Title: Towards a new kind of science and technology | Locklin on science Description: Reading the history of science and technology, one is struck by how important two branches of physics are: electricity and magnetism, and thermodynamics. The former is mostly how we conveniently pipe energy around. The latter is .... how we get the energy in the first place, as well as making most of chemistry possible. Even… Keywords: No keywords Text content: Towards a new kind of science and technology | Locklin on science Skip to content Skip to search - Accesskey = s Locklin on science Towards a new kind of science and technology Posted in physics, Progress by Scott Locklin on December 18, 2024 Reading the history of science and technology, one is struck by how important two branches of physics are: electricity and magnetism, and thermodynamics. The former is mostly how we conveniently pipe energy around. The latter is …. how we get the energy in the first place, as well as making most of chemistry possible. Even humble inventions such as the small electric motor had enormous implications in how things get done. Before the invention of small electric motors, for example, machine shops or printing presses were powered directly by thermodynamics and mechanical connections: usually leather belts driven off a line shaft. Other sorts of mechanical connections were also used: wire rope systems (elevators still use them which is kind of an odd anachronism) or hydraulics. Now we pipe electricity around and electric motors turn the power into motion. Must have seemed like magic at the time; it is pretty cool when you stop to think about it. Burn something in one place, pipe the energy via thin bits of metal into little motors which do useful work right where you need it. Much better than strapping leather belts to the output shaft of a steam engine, with dudes shoveling coal into it in another room. Thermodynamics is how we get power from heat. It is also how we design chemical reactions to make useful substances. I could imagine a modern world without modern chemistry (there would be fewer people without the Haber-Bosch process); even without electricity (certainly without computards: life was more fun without them), but not without thermodynamics and heat engines. Human standards of living are essentially proportional to the amount of heat converted into power which humans can use to do useful work. Without heat engines and the science which drove their creation and perfection, we’re back to Renaissance or Roman times where virtually everything is moved and built with muscle. It is now possible to get electricity directly from the sun and we’ve had wind and water mills for millennia. There are exotic ways of extracting electricity directly from heat: magnetohydrodynamics for example: still thermodynamic ultimately. Most accessible human power comes from thermodynamics.  We live in a thermodynamic age: without it, we go back to subsistence farming and chattel slavery. Where the modern world was born: Soho Engineering works in Birmingham Thermodynamics has a sort of reductionist version which is what is mostly taught in schools: statistical mechanics. It’s “nice” for physicists to think about things in this way, as we can derive most of the ideas of thermodynamics from more simple ideas in probability and statistics, and knowing that matter is made out of atoms. We like to think about it as being more fundamental for this reason, but I’m not sure it’s more general than the kind of thermodynamics developed to optimize heat engines. It is neat that we can derive all this thermodynamic stuff from simpler ideas, and this represents a great intellectual achievement. None the less, something was lost in didactics in thinking only about the statistical physics stuff. I think it’s within the realm of possibility that the statistical physics stuff also makes us blind to things we should be able to see if we approach the problem differently. Physicists probably wouldn’t have invented statistical physics and derived all those cool canonical ensembles if they hadn’t first invented thermodynamics. It’s not something that comes naturally from ideas about probability and atoms, though everyone who worked on thermodynamics had the notion that it was probably something like that. Thermodynamics does come naturally from thinking real hard about making better steam engines (before that, better cannons). That’s how we got there. Practical observation of nature, not smoking the pipe and thinking big thoughts about mathematics. Statistical physics was a cleanup operation; a very successful one, but it came about after the laws were discovered. There are other attempts to embed thermodynamics in some other kind of abstraction. Ruppeiner geometry is a way of embedding it into information geometry: I think mostly because it makes it easier to reason about thermodynamics in other differential geometric systems like General Relativity, though I haven’t made a study of it and I could be wrong. Other extensions of this idea may have more generality. Maybe not though: physics nerdoids like mapping ideas onto other kinds of models, especially geometric models (guilty). That doesn’t mean you get any extra insights from them.  Other examples of formulation of thermodynamics in higher forms: contact geometry,  Hamilton Jacobi theory (and here as well),  ET Jaynes MaxEnt attempts using information theory, Caratheodory’s axiomatic thermodynamics. Are they important? Don’t know: they don’t seem to have influenced much of anything so far, beyond, “that’s pretty nifty.” Thinking about the history of thermodynamics, it was essentially people trying to come to grips with the concept of heat. Heat and its absence is something we observe in nature through fairly humble kinds of observation. Drill a hole and everything gets hot: what means? People had been thinking about heat in various ways for hundreds of years before thermodynamics was formulated in the mid-1800s and finalized in the early 1900s. The kinds of observations of heat and its behavior are fairly humble stuff: looking at the microscopic theory is interesting and you can find odd effects you might not have looked for otherwise, but it was the basic stuff which brought about the biggest insights. Very humble measurements: pressure, temperature, volume, work. Retard monkey measure things, notice things are conserved or equal to combinations of other variables in funny ways.  None of the doofus “looking into the mind of God”  crap we’ve been afflicted with since the 20th century brought public relations to the physics community: just hammer and tongs science by the men who actually created the modern world. I postulate there are higher orders of “thermodynamics” which are discoverable,  yet undiscovered. Everyone knows there is something called non-equilibrium thermodynamics, which almost certainly has numerous undiscovered laws, and a couple discovered ones like the Onsager reciprocal relations. Note here, that the Onsager reciprocal relations are strictly formulated like all the other thermodynamic laws. He didn’t start from the microscopic version of statistical physics; he used normal physical conceptions of continuity and thought about what was going on in a manner mostly devoid of statistical mechanics until the very end of the second paper. You can read his original papers here and here for his reasoning. The impetus for the work was the thermoelectric effect, which had until previously been a classic subject of interest for thermodynamics pioneers like Lord Kelvin. Onsager finished the job; the first and so far only law of non-equilibrium thermodynamics. It’s criminally under-taught in school despite having extreme real world utility: I think because you’d have to be quite familiar with thermodynamics itself, which is also criminally under-taught in physics school. People mostly stopped serious thinking about this stuff with the advent of useful computers. If you have a non-equilibrium system you can figure out most of the stuff you need using computer simulations. The problem with letting computard look for answers in specific cases is you don’t get the higher order Onsager-type thing allowing you to look for effects you haven’t thought of yet. I’m talking about humble phenomena you can see. There are numerous phenomena in matter which show order which are not easily described by higher order thinking or some kind of Onsageresque thermodynamic relationship. They may be describable by some kind of simulator, or they may not. When things behave in an orderly fashion they should be describable by mathematics. Cloud streets: an example of self-organizing systems you can occasionally see There was this guy Herman Haken, who to my surprise only died this year (at 97) who wrote books on something he called Synergetics. It was an exciting series of books to read as an undergraduate in the early 90s, as it seemed to tie together a bunch of stuff which bothered me about physics, and promise a way forward (it also redpilled me that information and entropy were the same thing). Haken’s ideas came from studying phase transitions, and particularly the self-organization of laser dynamics. He and his colleagues were interested in things which self-organized: stuff like turbulence, patterns in fluid mechanics and plasmas, brains, Fokker–Planck equations. He assumed (as I do) there must be some unifying mathematics behind this sort of weird stuff that looks familiar and classically mathematical. I don’t think anything useful came of these  efforts; in part because it was long on qualitative stuff, fiddling with differential equations and interdisciplinary work, and short on trying to make something in the world of matter function properly, unlike thermodynamics and steam engines. You can’t blame them for trying though; Ilya Prigogine won a Nobel in chemistry around this time for his work on self-organizing systems, Fractals were a big thing, and results from Chaos Theory were pouring in, giving mathematical order to another kind of complex system which appeared to have self-organizing properties. Simple computer simulations of flocking birds and ant-swarms were newly possible and showed enticing order from simple rules. Solitons were something people studied back then: the kind of self-organizing system you could make in a little tub of water. Sure it’s described by the Korteweg De Vries equation, problem solved, right? Nah, not really. The phenomenon is more universal than that, and it’s basically only understood via computard simulation. For a while people thought maybe solitons could be a hidden key to what’s really going on in quantum mechanics. It seems a stretch, but it’s as good a guess as any I know of. When I started my physics career in the early 90s, this stuff looked like it was the future. Unfortunately it remains something for the future: as far as I can tell people have mostly ceased to think about these things, and ultimately never really did give them a good think. Now a days, physicists seem to prefer fiddling with neural nets. Of course it’s possible that all this spontaneous order is just a coinkidink and there is no over-arching principle to such things. I don’t think so, though. I think it mostly remains unstudied, excepting perhaps outside of classified work in naval and aerospace laboratories, where it isn’t likely to help anybody. Though there are some interesting exceptions to this where new results come out of military studies of turbulence. It is perhaps over-ambitious to think something like turbulence is connected to non-equilibrium chemical reactions, chaos or solitons, but I think Haken and his friends were onto something, and there is stuff there which can unify many seemingly unrelated unusual behaviors of matter. These sorts of things may be used to solve practical problems, and for progress in this, we probably should use one or more of them as test cases, if we want to figure stuff out, the same way we figured out thermodynamics from thinking about steam engines. I think Haken’s project was a failure mostly because it was a late career vanity project, done in the usual dreary “make conferences, publish proceedings” way,  but I think his assumptions of an underlying order are probably right. This kind of thing probably isn’t going to make progress by throwing money at it and having conferences (no other problem in human history has been solved in this way): it’s probably going to be attempting to build something practical that involves this sort of self-organizing system. Or at least a deep Onsager-like study of some particular one. Figuring out what that might be almost certainly won’t happen in the “physics community,” for the same reasons nothing else happens in the “physics community,” but it should happen. Humanity is leaving money on the table otherwise. Share this:FacebookXRedditLike Loading... Related 37 comments « Fall 2024 books 37 Responses Subscribe to comments with RSS. T Green said, on December 18, 2024 at 5:26 pm A physicist friend recommended ‘The Tragicomical History of Thermodynamics, 1822 – 1854’, (1980) by Clifford Truesdell for outlining the difficulties that scientists and engineers had in understanding and formulating thermodynamics in the early part of the 19th century. Have not read it yet, but Truesdell himself seemed a very interesting character… Reply Scott Locklin said, on December 18, 2024 at 6:00 pm That looks banger, thanks for suggesting. Reply chiral3 said, on December 18, 2024 at 11:50 pm That does look awesome. If I buy it I may need to put it next to this chestnut, which resides in my “reserved” collection. Reply Scott Locklin said, on December 20, 2024 at 1:15 pm I dunno though, two chapters in and the dude is still clearing his throat with equation baloney, making fun of his betters (I mean, Laplace, Poisson, Fourier aren’t good enough for this reddit knob) for not having clean mathematics like him. Also half or three quarters of each page is a “footnote” because he’s too convoluted to think in complete paragraphs. The author embodies this meme and is definitely part of the problem: Maybe I will read potato book instead. Reply T Green said, on December 20, 2024 at 2:13 pm Apologies, should not suggest books I have never read! In a bio of the author it mentions his obsession with mathematical rigor and disdain for experiments, something that would put me off. The potato book looks cool… Reply chiral3 said, on December 20, 2024 at 3:25 pm It’s a great name for a book written in a time when people were less sarcastic. Like Poor Man’s Explanation of Kalman Filtering: Or How I Stopped Worrying & Learned to Love Matrix Inversion. Scott makes a good observation that thermo was always the odd duck of physics. And if mathematicians get all bent out of shape about path integrals and delta functions being abused the whole dU=dQ-dW stuff should make them downright apoplectic. Reply Scott Locklin said, on December 20, 2024 at 7:22 pm It was worth a look, but baleeted and won’t look back: you did say you hadn’t read. The guy’s tone was supercilious and annoying even from the first few lines. If he were still alive I’d stuff him in a locker. I knew a guy like him worked on theoretical stat mech; same character flaw. One of the former marines in grad school described him as “a waste of skin.” Reply chiral3 said, on December 18, 2024 at 5:52 pm Haber-Bosch is super-interesting for historical reasons. The stories of the eponomous Haber and Bosch are also interesting (Nazi Nobel was interested in melting people, not making food, and the other dude unalived himself out of guilt). The population of the earth would be far smaller without. Early choas self-organizing days were interesting too. I spent my fair share of time nerding out. Doyne Farmer, Per Bak, et al. It never got legs, though. Like that liminal space between walking and running we just don’t understand the relation of micro-structure to the whole that well. Reply Scott Locklin said, on December 18, 2024 at 6:22 pm My thesis projects (all three of them, lol) all had strong connections to the Chaos stuff. I hadn’t quite got how it worked when I first looked at Synergetics. Could probably still give a decent lecture on it now, and its relation to quantum things. All that said, I think most of the excitement was due to early computard results. Anyone could run an “experiment,” like with fractals. Look at my Poincare map. Look at my Husimi distribution. Then, like with the Synergetics foray, they all found other things to think about. One of my former colleagues in all this stuff has moved on to studying rat-brains. Reply chiral3 said, on December 18, 2024 at 7:06 pm Me too. All the fluid shit I did was chaos curious. More stupid tidbits extrapolating the arc of Haber-Bosch Entire industries owe their existence (and salt) to the industrialization of ammonium salt production.   There’s another way to make ammonium nitrate on a large scale involving nitric acid, but it never got industrial legs.  The inventor was another German who, ironically still, was associated with a pacifist movement.  He died one year before the fall of Weimar. It is worth noting that ammonium nitrate lags another fertilizer in popularity, primarily due to the economics of transportation: urea has more nitrogen per unit of volume than any other compound and was first synthesized by the German chemist Frederick Wöhler in 1928.  Another German.  They must have had the same inspirational teachers in their respective gymnasiums.  Urea nitrate can be used as an explosive. The six individuals that perpetrated the 1993 World Trade Center bombings had handlers that knew this fact.  They were reportedly pissed off about US-Israeli sympathies and figured out how to make bang-bang without blowing their dicks’ off.  Despite avoiding blowing themselves up their efforts were not without comedy.  Mohammed Salameh, the esteemed driver of the truck housing the explosive devices, failed his New Jersey state driver’s test no less than four times leading up to the attack, arguably protracting the timeline by not being able to rent a car with a valid NJ state driver’s license.  So mystified was he by the complexity of the modern automobile that he almost killed his passenger and better-known co-conspirator, Ramzi Yousef, just thirty days prior to the attack in a car accident.  Then, just two weeks before the attack, he got into another fender bender totaling yet another rental car.  Logically, when assessing the team’s talents, one would assume Salameh was deemed the best driver of the lot.  Without chariot the team almost had to roll the fertilizer drums by hand to the World Trade Center.  In Dinkins’ 1993 New York this might not have raised suspicions in places like Times Square, but the financial district would have taken notice.  Khalid Sheik Mohammed, the money and brains behind the operation, perhaps chalked this misfortune up to the price one pays for recruiting the dumb and impressionable.  When General Mao wrote of one training ten training a hundred training a thousand… he never said the resistance had to be smart.  Only the first guy has to be smart.   The rest need to be able-bodied.  Smarts are a sufficient but not necessary condition: whether you’re killed by a genius or a dullard doesn’t matter, you’ve lost at that point and all the IQ points in the world aren’t going to bring you back.  I often think about the NJ Motor Vehicle Commission employee that finally capitulated on Salameh’s fourth driving attempt.  Salameh was finally caught on his second attempt to get his deposit back from the rental company. Inshallah. Reply Scott Locklin said, on December 18, 2024 at 10:11 pm I remember some of those WTC-1 facts. German chemistry used to be impressive. Now they have to do stuff like buy Monsanto. Any idea why they’re trying to make nitrogen use on the farms illegal? I can’t parse the reasoning of climate hysterics, mostly because there typically isn’t any, but it is making a big stink with farmers. Seems even more absurd than trying to run steel mills using solar panels or whatever other nonsense these clowns are touting, but perhaps there is some germ of a real reason there. Reply toastedposts said, on December 18, 2024 at 10:17 pm In Europe, the big push seems to be driving farmers into bankruptcy so various sorts of sharks can buy up their land cheaply for development. Reply Scott Locklin said, on December 18, 2024 at 10:51 pm That’s certainly the case, but I assume they have a better excuse than “it’s global warming: trust us.” Reply chiral3 said, on December 18, 2024 at 11:30 pm Dunno. Oceans, rivers, aquifers? Effluent run off and hypoxic water sources? Endotoxins from increased flora and tainted water sources? That’s my best guess. Reply Scott Locklin said, on December 19, 2024 at 8:57 pm They really don’t explain anything, but it’s some kind of pan EU objective at least. Reply toastedposts said, on December 18, 2024 at 10:07 pm My favorite bit of thermodynamics heresy is the Fermi-Pasta-Ulam numerical experiment. Every so often I like to bring up that we assume that thermalization is a thing that always terminates in equipartition of energy after enough time. (It ends up *not* being equipartition exactly for nonlinear systems, but that’s a side issue). In certain weakly nonlinear systems, thermalization as we assume it to happen … doesn’t. I should probably spend more time on this at some point. I was using toy computer models to look at the sort of noise spectra given off by systems in contact with a “thermal” random input, and noticed physics enforced cutoffs and resonance peaks in the output noise of the system. —– Undergrad engineering thermodynamics was taught along pre-theoretical historical lines. It also probably didn’t stick in my memory for exactly that reason. There are only so many arbitrary seeming came-out-of-nowhere two-parameter functions that can stick in my brain, if I don’t have a reason for why they are what they are to hang it on. Now I know what a Legendre transform is, and what the hell they were doing juggling various kinds of energy-like-functions, so maybe I should revisit it. OTOH, we could calculate the efficiency of a Brayton cycle without reference to planck’s constant and arbitrary zero-point nonsense. Reply Scott Locklin said, on December 18, 2024 at 10:50 pm I think the thermodynamics stuff is kind of an embarrassment for physics types because nothing else looks like that. It’s also almost certainly is the case that something like enthalpy would make more sense if they showed you how people figured it out rather than just blindly doing calculus of variations on it or whatever it is they say. Reply Cameron B said, on December 19, 2024 at 7:08 am “‘looking into the mind of God’ crap” as I glance over at my bookshelf where sits “The Mind of God” by Paul Davies. Thanks for the post, much enjoyable. Reply Brutus said, on December 19, 2024 at 3:58 pm Caratheodory’s axiomatic thermodynamics did lead to an interesting theory. The late Pharis Williams explored it in different limits and argued that you get back much of relativity and quantum mechanics when you do so. I can’t help thinking he introduced hidden assumptions, but I’m still inclined to think it’s a vastly better approach to unification than strings or super symmetry. Reply Brutus said, on December 19, 2024 at 4:07 pm Here’s a nutshell overview of Williams’ work. Shannon, James O., Warren Russell Maines, David Mathes, and Paul Murad. Memorial and Thoughts of a Man with Great Ideas-Pharis Williams. https://www.osti.gov/servlets/purl/1248828 Reply karocann said, on December 19, 2024 at 10:05 pm I did a stint as a researcher in the navy once. They were interested in magnetic hysteresis, particularly that induced by stress (you can imagine why). I subsequently discovered that macroscopic hysteresis was largely not understood. Some vaguely robust phenomenological models existed but they broke down quite quickly. No statistical mechanical theory accounted for the observations. Computards were of little use due to the colossal length scales and multi-scale nature of hysteresis (it is not so easy to make inferences about hundreds of meters from nanometer scale models, as it so happens).The team wound up doing a range of “boring” experiments looking at how stress influenced hysteresis studying pipes and whatnot, which it turned out, in spite of being cheap and simple, no one had done. To interpret the results and try to futz them into our electromagnetics codes, we dug up a crusty old WASP who had been toiling in obscurity in a naval academy, outlasting his rivals to become one of the world’s few remaining experts on the subject (a peculiar character, to say the least). Much of what we did was phenomenological thermodynamics, to adopt the language of physicist’s. It astonished me how little was understood (indeed, it turned out that NASA had figured out had to make precise flux gate magnetometers decades ago, but that this knowledge had been lost, and a great struggle ensued to rediscover it, which failed). I must confess that ultimately not much progress was made, but it buttresses your suggestion that “boring” real physics is not only important, but criminally understudied. The archaic WASP I mentioned had fewer citations in total than most noodle theorists have on their most cited papers, in spite of the vastly greater relevance of his work. I should also note that we briefly attempted some ab initio calculations to parameterize a microscopic approach to the problem, only to discover that DFT is garbage. I already knew this, but like a battered wife, I always come back to the quantum world for another beating. It vexes, perplexes, and flummoxes me that quantum calculations are still largely useless. To your points about non-equilibrium physics, I concur that Onsager’s relations are criminally understudied, although chemists and engineers are more familiar with them than fizzy-cysts. However, I would say that while there is really one general way to be in equilibrium, there are many ways to be out of equilibrium, so what one finds in the applied physics literature is theories which apply to certain non-equilibrium states. One can obtain a nice thermodynamics for, say, a multi-valley semiconductor, but it is peculiar to such a situation. Different specific thermodynamics are needed for things like the Boussinesq force or electron transfer in enzymes. Regarding esoteric geometry and its “application” to thermodynamics, I can never decide if I am a man of limited intelligence or if it is an academic way to make something otherwise mysterious but intuitive more complicated than it needs to be. I spent a few years in the computational biophysics world, and in their mounting desperation for any technique to calculate anything of relevance, many turned to goofy techniques developed by computer twinks which sometimes relied upon differential geometry (or, worse, muh sheen “learning”). These often reduced to a simpler, stupider method recast in more obtuse and confusing mathematical language, and simpler, stupider methods often obtained similar performance with less mathematical flair. I have since left that world, and perhaps the computer twinks have since dominated it (the MEGA-twinks at DeepMind seem to think so), but it turned me into a Luddite with regards to abstract formulations. Why think about distance between, say, protein configurations using abstract high dimensional geometry and metric tensors when you could think about path integrals, which themselves can be interpreted with stochastic processes? At least to me, that seems more intuitive. Information and thermodynamics (and statistical mechanics) is a subject that bewilders me every time I revisit it. A physicist friend of mine with a penchant for this thing coaxed me back into pondering about the Szilard engine and Landauer’s principle recently, and I always leave the subject with the same feeling that most of the conclusions reached are acts of semantic mental masturbation. Often information is redefined in a way that bears no relation to Shannon’s original theory, and what is more, I never get a clear, demonstrable conclusion of the significance of this alleged connection. Information is any signal. If I stand in a pool of water, and a slight breeze causes little waves to lap at my ankles, that is information. Any signal is physical. So what? Grandiose conclusions about the limits of information processing, vain attempts at general theories for deriving models of intelligence or biophysics are constructed, and no results that make much sense to me emerge. Then again, that could just be proof that I’m better off squeezing steel pipes to see if I can make the magneto-numbers go bigger rather than thinking Big Thoughts. (Good thing there is no subject which combines both information theory and differential geometry. Perish the thought! It would be the death of me, or someone else depending on my mood. That reminds me, I need to bone up on some statistics. Maybe I’ll go back to the early 20th century greats. How about I start with Sir Ronald Aylmer Fisher?) Reply Scott Locklin said, on December 20, 2024 at 3:11 am Very stronk poast, thank you. Any sort of magnetic anything is mysterious. I remember working at the data storage systems center at CMU for my second real physics job (summer internship); first one of consequence for cash money. We were making magneto optic disc substrates. You’re old enough to remember those dumb magneto optic drives that filthy hippy demanded be put in the NeXT Cube, but which never quite caught on (because they were retarded). Being an eager undergraduate I figured some theorist would give us guidance on how to make better substrates. Nah, they had real good ones they could make using molecular beam epitaxy (by accident), but sputtering is the only way to mass production, so we effectively baked 1001 cookies with slightly different partial pressures of cobalt and platinum and substrate spin rates in the sputtering chamber, then stuck them in torque magnetometers, SQUIDS and electron microscopes to see what happened. So, it turns out Juggalos are right: magnets are magic. Semi-reproducible magic. Like making a nice choco cookie: can’t be derived from first principles beyond “use some choco.” >Good thing there is no subject which combines both information theory and differential geometry. Perish the thought! It would be the death of me, or someone else depending on my mood I assume you’re trolling me for thinking about this, and it taking me two years of learning differential geometry formalism I should have learned in Carlo’s GR Class to figure out all it is, is fiddling with profile likelihood gradients, which is a completely retarded thing you could do manually without any fancy math (and you should do something else anyway, like use a Venn conformal predictor). Information theory is legit, but looking at Shannon’s collected papers, even in his day people were woo-ing it into a whole pile of bullshit which was both irrelevant and bullshit. He complained about it. Then took up juggling. All the best recent results in machine learning (not LLMs) have a strong info theory component to them. Information theory was invented to come up with stuff like Golay codes, and ended up being useful for all kinds of stuff involving uncertainty. Info geometry though, is almost entirely wankery. In the past of course, thinking of using binary math, noise and reasoning about error correcting codes was kind of the main project of information theory. I sorta like the Landauer thing but had to google what a Szilard engine is. Some people say Landauer limit isn’t a hard limit because “muh adiabatic whatever” -which is (IMO) basically just saying if you build a perfect reversible computer it doesn’t apply. This is true to some small scale; as long as you keep around all the intermediate bits in your calculation. In other words, practically speaking, this is not possible. Reply Lev said, on December 20, 2024 at 10:53 pm I sometimes fantasize about finding a way to get piezomagnetics to enable pure fusion nukes. Would be a lot of fun. While thermodynamics usually baffles me, following your mention of Bill Dreiss got me to read his ‘einstein’ spiel, and I really like the volume based heuristic. It makes thermodynamics somewhat intuitive, and has allowed me to explain to normies why certain harebrained ideas can’t work. Thinking about the volume(in phase space or whatever) where energy has spread out, clues you into how much work you’d need to get at it. I assume this is also relevant to information and Muh sheen learning, but I can’t be bothered to find out how exactly. Reply maggettethekraut said, on December 21, 2024 at 12:42 pm Even though I only understood half of it, fascinating read. Thx Reply maggettethekraut said, on December 21, 2024 at 12:42 pm Even though I only understood half of it, fascinating read. Thx Reply WMBriggs said, on December 19, 2024 at 11:04 pm There has to be a good joke in swapping Julian for Edward in MaxEnt, like the bicameral brain maximizing entropy and becoming the unicameral brain via thermodynamical processes, but it evades me. Reply Scott Locklin said, on December 20, 2024 at 3:18 am How do I bullycide BAP into reading Julian Jaynes? It is the ultimate semi-convincing schizo theory: somewhere between Drews and Spengler. Also thank for typo catch. Reply WMBriggs said, on December 20, 2024 at 11:15 am Quote one or two of the loftier passages from his early history. Jaynes’ gorgeous language alone is enough to convince anybody to keep reading. I’m so good at finding typos because my enemies insert many into each one of my posts. Reply OrangeFrog said, on December 20, 2024 at 7:35 am Another great post, Scott. It’s been a while since I took my physics degree but, I always felt there was a disconnect between a huge amount of the theory we learnt (especially thermodynamics) and what it meant from a practical perspective. I think I had one class on experimental physics, and it was a short one; but for me it was really enjoyable. I recall reading about R.A. Millkan’s experiment to measure the charge on an electron, and we got to repeat it. It was, to me, incredible that such a fundamental, microscopic thing could be measured in a ‘simple’ macroscopic way. I think Dover publications actually have Millikan’s book on the subject for cheaps. Even at school, we often used little wagons on inclined planes to study kinematics. Simple, yes. But I found it very enjoyable, and it was often satisfying to see just how closely (or not) reality conformed to theory. But this practical emphasis seemed to diminish as my education progressed; I mean, at one point I could do all the fancy mathematics with Maxwell’s equation… but I couldn’t build a crude motor or voltmeter. It always bugged me. I think that many students struggle to grasp a lot of physics because it is simply not presented to them in the manner that it was originally discovered. They think: “How would anyone know this? Did some guy sit in front of a whiteboard and just think this?” – but no, often discoveries were accidental and purely empirical. I think this is doubly so for thermodynamics (a subject which I enjoyed very much, in it’s more basic form). Guess another problem is this: in a typical physics undergraduate degree, we’re trying to cram what? 400 years of knowledge into 3 years? Of course, somebody like Faraday understood the practical phenomena of electricity better than us… it was his life. Probably wasn’t too hot at the calculus though, but so what? The great Scottish engineer Thomas Telford (highly recommend L.T..C Rolt’s biography of him and his works) would often undertake laborious and expensive experiments on things like different forms of concrete for harbour walls, or suspension chains for his bridges. There was no other way, and it provided him with the confidence he needed to build things that were unparalleled at the time. This is not to scorn at theoretical work, but one of the reasons that I never wanted an academic career in this field was because it seemed to be getting more and more detached from practical reality. And, on closer examination, many modern theories have more holes in them than Swiss cheese. Actually, as for characters that liked to experiment, Gerald Bull was a fascinating guy, for sure. Reply Scott Locklin said, on December 20, 2024 at 8:43 pm I had a couple of lab courses as an undergrad: at least 4; even had a machine shop course, and one (two semesters) as a grad student. My biggest problem with undergraduate was the junior/senior years were a wasted opportunity. They teach you half assed shit instead of what you actually need to know. Then you repeat it all in grad school. Also you take a bunch of mega gay coursework no adult should be forced to take as an undergrad, to keep the liberal arts side of the school in business. Only one I got anything out of was the Art History class, which, of course, was taught by a physicist. Like one, I picked “Nietzsche Marx and Freud” because I had already read all of Nietzsche and half of Freud. It made me realize the professor didn’t understand Nietzsche, she liked Freud because of her cummies, and Marx was her religion. Basically, the professor and most of the students were profoundly retarded, and you’d get more out of speaking with average football fans in a sports bar. I did often find myself wondering how they came up with stuff when we didn’t repeat the experiment someplace, but usually figured it out reading original papers. Original papers are almost always shockingly straight forward. Schroedinger’s equation for example. Reply karocann said, on December 20, 2024 at 8:48 pm Schrodinger’s Stackoverflow code-copying tier reverse engineering of the Eikonal equation while plowing his postdoc’s wife in the alps remains one of academia’s greatest coups.Speaking of old stuff that is criminally understudied, I learned about our favorite quantum pervert’s approach from an ancient and obscure tome on geometrical optics called Kline and Kay. There’s a subject nobody cares about anymore which is nevertheless of great importance. Reply Scott Locklin said, on December 20, 2024 at 9:05 pm Links to Kline and Kay or it never happened. The only hit I got was a 40 page white paper dating from 1962. Reply karocann said, on December 20, 2024 at 9:10 pm They don’t state explicitly that it was Schrodinger’s program to map the Eikonal equation to quantum, but it seemed strikingly obvious to me that he did after reading their treatment of the relation between the two subjects. I believe I heard another account but maybe it’s a one man Mandela effect. Another interesting connection I found in this book was between Sommerfeld’s program of deriving geometrical optics, which must have inspired Bohm’s later work on his mechanics. Reply Scott Locklin said, on December 21, 2024 at 1:19 pm Thank, I’ll have a look. Sommerfeld’s lecture books are a series of treasures, somehow mostly forgotten. He was professor to 7 nobel prize winners and pretty much everyone else he trained did something of note. You’d think people would continue to take note of his lectures. Reply OrangeFrog said, on December 20, 2024 at 9:14 pm Heh. Yeah, you can say that again about original papers. I’ve been boning up on DBA skills because “profession” and “job” reasons; so got to looking at the early work on relational theory applied to data banks by Codd. Then, even better, the original (maybe 1974?) paper on SEQUEL (later to become SQL, SEQUEL apparently being a trademark of Hawker Siddeley.). I was surprised at how simple it was. Really was a pleasure to read. Opened up a new (albeit old) world to me: old IBM DBMSs and the like. Never cared for quantum. At uni, I just wanted to skip right to fluid mechanics. Then again, my uni was a bit lame, we didn’t even learn Lagrangian mechanics in our entire three years there! Fancy that! Reply toastedposts said, on December 20, 2024 at 7:47 pm So if there’s a theme to these posts on this blog it’s that interesting discoveries come from immersing yourself in a world of interesting phenomena, and then noticing stuff. Not from staring hard into Platonic space to perceive the music of the spheres. Noticing that your film got fogged when you left it in the desk drawer over the radioactive rock —> lit-fuse —> nuclear revolution. So we need to find interesting phenomena somewhere. I am unfortunately locked away in what I call “abstracto-land” at work. (Trapped in the computer with no escape. They get nervous when I come in with a soldering iron.) Any (additional, you post on this pretty often) interesting phenomena that come to mind that 100 years of theory hasn’t already laid claim to? I imagine everything in biology is like this, because biology is fantastically complicated and resists all attempts to turn it into smooth functions. Reply Scott Locklin said, on December 20, 2024 at 8:26 pm There’s a whole category “Open Problems” on my blerg where you can find lists of these. Water’s heat capacity. Turbulence (aka incense smoke spirals). Barred spiral galaxies. Spiral galaxies. Dork matter shit (I am almost certain Doc Ludwig is correct: it’s gravitomagnetic). Globular clusters. Anisotropies in astronomy. https://scottlocklin.wordpress.com/category/open-problems/ Reply Leave a comment Cancel reply Δ About me: Stuff I like The Futurist Manifesto About Scott Locklin  Past blogsPast blogs Select Category astronomy big machines Book reviews brainz chaos Clojure cold fusion Corliss corona-chan Design econo-blasphemy econophysics energy finance finance journalism financial patents five minute university fraud fun Gambling systems Genetic data health history history information theory investments J Kerf Locklin notebook Lush machine learning manhood metalshop microstructure models nanotech non-standard computer architectures Open problems patent law patents patrician-entertainment philosophy physics physics anomalies privacy Progress Q reviews SBIR semantic web statistical tools stats jackass of the month systematic trading tools Uncategorized War nerding Wolfram Alpha Email Subscription Enter your email address to subscribe to this blog and receive notifications of new posts by email. Email Address: Sign me up! Join 2,401 other subscribers RSS link thingeeRSS - PostsRSS - Comments Blog at WordPress.com. Comment Reblog Subscribe Subscribed Locklin on science Join 1,554 other subscribers Sign me up Already have a WordPress.com account? Log in now. Locklin on science Customize Subscribe Subscribed Sign up Log in Copy shortlink Report this content View post in Reader Manage subscriptions Collapse this bar   Loading Comments...   Write a Comment... Email (Required) Name (Required) Website %d