asystemic

paradigm, for each problem, calls for discovering or evolving “a holistic unit of being” doing what we want — something elemental thats simply there when needed. An ideal asystem is simple (or its complexity is irrelevant), structureless (or its structure is irrelevant), modeless (at least on the outside); often fractal, i.e. appears or acts similar at different scales; most importantly it is inevitable — works because, put together this way (not necessarily exactly), it cannot but work. This contrasts to the old approach of composing a system out of (themselves often systemic) actors, each with its own isolated objective; a systems functionality comes as an orchestrated culmination of actors' state transitions, each critical to the ultimate success. ■    Both systemic and asystemic tools can be reasonably standalone and self-maintaining, but on efficiency and (even, counterintuitively) reliability, systemic ware tends to win — at least in the short term. And yet, starting with the electronic-age aversion to “moving parts,” the systemic mindset — “lights aflash, buttons amass, wheels and cogs and pins and brass” — slipped from favor: over time, even the best systems accumulate imbalances, suffer avalanche failures, require ever-costlier maintenance; just as importantly, evolving systems suffer from “creeping bloat” — complexifying for short-term gains at the price of long-term adaptability. Systems are multiplicative: any components failure can zero the outcome, fragility grows with size — making, sooner or later, redesigning from scratch inevitable; by contrast, an asystem (when complex at all) is additive: you may be able to improve it by tweaking parts but theres little risk of disabling it outright.  ■    Lazyball remains the archetypal asystemic solution: a discovered phenomenon you cant disassemble, upgrade, break or fix — it simply is; knowing its properties, we get better use out of it than any systemic device. In evolving organisms, millions of generations grind down the systemic joints; Change furthered this trend in humans, stoked the already flaring “asystemity revolution” in technology — and demonstrated that one road to asystemity goes, paradoxically, through steadily increasing complexity until that complexity “conquers itself” and becomes irrelevant. The post-Change human isnt any simpler and may not even be more efficient in absolute terms (e.g. the sense of time only grows more subjective as the speeds of biological processes diverge) but is vastly more self-sufficient — “unbreakable.” ■    More than any single organism, the whole planets biology with its evolutionary adaptability is the ultimate asystem: once “put together” a certain way, it “cannot but” survive within a very wide environmental corridor. Similarly, Arf as a whole aspires to the asystemity ideal of a structureless, hard-to-disrupt entity that cant fail to take care of itself; a single arf cell — a legacy of much older times — is still largely systemic but theres ongoing work to simplify it and replace complex quasi-intelligent subsystems with something “as reliable as laws of nature” — with “first-principles alignments” to be discovered rather than evolved (or at least evolved from a discovered core).  ■    The asystemic approach dictates not only how but what is or isnt being solved. When something appears to be calling for a systemic contraption, “step back and look for a general problem solvable asystemically”: dont even solve but search for something “simply there” under which your original problem reduces to a side effect — or disappears. Thats a major line of critique from the systemity counterculture: stop being lazy and proud of it, stop blindspotting or downplaying challenges for which we dont have an asystemic key; whats worse, asystemity may be eroding our criteria: in striving to emulate the ideal of “cannot but work” were willing to relax — dangerously so — our standards of “work.”

< artificium  |  axial tilt >

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License