Pace Layering: How Complex Systems Learn and Keep Learning | Hacker News Reader

Pace Layering: How Complex Systems Learn and Keep Learning

jods.mitpress.mit.edu

96 pointsakakievich9 comments

A few supplemental links that may be of interest to the HN crowd:

A Pace Layering seminar: http://longnow.org/seminars/02015/jan/27/pace-layers-thinkin...

Doc Searls riffed interestingly (in 2008!) on pace layering (aka "Layers of Time"), the Internet, the Burton Matrix[0], and Open Source: http://commons.oreilly.com/wiki/index.php/Open_Sources_2.0/B...

[0] Craig Burton's Open/Closed Proprietary/Public Domain matrix isn't often explicitly named, it's insights having been so thoroughly absorbed by our community: http://www.linuxjournal.com/content/understanding-infrastruc...

What are there tools that help simulate complex systems? Things like biochemical balances in the human body, economies, enzyme actions and homeostasis, environment and climate? Etc etc...? What tools are used by those in the profession?

That sounds like an incredibly broad domain of tools.

The tools are probably probably domain specific as well. But are there any standard/OTS commercial tools for this?

I guess simulations are commonly just large number crunching machines based on the matrix-algebra, but for example fluid-simulations can gain quite a lot of power from algorithms that divide the space into sub-cells and focusing on areas with more detail.

Agent-Based Modeling is an approach to understanding complex systems by simulating interactions between actors in a population. The actors can be modeled with varying properties, and the output of the model often produces emergent behavior.

https://en.m.wikipedia.org/wiki/Agent-based_model

Check out the NetLogo ABM environment for several pre-built example models.

http://ccl.northwestern.edu/netlogo/

The following research that uses Harel Statecharts for cell-level modelling may be of interest:

Bloch N, Weiss G, Szekely S, Harel D (2015) An Interactive Tool for Animating Biology, and Its Use in Spatial and Temporal Modeling of a Cancerous Tumor and Its Microenvironment. PLoS ONE 10(7): e0133484. https://doi.org/10.1371/journal.pone.0133484

Intresting stuff

The author tries to develop a perspective on the interaction of spaces over time. It is a long stream of consciousness republished from a decades old piece. A few examples are used to identify the idea and model the theory onto the whole world from a social point of view. In the process he reinvents the idea of homomorphisms. Fashion, as is quickly diagrammed, would be the terminal element in a category of ordered group actions up to isomorphism, reacting to catastrophe by isomorphisms as isomorphous shielding layer without modifying the underlying, generating functions, in fact preserving them. Without side-effects.

Only in the end an old example of different levels of longevity of a building are pointed out, revealing that the whole issue concerns planning and that "layers" are meant literally and figuratively.

Critically, that seems to be exactly the weakness of the exposition, it reveals no real insight. The choice of terms is either mundane (literally, cf. french monde) or subconsciously influenced by the overarching theme of layers, as in layers of fashionable clothes.

In essence, the whole article tries to frame the question to "explain the mechanism" to "absorb shocks and in fact incoporate them". And the answer, without it being named, seems to be hierarchy ... (when in fact, that couldn't be further from the truth. It's a state analysis, not dynamical. It's also ideologically colored, reiterating the old themes of the invisible hand of the market and the wealth trickling down.)

Disclaimer: I'm just joshin' around entertaining the idea to translate the article to abstract maths. See, everyone can write absurd amounts of text to say very little, but not everyone is going to read it. And not everyone can keep it short because it feels soooo god to hear yaself talkin.

Reply on news.ycombinator.com