to never be as “objective,” or more accurately “empirical,” as something akin to a Pollock drip.

It strikes me as irresponsible not to foster the relationship between idiosyncratic expression and improved comprehensibility, rather than dwell in the difference (originality and comprehensibility are not mutually exclusive, after all).

“A large portion is beyond control”

Melville sets out to write a fully accurate sense of whaling, essentializing and hyperbolizing the hunt to construct his story. In this abstracting process the thematic structure of the novel becomes quite receptive to metaphorical overlay. Interpretations abound- the white whale representing perfection, completeness, immaculate revenge, ineffability- these are both present and projected depending on the interpreter’s idiosyncratic knowledge and experience. Even the intended achievement, whatever that may be, is subject to this. To say that one allegory or metaphor stands over another could be compared to considering the purest potential of a quantum system without the definitiveness of its context. The feminist and psychological and religious interpretations of the book are hardly different in nature from the exacting, exceptional experimental setups that reveal but one property of a quantum system at a time, contingent factors constrained, amplified, or nearly excluded. The fuller truth may be impossible to address.

An origin of progress in a formal domain of thought

It is interesting to realize that what are considered to be objective thoughts and programs, in the proper sense, undergo great progress via the genius of idiosyncrasy. The way in which a concept is employed is related to the semantic framework of the thinker. It is put in relation with particular attitudes, processes of learning, other understandings, etc. The unique understanding of a concept may facilitate new insights, and through the application of those insights: progress. This is what Bloom calls a “strong misreading”, at least in terms of my strong misreading.

now how, until silly

Some considerations for the primary search and display algorithm of a software system I’ve been working on:

In order to be sufficiently efficient, the algorithm designed to retrieve notes relevant to a query must restrict itself to a narrow subset of all archived Notes. An index of every instance of every token in the archive is regularly maintained, the entry for each token instance an array containing the unique ID of the parent note and the semantic weight of the instance relative to that parent. With the initial query parsed into smaller tokens, the algorithm rifles through the Index in search of a) direct matches, b) grammatical variants, and c) semantically related terms (though it will be feeble at first, continual use of the tool and careful appraisal of search results will refine the program’s “semantic sense”. More sophisticated algorithms won’t hurt, either)

Suppose an archived note N₁ is returned for a particular query, Q. It is for whatever reason arbitrarily decided that N₁ should no longer be returned under Q. What to do, without disrupting the elegance of the search process? Though at first it seems to be a potentially sloppy thing to work around, I think there might be an elegant solution nestled right within the framework already in place.
A few calculations are performed in order to decide what to display. If Q is the set of all returned notes {N₀, N₁, N₂… N_n}, themselves each being a set of tokens, the program first calculates the union of all notes in Q, U(Q) = N₀∪N₁∪N₂∪…N_n, then the intersection I(Q) = N₀∩N₁∩N₂∩…N_n. Both of these calculations would involve the relative weight of each token instance. The coherence of Q, and by abstraction Q, is then expressed as a ratio: I(Q)/U(Q) which can easily be quantized. Now we are getting somewhere. The coherence of each note relative to the collection Q can then be evaluated as N_n∩I(Q)/I(Q), meaning “the % of features this note has in common with the average shared features of the entire search”. This value is compared to a threshold, calculated by some yet-to-be-drawn relationship between the coherence of Q and a sensitivity arbitrarily set by the user. (The idea is that this sensitivity could be adjusted on the fly, frequently, similar to depth of focus with a camera lens). If the coherence of N₁ is below the threshold it isn’t displayed. Simple as that.

When a note N₁ is retrieved and determined to be above the display threshold, but the user indicates N₁is insufficiently related to Q, the computer does some special background work. Though there is much more to think through, I imagine the relative complement I(Q)\N₁ , the tokens possessed by the average of Q but not to be found in N₁, would be used to redistribute the semantic weight of N₁‘s tokens until the terms relevant to Q fall below the threshold for initial retrieval.

There is a quagmire yet to be resolved if any of this is to actually function. The initial search routine combs the index for tokens determined to be relevant, relying on the “semantic weight” of the instance. In my thinking this value was always relative to the parent note, as opposed to the domain of the search query. Solving the issue of dissociated notes brought the inadequacy of the indexing scheme to the immediate awareness; a philosophical solution precluded a principle of efficiency. How might earlier thoughts relating frequency of word use to domain of analysis address this new situation?

thing as a thing

A thing’s status as a thing depends on the context, depends very much upon the domain {from, in} which one addresses it. One system’s native definition of a thing may preclude the existence of some other system’s definition. This is as true of communication in computing scenarios as it is in biological ones.

In a less damning way, it may be noted how much of a struggle it can be to communicate with someone quite apart from your own way of thought (“orientation”). The amount of clarification or definition some terms and phrases require in order to be functional may be daunting.

When a word is used in conversation, the meaning could be mis-perceived- perhaps because it has multiple meanings, or perhaps gradations of meaning are determined (to various degrees) by context, whatever that means.
The term with lowest possibility of mis-perception is usually the more dependent term overall. It is a rod in a structure much more “abstract”, much larger, than any sentence. It achieves its symbolic clarity by way of rhetorical uniqueness. Necessarily, these terms don’t kick around in the average lexicon, and accordingly require specialized knowledge that must be acquired externally.

This thought translates into computational linguistics:
0. assemble a vast textual corpus
1. analyze the frequency of word use
2. for words used most frequently, restrict the domain of semantic analysis to the sentence or less.

The domain of analysis is inversely related to the relative frequency of a term’s use. Terms with the highest frequency of use, “the”, “of”, “a”, are practically self-evident. The more recondite terms refer back to and rely upon a Mode, the sum of which is diffused across many papers, notes, and heads. To “make sense of” a high frequency term the domain of analysis begins small and expands outwards. The inverse is true of hapax legomena.

The Ocean and the Self

In Emerson, Whitman, and Stevens, there is the image of the sea- the vast mother shored by the father and abstracted by the clouds of the intellect. These abstractions are words, once crafted by an original, vital poetic voice and employed later as trope by the speaking poet and those who come after. As Emerson writes in The Poet “Language is fossil poetry.”

Even in artificial languages, the Word (in the Sausserian sense) is an idiosyncratic product of one Self- the product of one collection of interrelated abstractions, one collection of sensory impressions influenced by extant tropes (a Whorfian stance reinforced by the research of Boroditsky). The metonymy of the ocean is a manifold sensory impression uninfluenced by the semantic framework of an experiencer’s mind. It is, of course, an ideal. However, it is something we, each and all, may approach asymptotically. Wading into the ocean, new beholdings attune one to inadequacies in the impressed frames, and in an outpouring of originality, a new, more effective term or usage might be expressed. This is the genius of the painter, the philosopher, the mathematician, and the poet.

New for me is the resolution of Whitman’s use of “the soul” with the notion of “semantic framework” and, again, this ideal of experience, “the ocean”. “Spirit” is something different. “Soul” is the unique semantic framework, the unique hash value over the complete file contents, the signature of a singular character. It floats above the ocean as a system of clouds, independent of the water but inescapably of it. Emerson (at least in the earlier years, as seen in Essays, Second Series (1844) and the poem Seashore (1856)) holds that, being so inextricably of it, we cannot “sing beyond the genius of the sea,” as the singer in Stevens’ The Idea of Order at Key West(1936) did. The nature of Mathematics, too, seems to suggest that we may work beyond our origin, if only in the vapors of abstraction.

…though the voice that is great within us cannot be our own, we are under the transgressive necessity of being able to locate it nowhere else.

Harold Bloom, Ideas of Order , 1976

more vaporous wanderings re: probability theory

I have some deep suspicions. There is a strong relationship between the normal distribution, Probability Theory (both as concepts as opposed to applied frameworks) and the semantic apparatus of the mind.

Salience develops together with the ascription of meaning/senses beyond those of immediate experience. There is a complex association between words, appending notions and stitching relationships. On a more elementary level, there is also repetition and coincidence across the immediate sense inputs. The behaviors of these associations, their resolution and realization, can be described quite elegantly with Bayesian logic, among other statistical programs. Our brain, all brains, are elegant heuristic devices for creating discreteness.

I’ve recently come to appreciate the true fullness of the abstraction mathematics represents: to know that the Bayesian logic of the mind relies on a more primitive logic, one that follows the physical, and that this most fundamentally discrete logic also gives way to something else.

thinking/sloshing about Probability

‘The Law of Large Numbers’ and ‘Central Limit Theorem’ are representations of Probability Theory. According to Jaynes, the behaviors of any system in question appear “random” because we are not privy to the complete composition, behavior, and influences upon that system. Given many examples of the given system’s behavior, however, certain trends stand out- characteristics more essentially of the system than others; this may be more accurately stated: “characteristics definitive of the system”.

Obviously, Probability Theory was a coping method acquired by the developing ocean beast simply to function more effectively in its unfurling World.

More strangeness radiates from the unusual, inexplicable asymmetries of Thermodynamics when ‘statistical mechanics’ is regarded in light of this take on Probability. Deep quantum understanding and non-relativistic physics do a fine job of indicating the fundamental possibility or truth of bidirectional “time”. However, action in the universe proceeds unidirectionally; this is a basic tenet of Thermodynamics. How curious that this framework of understanding, to our best knowledge, is inseparable from the epistemological workings of Probability.

(sounds of 4/4 strumming)

My neighbor, in the apartment above my room, listens to some lame music. Every twist, chord change, bridge, and snare is utterly predictable.  By mishearing a song earlier, I tricked myself into hearing the music change into something akin to the band 311- I was immediately struck by its funkiness and character. I do not have the most positive feelings about 311, and that I reveled in its qualities should indicate how bad the neighbor’s music actually is.

This is not all to say that uniqueness is of virtue in its own right- that would be a grave, shallow misunderstanding of originality.

Because the neighbor’s music is so homogeneous, and of such undifferentiated presence in the sphere of music, little power inherent to “the literary artifact” is realized; it appeals basely.