wildiris

2.3K posts

wildiris

@wildiris19

Retired engineer: embedded systems hardware design. Education: physics and math. Interests range from Agricultural Robotics to Origin-of-Life questions.

California, Monterey Bay Katılım Mayıs 2022

176 Takip Edilen409 Takipçiler

Sabitlenmiş Tweet

wildiris@wildiris19·6 Eki

Nested agency and grammar-based language models (39/n). • Formally titled as “Nested pushdown automata and context free grammars (1/n”). Renaming to better reflect the design project’s origins as an exploration into physical computation.

wildiris@wildiris19

Nested pushdown automata and context free grammars (1/n). I am one of those people that needs to build something before I feel I understand how it works. In an effort to take a deeper dive into the Theory of Computation I’ve put together a nested stack machine in a FPGA. At this point the hardware seems to be working. Now the journey ahead is to start programming this bad boy and see how such an alternate processor architecture will work out in real world applications such as robotics.

English

1.6K

wildiris@wildiris19·18h

Cesar Chavez was a by-the-book union organizer. He treated illegal immigrants with the same distain as the Teamsters or Longshoremen's unions treated "scabs." With this move, the progressive liberal left of the Democratic party is signaling that traditional blue-collar labor union members are no longer wanted nor needed. Don't be surprised if other traditional blue-collar unions, like the Longshoremen, get "thrown under the bus" next.

English

765

wretchardthecat@wretchardthecat·20h

There is now a rush to expunge Cesar Chavez from progressive history. The statues, the USNS ship, the streets, the holidays. All must go for they are reminders not just of Cesar's failings but of those who made his myth. Yet surely his historical impact, if genuine, is objectively separable from his personal life. Many consequential figures, some of whom saved their city or purged some great evil were personal creeps. Napoleon was by some accounts the most baleful figure in modern history until Hitler. Perhaps the old way of thinking was better when a man's impact on history, which we could know, was separate from his standing before God, which we might guess but never know for certain. That is certainly better than the Orwellian practice of creating truth to order and then when fashions change, of dropping it down the Memory Hole. latimes.com/california/sto…

English

wildiris@wildiris19·21h

I live in a farming community and have a long time interest in agricultural robotics. The "right to repair" is a major legal issue facing farmers. May I ask, were does Menlo Research and "Asimov" stand on this question? Commercially, it's a deal breaker for any large scale deployment of autonomous robots.

English

RoboPapers@RoboPapers·2d

Robotics research is moving fast, and being able to modify and improve upon hardware is crucial to maintaining velocity. That’s why Menlo Research has started working on their own open-source humanoid project, Asimov @asimovinc. And they are moving fast. It’s been roughly six months since they started the project, and they already have full humanoid with arms, legs, and a head, which can walk forwards and backwards. @selim__1903 and @AlexE_00 of Menlo Research join us to talk about the development of this open-source humanoid. Watch episode 67 of RoboPapers, with @chris_j_paxton and @DJiafei, now! And follow @asimovinc for daily updates on humanoid robot development.

English

10.6K

wildiris@wildiris19·22h

@thesuperegonaut Welcome to the world of AI generated content.

English

Gabriel Stuckey@thesuperegonaut·22h

Are words dead on arrival if they lack the authoritative threat of a conclusive justice? What end is there for a story without logos? 🤔

English

wildiris@wildiris19·1d

Now why would a cable harness even need replacing? Once installed, there should be no reason to even touch, let alone replace, such a component. Did a wire break from too much handling? Or have some of the connector contacts gotten corroded from too long an exposure to salt air? This is not as trivial as it is being made out to be.

English

Richard P. Gallagher@rpg571·1d

Rollout Number Two for SLS Artemis ll NASA’s Kennedy Space Center teams are preparing for the rollout of the Space Launch System (SLS) rocket and Orion spacecraft from the Vehicle Assembly Building (VAB) to Launch Complex 39B tomorrow evening, Thursday, March 19, 2026, beginning around 8:00 PM EDT. Earlier this year the stacked vehicle was rolled back from the launch pad to the VAB after a helium flow issue was discovered in the upper stage (Interim Cryogenic Propulsion Stage). Engineers traced the problem to a defective seal in a quick-disconnect fitting and completed the repair in the controlled environment of the VAB. During final closeout preparations, technicians also identified and replaced a faulty electrical harness tied to the core stage’s flight termination system. The swift fix allowed NASA to recover the schedule and return to the original Thursday rollout target, preserving the opportunity for a launch attempt as early as April 1. #ArtemisII #NASA #SLS #Orion 📷@rpg571 for 📺What About It @FelixSchlang

English

5.9K

wildiris@wildiris19·3d

Robert Trivers had many original ideas/observations that he never published. Unless you were fortunate to have heard him speak, whether in a lecture, conference, or conversation, these ideas may sadly disappear from memory. • It would be a wonderful opportunity, right now, for those that were close to him and that still remember a lot of his original work, to put these unpublished ideas down in writing and give us a volume of his “collected unpublished works.” • My personal recollection was his thoughts on parasites and their role in the evolution of species. I only heard them twice, both times in the forum lecture format. They went way beyond what W. D. Hamilton ever proposed. But as far as I been able to tell, Robert Trivers never formally published any of this.

English

264

Steve Stewart-Williams@SteveStuWill·5d

RIP Robert Trivers - the Einstein of evolutionary biology and one the greatest thinkers of our age. Among other things, Trivers came up with parental investment theory, reciprocal altruism theory, and parent-offspring conflict theory.

English

267

1.2K

191.6K

wildiris@wildiris19·4d

@datagenproc Yes indeed. But, IIRC, eventually Newton personally asked him to leave out of fear for his safety.

English

jsd@datagenproc·4d

TIL Trivers was Huey Newton’s thesis adviser?!

wildiris@wildiris19

Was at UC Santa Cruz as a grad student when Robert Trivers was there. To describe him as a "character" would be an understatement. He was a member of the Oakland Black Panthers. Huey Newton's thesis adviser. And Newton was godfather to one of his daughters. I still remember my first impression, upon my first encounter with him, was Leisure Suit Larry. But his forum lectures were amazing! He was Bret Weinstein's undergraduate mentor during Bret's stay at UCSC. Interestingly, Angela Davis was on the faculty at UCSC back then too. I've often wondered if Bret had any interaction with her through Robert Trivers.

English

257

wildiris@wildiris19·4d

I remember him lecturing on the generational time scales that must be at play for natural selection as an evolutionary process to work. • That is, whatever drives evolution must appear as a series of alternating cycles of population winnowing, to select out traits, then followed by a period of population renewal, to reinforce these new traits. • With such cycles taking place on the scale of a few generations each. • The traditional idea of slow, incremental change doesn't work mathematically. • Thinking this way rules out forces that vary on geological time scales. Which leaves what? • Don't know if he ever published anything on this. But would be extremely relevant to any treatment of evolution as an emergent process.

English

Paul Sas@hormetic·5d

Robert Trivers z''l His nimble theory of Reciprocal Altruism ain't even 1/64th of his prodigal fountain of original ideas An OG trickster Documentary bio on @hooplaDigital

wildiris@wildiris19

English

188

wildiris@wildiris19·5d

English

3.9K

wildiris@wildiris19·5d

Depends, if you think schematics should be pretty, or should be useful. When troubleshooting a PCB, while looking at pages of schematics, it helps if the part decal pinouts match what you see on the board. That way you don't have to do repetitive mental arithmetic every time you want to check a different signal line with your oscilloscope probe. It should be required experience for every PCB designer to spend time as a bench tech, having to troubleshoot and rework the PCBs they designed.

English

192

BlindVia@blind_via·6d

Is anybody using this pin combination feature yet in KiCAD PCB yet? V10 is available now. Still trying to figure out my tool bar problem that I had, there are a couple notes about it in the issues list, but I still new to KiCAD and trying to figure what some of the notes mean

KiCad PCB@kicad_pcb

@i2cjak New V10 features coming...

English

4.6K

wildiris@wildiris19·6d

You're correct. Systems don't have to be cooperative to form a collective intelligence. Interestingly, competitive systems (even ones at war with each other) will also form collective intelligences. • Regarding FEP, if you've followed me for any time you'll know that, while I think the FEP is a useful model, in the end, it's not a blueprint that can tell you how to build something. • And regarding QNLP, good link! • But you don't need quantum computing to work with natural language. Nested stack machines along with indexed-grammars potentially also offer a suitable platform. Which is why my interest, as a hardware developer, is in them. en.wikipedia.org/wiki/Nested_st…

English

wildiris@wildiris19·6d

The authors and I start down the same path but soon go separate ways. • My interests are in hardware design and what it will take to engineer physical systems that can instantiate agency. • For this reason, I end up asking different (nontypical) questions about agency. • The authors focus on a statistical analysis between semantic and syntactic information. But what I’ve found with hardware design is that physical systems that instantiate agency are far more complex than the author’s basic mathematical analysis can be stretched to cover. Observations as I read through the paper... 1). Agential systems don’t want to stay isolated. They will naturally coalesce cooperatively to form more complex extended agential systems. So, in practice what you find are hierarchical systems of nested and distributed agency. • When dealing with systems of nested agency, there is no single clean boundary that will unambiguously separate semantic from syntactic information. One nested layer’s semantic information is another layer’s syntactic information. • That is, the same information can/will take on either designation depending on which nested layer is recognizing it. I’m not sure the author’s statistical analysis can be extended to include this working situation. 2). Also, semantic information takes the form of modifiers, nouns, and verbs. While each of these semantically can be represented by a single symbol (“word”), they each correspond to categorically completely different syntactic objects. • For example, a “noun” might be seen syntactically as a particular static pattern in a memory state space. • But a single “verb” can/will correspond to a sequence of choices spread out over time, which could be hundreds of lines of code with dozens of conditional sub-branches. royalsocietypublishing.org/rsfs/article/8…

English

121

wildiris@wildiris19·13 Mar

The authors and I start down the same path, but soon go separate ways. • My interests are in hardware design and what will it take to engineer physical systems that can instantiate agency. • For this reason, I end up asking different (nontypical) questions about agency. • The authors focus on a statistical analysis between semantic and syntactic information. But what I’ve found with hardware design is that physical systems that instantiate agency are far more complex than the author’s basic mathematical analysis can be stretched to cover. A couple of early observations as I read through the paper: 1). Agential systems don’t want to stay isolated. They will naturally coalesce cooperatively to form more complex extended agential systems. So, in practice what you find are hierarchical systems of nested and distributed agency. • When dealing with systems of nested agency, there is no single clean boundary that will unambiguously separate semantic from syntactic information. One nested layer’s semantic information is another layer’s syntactic information. • That is, the same information can/will take on either designation depending which nested layer is recognizing it. I’m not sure the authors statistical analysis can be extended to include this working situation. 2). Also, semantic information takes the form of modifiers, nouns, and verbs. While each of these semantically can be represented by a single symbol (“word”), they each correspond to categorically completely different syntactic objects. • A “noun” might be seen syntactically as a particular static pattern in a memory state space. • But a single “verb” will correspond to a sequence of choices spread out over time; which could be hundreds of lines of code with dozens of conditional sub-branches.

English

wildiris@wildiris19·12 Mar

Looks intriguing. Will read through it today, if I can. Based on the abstract, I think I might be sharing a sizable Venn diagram intersection with the authors. In my post, I was reluctant to make the association of Shannon with syntactic and Landauer with semantic; even though that’s where my ideas want to go. But it looks like that’s what these authors are going to do. Will keep you posted.

English

wildiris@wildiris19·12 Mar

(3/3) For too long now, mathematical theories of information have been used as a basis to characterize and understand agency. • But the physics community, in defining the mathematics of information, has already presumed the existence of agency. • If there seems to be a circular argument at play here, there most certainly is. No doubt this explains the lack of progress that @CAS_ReproLab senses in his post. • Shannon treats "information" as a thermodynamic potential; that is, as a sum over states within a bounded state-space system. In which case, agency becomes a passive dynamic response to gradients in that potential. • On the other hand, Landauer treats agency as an active process with the designation of "information" being reserved for specific measurable patterns appearing in an agent’s environment. • In a sense, Shannon is an “acting-globally” interpretation of agency. While Landauer is an “acting-locally” interpretation of agency. • But, until physics can offer some kind of a comprehensive theory of agency, these will remain open questions.

Lars Herrmann@herrmann_l0578

The gap is real. Every framework assumes information first, then action. But in complex coordination the sequence runs the other way — someone has to move before the system produces the signal. Agency without a theory of agency is the oldest unsolved problem in project physics too.

English

330

wildiris@wildiris19·13 Mar

I spent a decade of my engineering career in medical device design. Talk about "being in the kitchen and seeing how the sausage gets made." There are so many things that could be done to reduce the costs of healthcare; ..but, for too many political reasons, aren't going to happen. Tort reform in the USA's legal system alone could bring costs down by 20%, easy. Also, the USA is responsible for roughly 90% worldwide of all new medical procedures, devices, and drugs. USA's healthcare costs are thus supporting the medical R&D budget for the rest of the world. I could go on, but I'll stop here.

English

Cameron A. Schmidt, PhD@CAS_ReproLab·13 Mar

I often see healthcare costs blamed on 'big pharma' or 'insurance'... No one really talks about how plastic petri dishes suddenly cost $1000 per box when they're marketed for IVF use🤔

English

165

wildiris@wildiris19·11 Mar

@herrmann_l0578 Exactly!

English

Lars Herrmann@herrmann_l0578·11 Mar

English

371

wildiris@wildiris19·11 Mar

(2/n) Information is not a thermodynamic quantity like entropy. Though the mathematical similarities make that a seductive assumption. • The term information gets used at least three different ways in physics. There’s probably more if one stopped to think about it. 1. There is information in the Shannon sense. 2. There is information in the Landauer sense. 3. And then there is whatever the cosmologists and blackhole theorists are doing? • Information in the Shannon sense presupposes the existence of an encoder and decoder that sources and sinks information flow. But the encoder and decoder are agential systems that therefore must preexist any Shannon information. • Information in the Landauer sense is the Maxwell’s demon’s view. Again, Maxwell’s demon is the archetypal agential system. • In physics, agency always precedes information. But physics has no theory of agency. @CAS_ReproLab

English

176

wildiris@wildiris19·10 Mar

Yes, but it still doesn't explain the mechanism(s). The speedometer in my car is a faithful measure of the car's speed. But moving the speedometer needle is not what controls that speed. Same with information. While it may be a valid measure of "the drive", it's not what controls it.

English

wildiris@wildiris19·10 Mar

(1/n) Your observation raises yet again that nagging question about information relative to living systems and the origin of life. • Is information foundational and the driving force behind living (agential) systems? • Or is something else more fundamental at play, with information only a downstream byproduct of that fundamental process? • The concept of information, in the Shannon sense, is the mathematical sum (or integral) over some bounded thermodynamic state space. • That is, information is a "global" measure and represents only one element of a "black box" description of a system. • But as a global quantity, information tells you nothing about the local conditions and interactions going on at the level of the individual elements making up that thermodynamic space. • ..This is the point I think your observation is picking up on... • Rather, whatever drives agential behavior at the level of the individual elements, is what should be taken as primary; with information then becoming a downstream secondary effect. • Of course, every agential system is also a physical system. It’s not unexpected then, when treated as thermodynamic systems, agential systems will still follow the same laws. • But that doesn’t mean that the interactions among the individual elements of an agential systems, are either driven or explained by recourse to generalized thermodynamic quantities.

Cameron A. Schmidt, PhD@CAS_ReproLab

Trying to think of a scenario where information theory adds something new. Seems to just rescale existing models. E.g. optimal allocation of search effort becomes optimal acquisition of information. Maybe more intuitive framing, but not really different🤔

English

184

Keşfet

@asimovinc @selim__1903 @AlexE_00 @chris_j_paxton @DJiafei @thesuperegonaut @rpg571 @FelixSchlang