Even as a huge Linux tinkerer I've been really pleased with Zo setting things up for me that I wouldn't have expected to get done on a server like this. I've started to experiment with installing Guix and serving Lem, and made a personal Telegram bot as an alternative interface. They really focus on keeping out of your way, and the Zo agent is quite persistent and gets things done that I didn't think was feasible.
The price is right for the specs of two hetzner boxes plus a convenient agentic web interface (I like my Cockpit, but it is honestly so nice to be able to just send a Telegram message to get a service set up or to grab a file just by speaking). I use all the LLM apps but, it's like comparing SMS vs. Telegram, it's just not the same?
I was a bit wary of adding my google accounts at first, but I've tried using it to send me daily images through Telegram and it's actually pretty dang nice!
One thing I would really like though is Whisper on Zo. Oh, maybe I should just get that set up on Telegram...
Sadly, a third of American adults do not understand fractions. A big part of this isn't just the particulars of the educational system -- it is also caused by the way we conceive of and teach mathematics as disembodied, symbolic manipulation.
James Algebra (by Jeffrey James and William Bricken, one of the O.G. researchers in virtual reality) is an 'iconic' notation for K-12 mathematics -- it makes numbers look like what they mean.
I'm working on a mobile puzzle game to publicize this. Check out my WIP at https://house.valeriekim.ca -- though I need to get it to a self-demoing state.
Provides a historical overview and introduction to applications of cybernetics in social sciences.
Crucially, Umpleby's history mentions the exact reason why Cybernetics never quite took off in America (well, aside from the fact that it was just difficult to fit into a disciplinary box). Heinz von Foerster, head of the Biological Computer Laboratory, did not want to provide a b.s. military justification for Dept. of Defense funding after the Mansfield Amendment -- (which, was passed in response to student protests of divestment from the Vietnam War).
Huge thanks for the list! Interesting to wonder just how influential or generally just of what nature cybernetics would be today without funding issues, lack of categorisability and the field slowly drifting away from the original vision Wiener had.
While I agree that nitty gritty formalizations are important to grasp, it's been made clear that the formal sciences were often insufficient in facilitating insight on the type of complexity cyberneticians cared about -- especially that of process oriented circular causality, which often involved paradox and self-reference.
> Shannon was like Spencer-Brown a mathematician and electrical engineer. In his study of data transmissions, he has demonstrated how any medium generates background noise that interferes with the transmitted characters. To this day, mathematics has not perceived or not wanted to perceive the elementary consequences of this for mathematics and logic. To this day, mathematics is regarded as a teaching that is independent of the medium in which it is written and through which it is transmitted. Nobody can imagine that the medium could have an influence on the signs and their statements. Mathematics is regarded as a teaching that is developed in a basically motionless mind.
Having taken the last 2 years to really go through the discourse surrounding second order cybernetics beyond Ashby's introduction and Stafford Beer, I learned of a pivotal text called Laws of Form, which was at the heart of second order cybernetics. The formal system was directly incorporated in Varela and Maturana's thesis of autopoiesis, and Niklas Luhmann was also /obsessed/ with Laws of Form for much of his academic career. This is a progenitor of our current interest of enaction and embodied cognition!
With the book's 50th anniversary in 2019, the discourse has been seeing a rejuvenation thanks to some small conferences at https://lof50.com. I've seen some intriguing applications. Some are a bit far out, but that's the nature of systems thinkers, yeah?
A couple that may be of interest:
William Bricken's work on Iconic Mathematics, a system that covers K-12 math and bridges it to purely physical manipulation, shedding matters of complexity difficult that fuel general mathphobia such as: associativity, commutativity, division by 0, bases, functions, order of operations, the disambiguated meaning of equality. Instead, everything is a /structure/.
Bricken's work on computational implementations of Iconic Logic. One example includes a novel SAT algorithm / tautology verification algorithm called Virtual Insertion, which makes extensive use of the notion of semipermeable boundaries, in which the context of a boundary still pervades its content.
Gitta and Ralf Peyn on FORMWELT, a yet-released system aiming to facilitate precise, clear communication of nebulous natural language concepts through the use of injunction and self reference.
>While I agree that nitty gritty formalizations are important to grasp, it's been made clear that the formal sciences were often insufficient in facilitating insight on the type of complexity cyberneticians cared about -- especially that of process oriented circular causality, which often involved paradox and self-reference.
I wholeheartedly agree. Hence why I think we are dealing with an entirely new type of science, the basic principles and theorems are yet to be discovered.
Lots of interesting pointers and links throughout. Thanks again!
I would say that the first-order Cybernetics (as exemplified IMO in Ashby's book) is all about symbolic formalism for "circular loops in causality", and that Second-Order Cybernetics is a species of mysticism (I hasten to add that I don't mean that in a derogatory way. I'm a Mystic myself.)
You could imagine a spectrum from logic to cybernetics to philosophy to mysticism. It's all fine, just I believe that it's important to be clear where on the spectrum you're working. If you want to build machines that do things use Cybernetics and feedback/control theory, if you want to grok reality and self eat a mushroom and read "Gödel, Escher, Bach" or Tao Te Ching.
In re: "Laws of Form", yeah, he identifies the boundary or distinction between the mystic realm (non-form, non-distinct, non-dual) and symbolic logic, and then builds a lovely binary Boolean logic directly off of that. It's a tour de force.
The really interesting thing is that George Spencer-Brown figured out how to deal with circular logical systems by introducing the concept of imaginary Boolean values.
(That third rule in the calculus, discovered by Bricken, is the kicker!)
cf. "The Markable Mark" George Burnett-Stuart
http://www.markability.net/
GBS (not GSB, they're two different people) has managed to extend the system to Predicate Logic as well!
To sum up, we need symbolic formalism to communicate and build machines, otherwise we're just sort of reading poetry to each other, which can still be helpful, but in a different way than building machines. (And when I say machines in the context of Cybernetics I mean to include those made out of people! "Human use of Human Beings", eh?)
As an aside, there is a fascinating talk and book "My Stroke of Insight: A Brain Scientistʼs Personal Journey" by Dr. Jill Bolte Taylor:
> In it, she tells of her experience in 1996 of having a stroke in her left hemisphere and how the human brain creates our perception of reality and includes tips about how Dr. Taylor rebuilt her own brain from the inside out.
When "Laws of Form" talks about the beginning of logic from the pre-logical non-distinct or non-distinguished realm, the "mark" that divides the world into A and Not-A, it turns out to be quite literal, purely biological: the brain does it. When the part of her brain that "does logic" was rendered inoperative due to the stoke she was having, she reports an inability to tell herself from the world around her accompanied by oceanic bliss...
Mysticism is real, you just can't talk about it. The very first chapter of the Tao Te Ching says it right off: the Tao that can be talked about is not the Tao.
Total random comment, but for posterity's sake -- apparently, Agnikul is organized using a free and open source ERP called ERPNext. There's not been much discussion on it, but I'm doing my best to get my foot in the door using ERPNext and Frappe Framework (what it's built on) in North America.
I figured this 1994 paper would be relevant to the discussion about finished software, "Software Aging" by David Parnas. Read it for a software engineering course where the prof railed against the assumptions underlying the notion of finished software.
I'm part of an electric riding group in my city. We've had several members switch from electric skateboards and scooters to EUC without looking back. Two critical differences that make electric skateboards much worse than onewheels or electric unicycles: control mechanism and wheel diameter.
Electric skateboards and scooters have tiny wheel diameters. Especially with esk8, it's very difficult to go on any rough terrain: pebbled asphalt, cobblestone, grassy park, gravel, sand, dirt road, up or downhill. Scooter wheel diameters are also a safety issue due to that combination of easy brainless riding and small wheel diameter.
Electric skateboards and scooters interface with your hand, but EUC and OneWheels interface with your balance - your whole body. There's an intuitive sense of where you're going, and you have many degrees of freedom along your entire body that you can leverage to maneuver for incredibly fine tuned and synchronized control of your acceleration to the point where it feels like flying. The fact that acceleration and balance are intimately tied together without your hands interfacing eliminates a huge category of problems from accelerating or braking too hard vs. your centre of gravity. When you're riding in any situation where hand dexterity is compromised (i.e. wrist guards, considering none of these vehicles are any safer for your wrists should you fall), this is an amazing preventative safety element.
The "Easy ASP" [0] tutorial from Potassco can walk you through some examples, if you'd like.
The playlist is aimed at a general scientific/business audience, the presenter suggests that a lot of natural and business systems can be described in this manner. The presenter also mentions how a Clingo program was used, without modification, to optimize radio frequency band allocation.
Here's a repository [1] of ASP programs in clingo. Under problem classes, I see mostly: game AIs, graph problems, various puzzles, so on.
Not sure if related, to Schaefer's theorem, but I dove into Answer Set Programming [1] recently, which follows this approach, enabling the use of fast-ish SMT solvers, which are a generalization of SAT solvers! Boolean/Propositional Logic is to Predicate Logic as SAT is to SMT. There's a very nice course about it from the developers of Potassco, one of the best open source Answer Set Programming framework [2].
The syntax looks like Prolog, but predicate negations are a first class citizen, avoids infinite loops.
Prolog's approach is like a depth first search through a search tree -- ASP is like a nondeterministic turing machine, exploring all branches simultaneously from the bottom up.
The price is right for the specs of two hetzner boxes plus a convenient agentic web interface (I like my Cockpit, but it is honestly so nice to be able to just send a Telegram message to get a service set up or to grab a file just by speaking). I use all the LLM apps but, it's like comparing SMS vs. Telegram, it's just not the same?
I was a bit wary of adding my google accounts at first, but I've tried using it to send me daily images through Telegram and it's actually pretty dang nice!
One thing I would really like though is Whisper on Zo. Oh, maybe I should just get that set up on Telegram...