2023-10-15 00:00:06 <MrMobius    > hehe
2023-10-15 00:10:41 <KipIngram   > Oh, cool.  I was talking about making a cluster of raspberry pi's.  I just remembered all the possibilities - I could set those guys up wo that they booted over network, and then every time I turned it on I could clean configure the whole cluster.
2023-10-15 00:10:53 <KipIngram   > I could work my way toward bare metal if I wanted to.
2023-10-15 00:11:09 <KipIngram   > That really hits a lot of the bits of digital technology I find interesting.
2023-10-15 00:12:59 <KipIngram   > MrMobius: I also once at least THOUGHT ABOUT niggling around with cheap programmable logic like 22V10's and try to make a "serial computer.  That is, one that processed its words bit serially.  the idea would be to hugely minimize the necessary logic by basically making it a "bit slice" one bit wide.
2023-10-15 00:13:19 <KipIngram   > Of course it'd be extremely slow.
2023-10-15 00:14:08 <KipIngram   > I'd probably make a concession to full width and make the RAM data bus wide.  Store and fetch RAM words, but use the read data and load the data to write serially.
2023-10-15 00:14:30 <KipIngram   > The metric of interest would be how many gates I needed for that.
2023-10-15 00:15:07 <MrMobius    > KipIngram: neat idea. did you see the 8 chip computer that was hackaday recently?
2023-10-15 00:15:15 <KipIngram   > How small is it possible to make something that actually computes?
2023-10-15 00:15:16 <MrMobius    > 800 cycles per instruction, 1 bit ALU
2023-10-15 00:15:26 <KipIngram   > No, I didn't.
2023-10-15 00:15:35 <KipIngram   > Ok, that sounds like it's right up the same alley.
2023-10-15 00:18:07 <KipIngram   > Do you have a linK?
2023-10-15 00:18:43 <MrMobius    > https://hackaday.com/2023/08/20/only-8-chips-make-a-cpu/
2023-10-15 00:19:10 <MrMobius    > I thought about this with a 4 bit ALU but got stuck on buffering addresses. you need a lot of logic for that
2023-10-15 00:19:19 <MrMobius    > dunno how the one I linked does it
2023-10-15 00:23:35 <KipIngram   > I might g look for a serial RAM chip.  Shift in the address, shift the data in or out.
2023-10-15 00:23:53 <KipIngram   > Because yeah - full-width ram logic defeats the whole point, really.
2023-10-15 00:25:18 <MrMobius    > I have some 128K SRAMs in DIP8 that work pretty well
2023-10-15 00:25:51 <KipIngram   > https://www.microchip.com/en-us/product/23A1024#document-table
2023-10-15 00:26:06 <KipIngram   > That's a 1-bit serial ram chip.
2023-10-15 00:26:23 <KipIngram   > https://ww1.microchip.com/downloads/aemDocuments/documents/MPD/ProductDocuments/DataSheets/23A102423LC1024-1-Mbit-SPI-Serial-SRAM-with-SDI-SQI-Interface-20005142.pdf
2023-10-15 00:26:29 <MrMobius    > maybe the external parallel eeprom is the trick. you could feed the command byte for read into the serial eeprom or sram that way
2023-10-15 00:26:44 <KipIngram   > Eight pin package.
2023-10-15 00:27:03 <MrMobius    > there ya go
2023-10-15 00:27:09 <MrMobius    > same as what i use but mine is 3v
2023-10-15 00:37:01 <MrMobius    > the neat thing about those is you can put them in QSPI mode and get 4 bits out per clock cycle
2023-10-15 00:37:13 <MrMobius    > one guy was using them as a poor man's VGA
2023-10-15 05:09:16 <veltas      > Croran: I agree, gforth tutorial is actually quite good
2023-10-15 05:09:40 <veltas      > I agree with KipIngram as well, the best course is practice
2023-10-15 05:09:53 <veltas      > Practice writing code for solving problems
2023-10-15 05:17:05 <hello-operator> MrMobius: Do you know who that was? The VGA thing?
2023-10-15 05:17:22 <hello-operator> Here's a bit-serial CPU that runs Forth - https://github.com/howerj/bit-serial
2023-10-15 06:56:45 <veltas      > I am sick and tired of irssi, any terminal-based IRC client suggestions?
2023-10-15 06:57:11 <veltas      > Serious suggestions, not suckless IRC client which I'm not even allowed to connect to libera.chat with
2023-10-15 06:57:56 <veltas      > For some reason libera.chat don't allow unauthenticated connections from some IP's, which IMO is lazy moderation but it is what it is
2023-10-15 06:58:30 <veltas      > At least there's less spam... but IRC is meant to be a "wet string + blood + sweat" protocol
2023-10-15 06:58:59 <veltas      > But anyway ... any terminal IRC client suggestions?
2023-10-15 07:17:45 <MrMobius    > veltas: why terminal only? just curious
2023-10-15 07:19:00 <MrMobius    > hello-operator: https://fleasystems.com/flea86.html
2023-10-15 07:19:04 <MrMobius    > thanks for the link!
2023-10-15 07:26:27 <veltas      > Because it's the simple self-hosted option I can access anywhere on my server
2023-10-15 07:28:09 <veltas      > Everything else is harder to maintain, wants to sell me things based on what I say, or otherwise runs on "some computer" I don't trust
2023-10-15 08:52:57 <MrMobius    > :/
2023-10-15 08:53:33 <MrMobius    > I pay $6/month for IRCCloud. well worth it though doesnt seem to fit your requirements
2023-10-15 08:53:41 <MrMobius    > a lot of channels are logged anyway so I dont care about privacy
2023-10-15 10:36:38 <KipIngram   > veltas: I run weechat-curses on a rented virtual server.  Could you perhaps do the same thing using your Oracle free thingie?
2023-10-15 10:36:50 <veltas      > Probably
2023-10-15 10:37:00 <veltas      > That's probably my option, weechat
2023-10-15 10:37:17 <KipIngram   > I connect to it with mosh, which causes it to behave like an always on terminal session on my notebook.  I don't have to re-connect every time I wake the notebook up.
2023-10-15 10:37:20 <KipIngram   > It's just "there."
2023-10-15 10:37:24 <KipIngram   > Totally painless.
2023-10-15 10:38:45 <KipIngram   > I've just been so thrilled with mosh over the years - nice simple little tool that does exactly what it promises to do.
2023-10-15 10:39:34 <KipIngram   > You do have to putz a little with opening firewall ports for mosh when you first install it it.
2023-10-15 10:40:09 <KipIngram   > But there's no other pain - basically, install it on your server and your client, open the right ports on the server, and then just use mosh instead of ssh to open your connection.
2023-10-15 11:15:51 <veltas      > I'm fine with ssh
2023-10-15 12:12:46 <KipIngram   > Fair enough.  I regard mosh as likely my favorite piece of software.
2023-10-15 12:15:01 <KipIngram   > I particularly like that it was written by a grad student.
2023-10-15 12:15:32 <KipIngram   > I compare it to my fancy AT&T written VPN tool I use from work, and mosh just leaves it in the dust.
2023-10-15 12:16:05 <KipIngram   > It meets my definition of an "appliance."  Does just the one thing, but does it reliably and well.
2023-10-15 12:16:50 <KipIngram   > I have the same attitude toward my ereader.  I can't watch movies and play games with it, but that's not what it purports to offer me.  Its only purpose is to be a library I can carry around with me.
2023-10-15 12:17:15 <KipIngram   > Phones were once appliances.
2023-10-15 12:17:57 <KipIngram   > I used to use a Motorola Razor.  This one:
2023-10-15 12:17:59 <KipIngram   > https://www.macitynet.it/wp-content/uploads/2016/05/Motorola-RAZR.jpg
2023-10-15 12:18:18 <KipIngram   > That was an appliance, and I was actually WILLING to shove it in my pocket and walk around with it.
2023-10-15 12:18:41 <KipIngram   > it's been many years since I had a phone I felt safe routinely pocketing.
2023-10-15 12:19:09 <KipIngram   > I do slip my smartphone into my hip pocket now and then, but I have to take it out when I sit down.  I can't just pocket it and forget about it unless it rings.
2023-10-15 12:19:25 <KipIngram   > that doesn't actually feel like progress to me.
2023-10-15 12:32:42 <KipIngram   > So I've complained at times about how we teach our kids physics in a way that leaves them entirely clueless about numerous important things.  i've tried to be careful to note that I'm  not saying all those details should actually be TAUGHT at that level - I more just think an "awareness" should be conveyed.  It occurred to me this morning that a way of saying that is that we should avoid leaving the kids with
2023-10-15 12:32:44 <KipIngram   > a whole pile of "unk-unks."  *Known* unknows are fine - we can't teach them *everything*.  But unknown unknowns somehow strike me as equating more fully to "ignorance."
2023-10-15 12:33:02 <KipIngram   > Unk unks are what cause you to overrate your own knowledge.
2023-10-15 12:33:58 <KipIngram   > One problem with trying to make things more like I'd like, though, is that we probably couldn't find enough teachers.
2023-10-15 12:34:14 <KipIngram   > My guess is that most high school physics TEACHERS don't know these things either.
2023-10-15 12:34:58 <KipIngram   > It would be interesting to sample a few thousand high school physics teachers and find out how many of them even know, for example, that the cross product is technically only valid in three dimensions.
2023-10-15 12:37:35 <KipIngram   > I get Quora question after Quora question that clearly came up because the asker is assuming their physics training is a) complete and b) exact.
2023-10-15 12:38:16 <KipIngram   > By "complete" I don't mean "all of physics" - I mean "tells the whole story" about the particular issue at hand.
2023-10-15 12:39:48 <Zarutian_iPad> that is rather interesting ?cognative error?
2023-10-15 12:40:54 <Zarutian_iPad> bythe students and possibly teachers that is
2023-10-15 12:41:52 <Zarutian_iPad> perhaps it is partially at fault with how their textbooks are written
2023-10-15 12:43:22 <Zarutian_iPad> I recall comming across a chemestry textbook that simply stated that it was not the complete thing on chemistry and even pointed out stuff that was in active research when the textbook was written
2023-10-15 13:06:26 <KipIngram   > Yes - this is something I've given a good bit of thought to.  Just as an example, "momentum = mass * velocity."  Well, yes - that is how you determine the momentum associated with a massive object.  But most kids leave school regarding that as a "complete definition of momentum," as in that's the *only way* momentum can arise.  That's just not so.
2023-10-15 13:07:55 <KipIngram   > And all the equations you learn in high school physics are deterministic equations, and most kids leave school regarding the universe as a deterministic clockwork.  I don't think we should try to teach quantum theory in high school, but I do think we should at least mention it and make the kids aware that all the equations they learn to use are actually just (very good) approximations of a fundamentally
2023-10-15 13:07:57 <KipIngram   > non-deterministic situation.
2023-10-15 13:10:44 <Zarutian_iPad> I think it is because there is too much ?math-purity? that sneaks in to the thinking of the teachers
2023-10-15 13:11:20 <KipIngram   > Actually it's pretty easy to take the bits of EM theory that get taught in high school and show that when an EM wave impinges on a metal sheet, the electric field will accelerate electrons sideways, and then the magnetic field will act on those moving charges and give the sheet a push in the direction of wave propagation.
2023-10-15 13:11:40 <KipIngram   > In other words, that EM wave had to be carrying momentum, because it pushes the object that it strikes.
2023-10-15 13:12:23 <KipIngram   > It wouldn't be much of a step to actually mention that - it's super easy to see.
2023-10-15 13:12:29 <Zarutian_iPad> carrying momentum? more like light excerts light pressure
2023-10-15 13:12:35 <KipIngram   > Semantics.
2023-10-15 13:13:28 <KipIngram   > The momentum is actually in the wave, because it takes time for the light to move from its source to its target.  The momentum is lost by the source at the time of emission, and isn't gained by the target until it's absorbed.
2023-10-15 13:13:34 <KipIngram   > In the interim is "in the wave."
2023-10-15 13:13:40 <KipIngram   > s/is/it's/
2023-10-15 13:14:02 <KipIngram   > You have to look at it that way if you want momentum conservation to hold at all times.
2023-10-15 13:14:18 <KipIngram   > It's certainly totally fine to think of it in terms of "pressure" too.
2023-10-15 13:14:35 <Zarutian_iPad> no, because the intutive thinking is that you need mass to carry momemtum which impacts against "but photons are massless!"
2023-10-15 13:14:40 <KipIngram   > So long as you get the right numbers when you're done.  :-)
2023-10-15 13:14:55 <KipIngram   > That's EXACTLY the problem I'm highlighting.
2023-10-15 13:15:00 <KipIngram   > Precisely.
2023-10-15 13:15:12 <KipIngram   > The kids get the impression that m*v is the ONLY way to get momentum.
2023-10-15 13:15:18 <KipIngram   > That's WRONG.
2023-10-15 13:15:45 <Zarutian_iPad> in my experience, confusion often arises because crappy similies were used without giving caveats
2023-10-15 13:15:52 <KipIngram   > there's another simple formula for the momentum of a photon, but it's not typically taught in high school.
2023-10-15 13:16:56 <KipIngram   > All we'd need to do, really, is just point out that m*v is the formula for the momentum of an *object with mass*.
2023-10-15 13:17:07 <Zarutian_iPad> btw I find the curiculum tought in USA highschools often be less than is taught heree in elementary school
2023-10-15 13:17:08 <KipIngram   > Something without mass requires a different formula.
2023-10-15 13:17:17 <KipIngram   > I wouldn't be surprised.
2023-10-15 13:17:25 <KipIngram   > Our public education in the US is pretty weak.
2023-10-15 13:17:45 <KipIngram   > And I probably don't even realize how weak, because my calibration is from the late 1980's.
2023-10-15 13:17:57 <KipIngram   > I think it's tragic.
2023-10-15 13:18:51 <KipIngram   > As I noted earlier, though, I expect at this point it's horribly entrenched, because we now probably have most TEACHERS not really being good enough to teach the better material.
2023-10-15 13:19:01 <KipIngram   > So it's endemic now.
2023-10-15 13:20:38 <Zarutian_iPad> one thing I have seen whilist reading about 'logos on teaching' is certain ?arrogance/assumptions/pressumptions? made
2023-10-15 13:22:47 <KipIngram   > Um, not sure what you mean?
2023-10-15 13:23:19 <Zarutian_iPad> I am not interested in child-rearing of schools but in how to convey material/information about a thing
2023-10-15 13:24:59 <Zarutian_iPad> but the ?assumption? about a student certain kind of textbooks is on the line of someone immature
2023-10-15 13:26:44 <Zarutian_iPad> and this is usally from newish style textbooks whilist older ones just gets on with it.
2023-10-15 13:26:47 <KipIngram   > I still don't think I'm tracking you completely, but if your point is that the focus should be on conveying knowledge rather than on "raising children," then I mostly agree with you.
2023-10-15 13:29:09 <KipIngram   > I guess there's no way to 100% separate the processes of "education" and "socialization," but I definitely think that the further along a kid gets in school the more things should shift to "pure education."  Socialization should only be significant at all in the very early years of education.
2023-10-15 13:29:52 <Zarutian_iPad> a good textbook on trigonometry went a bit into the history of the problems that gave arise to it, how sinus and cosinud are described from the unit circle, etc.
2023-10-15 13:29:59 <KipIngram   > There was a TV show here in the US back in the late 1970s, early 1980s, called "The Paper Chase."  It was about a bunch of kids in law school.
2023-10-15 13:30:25 <Zarutian_iPad> cosinus*
2023-10-15 13:31:03 <KipIngram   > Now, obviously law school is not primary/secondary education, but the atmosphere of the law school that was preesented was a very focused one - there clearly was an attitude of "You will have knowledge presented to you.  You will sink or swim.  We don't care which one, but good luck."
2023-10-15 13:31:21 <KipIngram   > I found it appealing.
2023-10-15 13:31:21 <thrig       > various higher education teachers have complained about students not really being socialized
2023-10-15 13:32:22 <KipIngram   > The school (in that show) clearly did not regard itself as having the duty of "ensuring the success of every student."
2023-10-15 13:32:34 <KipIngram   > In fact, most of the students would ultimately fail.
2023-10-15 13:32:49 <KipIngram   > And I actually think that's an appropriate mindset in the highest levels of education.
2023-10-15 13:35:43 <Zarutian_iPad> one thing I never understood is when the switch from teaching certain classes/modules as units in themselfs to full huge semisters only
2023-10-15 13:40:22 <KipIngram   > Oh, hmmm.  Well, I've no exposure to anything other than the semester setting.  As in, after the dust all settled there was one grade per course per semester - the "internal assessments" that led to that just vanish once the final grade is turned in.
2023-10-15 13:41:42 <Zarutian_iPad> I have actively badgered teachers that just graded my work and did not give any feedback
2023-10-15 13:43:04 <KipIngram   > :-)  I had a professor in grad school that I took as much math from as I possibly could.  I'd had him for one course as an undergrad, and he was just that good.  But... oh my god.  The assignments you turned in just fell into a black hole, and you heard NOTHING from him.  I did go see him once and asked about it - he listened politely and smiled and sent me on my way, and nothing changed.
2023-10-15 13:43:09 <KipIngram   > That IS a frustrating situation.
2023-10-15 13:43:37 <Zarutian_iPad> a diploma certificate is just that, a cerificate but does not tell iff I or the individual in question actually knows the stuff
2023-10-15 13:44:07 <KipIngram   > I don't believe he ever CALCULATED anyone's grade.  I think at the end of the semester he pulled out their work, leafed through it, and thought something like "Ok, this guy knows what he's doing - he gets an A."  Or whatever.
2023-10-15 13:44:38 <KipIngram   > But he was extremely good at teaching the material.
2023-10-15 13:45:08 <KipIngram   > Right - not at any fine level of detail.  It's just a very rough statement that "this person survived the process."
2023-10-15 13:45:57 <KipIngram   > Employers and graduate schools can ask you for transcripts, and at least that shows individual grades for individual courses.  But that's the finest level of information that remains permanently available.
2023-10-15 14:04:06 <KipIngram   > Wow>  I just browsed the website of the place I bought my telescope from.  Apparently they don't make that particular line of telescopes anymore, at least not in the size range I own.
2023-10-15 19:44:14 <hello-operator> MrMobius: Cheers! I'll take a look!
2023-10-15 20:19:49 <KipIngram   > I'm watching a suite of videos on electric motor design.  Brings back some old memories.  Great application for finite elements.
2023-10-15 20:20:19 <KipIngram   > Specifically this set is on switched reluctance motor design, but it looks like they've got video sets on a bunch of different kinds.
2023-10-15 20:21:57 <KipIngram   > I've always found finite element methods really cool.
2023-10-15 20:27:20 <KipIngram   > Pretty simple kind of motor - basically looks like this inside:
2023-10-15 20:27:22 <KipIngram   > https://www.eeeguide.com/wp-content/uploads/2019/01/Switched-Reluctance-Motor.jpg
2023-10-15 20:28:10 <KipIngram   > The rotor pole is pulled toward alignment with whichever stator pole is energized at the time, and the pole count on the stator and rotor is such that there's always a clear "next pole" to turn on to continue the motion.
2023-10-15 20:28:28 <KipIngram   > So the controller just activates the windings in sequence, and the rotor spins up.
2023-10-15 20:29:37 <KipIngram   > In some ways all of this has gotten a lot simpler since we got microcontrollers - before that you had to work out some sort of angle dependent switching mechanism.  But with a uC you can just do pretty much anything.
2023-10-15 20:30:51 <KipIngram   > Switched reluctance is one of the easier to understand motor types.
2023-10-15 23:30:33 <KipIngram   > Ah, this motor design involves a semi-interesting optimization method.  The software lets you enter all the relationships among an arbitrarily chosen set of variables, depending on whatever it is you're doing.  But you can also flag variables as "optimizable" or not (some of them will just depend on variables you are optimizing over, and you don't want to independently change those).  Then when you have it
2023-10-15 23:30:35 <KipIngram   > all set up the software has the info it needs to run some general optimization procedure.
2023-10-15 23:30:47 <KipIngram   > Guess it's really more about how the interface works than how the calculations are done.
2023-10-15 23:30:59 <KipIngram   > Those extra layers of variables sort of become 'convenience variables.'
2023-10-15 23:31:14 <KipIngram   > But don't add any  new degrees of freedom.
2023-10-15 23:31:46 <KipIngram   > In a Forth system each of those could have a corresponding word that calculates them.
2023-10-15 23:32:21 <KipIngram   > You could collect them all into a self-contained vocabulary, and the optimization procedure could run against that vocabulary  name.
2023-10-15 23:32:37 <KipIngram   > It could update all of the variable values in the vocabulary.
2023-10-15 23:32:46 <KipIngram   > It's a nice way of "containing" a piece of work.
2023-10-15 23:33:14 <KipIngram   > I like ideas that leverage the dictionary.
2023-10-15 23:34:29 <KipIngram   > You could also put the optimizable variables in a particular vocabulary within the "problem vocabulary," rather than setting per-word flags.  That would segregate the variables just as well as flags.
2023-10-15 23:36:13 <KipIngram   > I'm watching this video and looking at all the morass of GUI widgets that are involved in telling the software how to know what to do - neat that a plain old Forth vocabulary structure could do it just as well.
2023-10-15 23:51:36 <KipIngram   > After all, a vocabulary is a linked list, and in each case the optimization variable itself would be at address *(*(p+cell)).  Probably about as efficient as any method of identifying a particular subset of a collection of variables.
2023-10-15 23:52:29 <KipIngram   > Oh, wait - that would be the value of the variable.  Drop the leading * to get the address.  And of course that's dependent on my implementation details.
2023-10-15 23:53:48 <KipIngram   > Anyway, the new wrinkle here that I hadn't thought of before is to actually use the dictionary organization as pertinent to the actual algorithm you're running, rather than just as a way to find words at compile time.