Leon P Smith

I ordered a LDO Micron+ kit from West3D today, which has a 10% off all LDO kits this weekend. #3dprinting

@NefariousAryq@hoosier.social I ordered my first roll of TPU recently, but I haven't yet tried to print with it. It's been a while, but I hope to be pushing plastic soon, probably PETG. I'm trying to get my 3d printers properly ventilated out the window, and need a nice window insert with a dryer vent on it.

@ai6yr@m.ai6yr.org For the large one I was inspired by 3D handyman's efforts on youtube. I love his introduction to laminar airflow systems. I remember hearing about them a bit in the context of various kinds of cleanrooms, but didn't quite realize how (conceptually) simple they really are. I may need to set a laminar air filtration system up at some point. For now I'm more concerned about getting decent quality extractors for my 3d printers and my soldering station. https://www.youtube.com/watch?v=zKeNf6fLSLA

@evacide @tomhead surely there has to be a fairly strong legal argument here, at least one of which revolves around one's right of publicity.

@josh @ChuckMcManis Two problems: that kind of documentation often does not exist, and if it does, that kind of documentation often gets ignored. "Too theoretical, not practical, etc. etc., but more often if they were self aware and being honest, it's more like 'I don't understand'" I mean, I agree, this kind of documentation can be invaluable, if employers are willing to let (and encourage) engineers to write and read it. Concurrency poses a real challenge, as most programmers don't even begin to comprehend the issues in play. I know I was taking graduate classes before that light bulb really turned on.

@futurebird@sauropods.win @johncarlosbaez@mathstodon.xyz Anyway, my point is that "decoding alien hieroglyphs" is actually a pretty good way to start to get an intuition for what an automorphism is. For example, if you were given all the alien hieroglyphs that correspond to their integers, and given access only to their addition table, then you'd be able to figure out what 0 is, as any identity must be unique, if it exists. Similarly you'll be able to figure out which two hieroglyphs correspond to ±1, in the sense that you could call one of them "1 up" and "1 down", and from there figure out what "2 up" and "3 down" are, but you'll never be able to decide if "up" correponds to positive and "down" corresponds to negative, or vice-versa. This intuition is captured by the fact that the group of integers under addition has exactly one non-trivial automorphism: we can negate everything everywhere and things will still work out. (And in fact, this is the only such change that is guaranteed to work perfectly in all cases.) Of course, if you then gain access to the alien's multiplication table, you can multiply "1 up" by "1 up", and that answer will correspond to the positive direction. Thus we can fully decode the alien's integers, which corresponds to the fact that the ring of integers exhibits only the identity automorphism: when multiplication is involved, we can't just flip everything's sign and expect things to work out. This intuition is a bit hard to operationalize, though, as the addition tables are infinitely large. In reality, if the alien heiroglyphs are truly capable of expressing arbitrarily large members of an infinite set, such as the rational numbers, the notation must involve some regularity. That regularity can provide insight into the alien's interpretation of their rationals in ways that don't correspond to what could be learned from their operation tables alone. The automorphisms of the group of rationals under addition correspond to multiplying by a non-zero rational number, capturing the intuition that you'll never be able to definitively decode the scale of the alien's unit of measurement from the addition table alone. But if you see something like 1/10000, you can guess it's probably not the unit, versus something much simpler like 1/1. However, the field of rationals exhibit only the trivial automorphism, meaning that you could fully decode alien rationals from their addition and multiplication table. Switching to a finite system avoids these complications, allows the use of a simple substitution cipher which prevents students from making inferences based on notation, and also is capable of providing much more interesting examples of automorphism groups than your more widely-appreicated number systems can.

@WiteWulf@cyberplace.social @SecureOwl@infosec.exchange if you aren't using a soldering iron with some firmware on it, you really are missing out on the benefit of modern temperature control loops. Generic T12 stations seem to be the best value for the buck, the precise temperature control makes a lot of otherwise tricky soldering job a piece of cake, such as surface-mount work or using lead-free solder. I don't think Hakko even makes a T12 station anymore, though they do still make T12 tips. Generic C245 and C210 stations seem to be where it's at, assuming you are willing to spend a bit more, and are willing to spend more on tips. Still way less than a genuine JBC station, though.

@ai6yr@m.ai6yr.org I have the things to build a small one, and soon I will have the things to build a much larger one.

@VeroniqueB99@mastodon.social https://www.youtube.com/watch?v=RGqA1lNXYhg

@mttaggart Unless you are unlocking your password manager every time you want to use a password, or you have a dongle-assisted password manager, at least in theory there's not much difference between storing passwords in cleartext in memory or storing encrypted passwords with the unlock key in memory. Admittedly keeping passwords encrypted with the key in memory does make confirmation much harder, and exploitation a teensy bit harder, but an highly skilled attacker with access to your password manager's working memory shouldn't find much difference. Virtual memory does throw a monkey wrench into this analysis, though. Virtual memory is one reason I am very interested in dongle-assisted password managers.

@sterophonick About two years ago the electronics in my Kinesis Advantage2 keyboard failed, so I quickly built a replacement using a Pico and QMK. At the last minute, I used a Pico W. This weekend I bodged in a battery charger, lithium ion cell, and piezoelectric buzzer. QMK doesn't support bluetooth, and ZMK doesn't support the Pico W. Fortunately the keyboard is still very useable in its current state, but none of my work will pay off until I get the software wrangled.

@stefanie @gregeganSF @ZachWeinersmith I had an elementary teacher who insisted that given the sum and difference of two numbers, the only way to find the original is to guess and check. I immediately knew how to solve that kind of problem, even though I wasn't familiar with linear equations at the time. I did have a good geometric intuition regarding the average of two numbers, which is how I solved it.

@futurebird@sauropods.win @johncarlosbaez@mathstodon.xyz addition table, multiplication table, it doesn't matter, its an abstract operation. But yeah, I do call it "addition", not multiplication, at least when introducing this stuff. I think I have a reasonably simple angle for introducing the symmetry group of the square, and that's (imperfectly) represented in the repo as it currently exists. You should print out the calculator front-to-back and play with it for a bit. I have somewhat developed ideas about how to move from the intuitive approach of my mechanical number line for D_4 to implementing the arithmetic of D_4 using pencil-and-paper calculations. Namely, I think the semidirect product, the 2x2 integer matrix approach, and the permutation-based (i.e. subgroup of S_4) approach are particularly notable. I don't know where I'd place the lesson on automorphisms, as honestly it need not depend on anything other than the intuitive approach. On the other hand, I'd probably want to prioritize at least one or two of the pencil-and-paper approaches to performing addition in D_4.

@futurebird@sauropods.win Actually I have a (very undeveloped) concept of a lesson with respect to the symmetry group of the square. Basically, after the class has been introduced at least to the intuitive approach to the symmetry group of the square, you give them a problem where they have to "solve" a substitution cipher from {a,b,c,d,e,f,g,h} or whatever to the symmetry group of the square, given the multiplication table of that substitution cipher. The lesson here is that this problem doesn't have a single unambiguous answer: rather you can solve the substitution cipher for a few elements like the identity element and the "rotate by 180 degrees" element, but you can only classify the rest of the substitution cipher up to the symmetry group of the symmetry group of the square, more technically known as the automorphisms of D_4. I was thinking maybe there's an angle to develop as like an alien linguist as part of a Star Trek science team, and perhaps even make it a trick question by making it seem like they are expected to find the one "true" solution. It turns out that the automorphisms of D_4 is isomorphic to D_4, which is definitely a very yo dawg moment, but it turns out this is very much accidental. Groups G that are isomorphic to their own automorphism group include all complete groups, but this is one of a handful of sporadic exceptions of a group that is not complete but also isomorphic to its automorphism group. This includes D_4, D_6, D_∞, and may include a few more unknown examples. It turns out that all the symmetric groups (i.e. groups of permutations of n elements) are complete except for n=2 and n=6. The n=6 exception actually pretty interesting, and @johncarlosbaez@mathstodon.xyz likes to talk about it. https://github.com/constructive-symmetry/constructive-symmetry/tree/master/D002_Book_of_Algebra https://math.ucr.edu/home/baez/six.html

@tila@girldick.gay Non-wellfounded set theory is a thing, where sets are allowed to contain themselves. It's relatively consistent with normal set theory. The deeper problem embedded in Russell's Paradox is assuming that the extension of a logical statement is a set. Now, there is a certain sense in which non-wellfounded set theory and more typical set theory can make the same statements, so it isn't intrinsically interesting to those interested in the foundations of math. On the other hand, non-wellfounded set theory did inspire some of my weirder work: https://hackage.haskell.org/package/control-monad-queue-0.2.0.1

@phonner@mathstodon.xyz @xenaproject@mathstodon.xyz Incidentally, I've long been deeply interested in #mathed. I wrote down a rough draft of a overview of most of my philosophy of math education over the pandemic in my constructive symmetry repository. Anyway, I remember thinking about proof checkers for the traditional two-column geometric proof, but never got very far with it. (Geometry was well outside my wheelhouse at the time) I'd love to hear more about your thoughts on two-column proofs, Patrick. https://github.com/constructive-symmetry/constructive-symmetry

@s0 I understood the CW and just smiled.

@funkylab @dlharmon @gsuberland @dysfun Yeah I worked on a bit of sketchy telephone system about 20 years ago, and ended up using some latex gloves just to avoid that particular unpleasantry, even if it wasn't immediately hazardous.

@mattblaze@federate.social @DaveMWilburn@infosec.exchange 40% is likely very low compared to what Trump will be demanding.

@futurebird@sauropods.win Yeah even the lore most mathematicians think is somewhat reliable... we really don't do history as mathematicians, at least not as a larger community. Some of our lore I've come to see as harmful, especially around Calculus. The actual history of Calculus is a lot more interesting than Newton and Liebnitz, but should also incorporoate Isaac Barrow, Archimedes, Madhava of Sangamagrama, and more

@revk@toot.me.uk Is that your first 3D resin printer? I have an Uniformation that I need to actually set up and use. That's been a much slower and time-consuming process than I hoped, in part because I really want to get a high quality ventilation system set up first. I mostly intend on printing things that I intend on handling a lot for longer periods of time, like palm rests and keycaps. I figure the improved texture will be well worth the time and effort, among some other possible advantages.