[FRIAM] new math of complexity

[glen]

Without looking at an actual publication by Rosas et al, I can't be optimistic or pessimistic. But I can object to Ball's examples of the 1st 2 types of closure: information and causal. The examples he lists: predictability of laptop output and controllability of laptop output are obviously flawed. The former is fairly easy to disprove. The typical aphorism is that all software contains bugs (barring the infrequently used formal methods). We have this entire, I think dominant, conceptual category of "bug" that should demonstrate a lack of that kind of "informational closure" ... at least in the wild if not in the lab.

The objections for the latter are more persnickety. Again, in the lab where the unprofessional techie can get away with saying things like "Well, it works for me" because there are detailed, documented methods. And if "it doesn't work for you", then you're just doing it wrong. But in the wild, "it works for me" is entirely inadequate. So, again, we have an entire conceptual category grown from the lack of causal closure in laptop controllability.

Maybe, though, this is simply validation of Ball's suggestion that life (or deeply interactive computation) is leaky ... maybe even very leaky such that any mathematical definition of computational closure we compose, it's Platonic, merely a useful fiction.



On 6/12/24 12:43, steve smith wrote:
> 
>> Speaking of emergence, any takes on Phillip Ball's article in Quanta?
>>
>> https://www.quantamagazine.org/the-new-math-of-how-large-scale-order-emerges-20240610/
>>
>> I really liked his summary of the current non-explanations for emergence, but I haven't had time to read further.
>>
>> -- rec --
> 
> 
> As a member of a group here roughly described as "Complexity Groupies" I am heartened to hear Ball's acknowledgement that "nobody" really seems to have a good explanation of "what emergence is".  It feels parallel to art and pornography in the sense of "I don't know what it is, but I know it when I see it".
> 
> Terrence Deacon's classifications of dynamic systems seems to provide some insight or fine structure to emergence, though I don't know if it is widely interesting or helpful to others.  He applies it primarily to life unto consciousness studies.  It seems particularly apt to Ball's references to "heirarchical" systems while his references to "leaky" emergence rhymes (a little for me) with Herb Simon's "partially decomposable" systems.
> 
> As an aside, I don't think I would have recognized Crutchfield... I haven't seen him in person since about 2009 when he was doing an art-project with Woody Vasulka and microphone/speaker/ambient-space dynamical systems at the old bank building downtown?  Or maybe it was a few years earlier...
> 
> Deacon's classification system:
> 
> *homeodynamic:*A system is homeodynamic if its spontaneous, natural or unforced path leads towards equilibrium. Homeodynamics erases differences (e.g., in temperature or pressure).
> 
> *morphodynamic:*A system is morphodynamic if it tends to spontaneously increase in order. This generally involves external perturbations, but does not involve external design or imposition of form. Morphodynamics subsumes many standard examples of self-organization. Morphodynamics amplifies differences.
> 
> *teleodynamic:*A system is teleodynamic if its organization becomes spontaneously end-directed. Teleodynamic systems employ homeodynamic and morphodynamic processes in the service of a self. Terms like ‘self-maintenance’ and ‘self-repair’ become natural and unavoidable in teleodynamic systems.

-- 
ꙮ Mɥǝu ǝlǝdɥɐuʇs ɟᴉƃɥʇ' ʇɥǝ ƃɹɐss snɟɟǝɹs˙ ꙮ