2023-11-02 01:07:59 <KipIngram   > :-)  Well, fanfic really doesn't ever fall into my sandbox.
2023-11-02 01:08:38 <KipIngram   > In case anyone is interested, i^i = 0.20787958...
2023-11-02 01:08:44 <KipIngram   > i = sqrt(-1).
2023-11-02 01:09:28 <KipIngram   > Interesting what a tangled knot all that stuff forms.
2023-11-02 01:29:16 <KipIngram   > Good first evening with Brave.
2023-11-02 01:29:27 <KipIngram   > Still loving the clear lower power operation.
2023-11-02 01:30:08 <KipIngram   >  By the way.  Saw a video last night on Shamir's Secret Sharing algorithm.
2023-11-02 01:31:17 <KipIngram   > Say yo have a number you want to keep secret.  Choose to integers M and N.  Imagine that your number is the y-intercept of an M-1'th order polynomial.
2023-11-02 01:32:54 <KipIngram   > You can choose a bunch of points that you want to lie, along with your number on the y axis, along an Mth-order polynomial.
2023-11-02 01:33:56 <KipIngram   > Or maybe it's M+1.  Anyway, Now choose N value of x, evaluate the poly at those points, and give the results to N people.
2023-11-02 01:34:11 <KipIngram   > The idea is that any M of them can recover your secret, but no group less than M can.
2023-11-02 01:34:32 <KipIngram   > To really make it work well and completely, you do it on a prime-sized finite field, but the idea is the same.
2023-11-02 01:35:04 <KipIngram   > <M people simply have no information that lets them make any progress toward getting your secret, but M people can immediately recover it.
2023-11-02 01:35:45 <KipIngram   > I've always thought that's how distributed storage should work - there should be many vendors of cloud-based storage, and you'd use Shamir's method to spread your data out across a large number of them.
2023-11-02 01:36:01 <KipIngram   > Some of them could go bust and disappear, but as long as M of them were still running you could read your data.
2023-11-02 01:36:15 <KipIngram   > But without M-collusion among the vendors, none of them could.
2023-11-02 01:39:42 <KipIngram   > I think you'd still want to use encryption on top of that; otherwise someone intercepting all of your traffic could see your data.  It would just be an additional level of protection.
2023-11-02 03:55:47 <olle        > Good morning, will check chat history :)
2023-11-02 04:01:50 <olle        > 18:16 < GeDaMo> What do you want to do with the xts?
2023-11-02 04:02:03 <olle        > Probably loop them over a collection, as one example.
2023-11-02 04:03:06 <olle        > KipIngram: Interesting use-case with op
2023-11-02 05:45:34 <olle        > Are there any Forth transpiler with "industry" production quality? Or is that just not a thing. Can't find anything online, realy.
2023-11-02 05:46:11 <olle        > Well compile to ASM is kinda transpile, perhaps.
2023-11-02 07:03:51 <olle        > Asking the gforth mailing list about hooking into it. :)
2023-11-02 07:04:50 <crc         > olle: http://forth.works/temp/pipe.txt would be a quick implementation of your pipe question in my konilo forth
2023-11-02 07:07:36 <crc         > (for retroforth, s:get/token -> s:get-word and d:address -> d:xt )
2023-11-02 07:31:29 <rendar      > let's think about this: in my Forth each word is mapped into an integer, which is a kind of byte code the vm executes. Now, when i declare new words, they will be vectors of Atom's (which are 1, 2, 4.0, 0.1, 'hello' and so on) and these integers representing words
2023-11-02 07:32:12 <rendar      > my question is: in defining a new word which uses an *USER DEFINED* word, should that new word copy all builtin integers?
2023-11-02 07:32:25 <rendar      > or it may just "call" the other chunk
2023-11-02 07:32:45 <GeDaMo      > That would be the equivalent of inlining, it might depend on how big the called word is
2023-11-02 07:33:06 <rendar      > exactly, what about always inlining?
2023-11-02 07:33:31 <gordonjcp   > rendar: probably not necessary
2023-11-02 07:33:38 <rendar      > if i always inline, i won't ever need a call stack!
2023-11-02 07:33:44 <gordonjcp   > true
2023-11-02 07:34:02 <rendar      > that would be an huge advance
2023-11-02 07:34:10 <gordonjcp   > it's probably a win where the function of the word is smaller than the setup and teardown for calling that word
2023-11-02 07:34:18 <rendar      > yeah
2023-11-02 07:34:28 <gordonjcp   > inlining arithmetic operators, LIT, and so on
2023-11-02 07:34:57 <gordonjcp   > again though it depends on how quickly you can jump in and out of a word, which depends on the CPU architecture
2023-11-02 07:35:14 <rendar      > well
2023-11-02 07:35:24 <gordonjcp   > which really comes down to how easy it is to implement NEXT
2023-11-02 07:35:34 <rendar      > for me is also a security concern, i will always have function that run KNOWN words
2023-11-02 07:35:38 <gordonjcp   > NEXT is a bit of a pig on a 6502, for example
2023-11-02 07:35:40 <rendar      > so it can be better analyzed
2023-11-02 07:35:59 <gordonjcp   > you can win a lot by not having to jump out and back in
2023-11-02 07:36:08 <rendar      > yeah
2023-11-02 07:36:34 <rendar      > wasn't 6502 Apple II GS CPU?
2023-11-02 07:36:47 <gordonjcp   > on say 68000 or 6809 where it's like two (long, many clocks) instructions, the code is clean and actually it'll save you like six clocks but inlining might not help much
2023-11-02 07:36:56 <GeDaMo      > The GS used the 65816, 16-bit extension of the 6502
2023-11-02 07:36:58 <gordonjcp   > no, Apple II up to the GS, GS had an
2023-11-02 07:36:59 <gordonjcp   > ye
2023-11-02 07:37:04 <gordonjcp   > 65816 was cool
2023-11-02 07:37:04 <rendar      > oh!
2023-11-02 07:37:25 <gordonjcp   > it ran like a 6502 "normally" but you could flip it into a mode where it had 16-bit operators
2023-11-02 07:37:29 <gordonjcp   > and more registers
2023-11-02 07:37:32 <rendar      > indeed it could run ProDOS 8 and 16
2023-11-02 07:37:38 <GeDaMo      > I thin k it could address 16MB of memory
2023-11-02 07:37:43 <gordonjcp   > ye
2023-11-02 07:38:04 <gordonjcp   > well it had paging hardware built in, it couldn't really but it could fake it well enough to look like it cared
2023-11-02 07:51:15 <olle        > http://www.complang.tuwien.ac.at/papers/ertl%26maierhofer95.pdf Translating Forth to efficient C
2023-11-02 07:51:20 <olle        > 1995 hm
2023-11-02 08:09:10 <olle        > crc: thanks :)
2023-11-02 08:11:04 <KipIngram   > olle: I think it's just that Forth "transpiling" isn't that big of a thing - maybe I don't use that word right, but it makes me think of a *compiler*.  Forth is primarily intended to be an interactive system, not a tool to compile source to binary like we usually do with C.
2023-11-02 08:13:52 <olle        > Yep
2023-11-02 08:14:09 <olle        > All compilers "transpile" to asm code after all :)
2023-11-02 08:18:05 <KipIngram   > Actually indirect and direct threaded Forth systems transpile to data structures.  You can, though, think of it as code for some virtual machine.
2023-11-02 08:19:13 <KipIngram   > In those types of system you don't make any new "machine code" unless you break out the Forth's assembly tools.
2023-11-02 08:22:04 <olle        > What is code for a VM?
2023-11-02 08:22:17 <olle        > The Forth words?
2023-11-02 08:23:16 <GeDaMo      > Forth is usually some form of threaded code
2023-11-02 08:23:28 <GeDaMo      > But could also be bytecode
2023-11-02 08:25:47 <olle        > Yea but compile-time behaviour makes it kinda not bytecode at all haha
2023-11-02 08:26:17 <gordonjcp   > I should start teaching my toddler Forth
2023-11-02 08:26:21 <GeDaMo      > compile-time and run-time are more of a spectrum than in most languages :P
2023-11-02 08:26:51 <gordonjcp   > he already talks in postfix notation a lot of the time, because he's saying it in German but using a lot of English words
2023-11-02 08:27:54 <gordonjcp   > he's at the age where we his parents pretty much understand him just fine, his friends at nursery understand him just fine, but the nursery staff don't always catch what he's saying
2023-11-02 08:28:07 <gordonjcp   > so he repeats it very slowly and clearly like he's talking to someone a bit slow
2023-11-02 08:28:28 <gordonjcp   > and quite often, slowly and clearly but in German, which actually just makes them understand it less
2023-11-02 08:38:42 <KipIngram   > VM - the code that navigates through Forth definitions and brings about execution of the correct sequence of primitives.
2023-11-02 08:39:32 <KipIngram   > In a code-threaded Forth that IS machine code itself - the forth compiler builds definitions out of machine code.  But in direct and indirect threaded systems it builds lists of addresses, and something has to scan through those address lists and call the associated code.
2023-11-02 08:39:38 <KipIngram   > That code is what I'm calling a VM.
2023-11-02 08:41:11 <KipIngram   > gordonjcp: I think Forth would actually be relatively easy for small children to learn, at least once they can read.
2023-11-02 08:41:21 <KipIngram   > It's a very simple way of communicating with a computer.
2023-11-02 08:45:23 <KipIngram   > olle: I don't quite understand your remark about compile-time behavior making it "not bytecode."
2023-11-02 08:46:15 <KipIngram   > Indirect and direct threading are distinctly different from bytecode.  When I think of bytecode I think of there being a byte-indexable table that you jump through to execute code associated with your byte token.
2023-11-02 08:46:58 <KipIngram   > There is no such table in direct and indirect threaded systems - instead you jump through your definition cells (direct threaded) and additionally through your definition headers (indirect threading).
2023-11-02 08:48:04 <GeDaMo      > You can think of all of memory as a table :P
2023-11-02 08:48:15 <KipIngram   > The new one I'm planning will have a jump table; I don't know yet if my indexing tokens will be bytes; they may be smaller.
2023-11-02 08:48:20 <KipIngram   > Yes, that's true.
2023-11-02 08:55:58 <KipIngram   > But I still think of the "entries" as having more "direct" meaning than in a bytecode system.  They're actually addresses.  You couldn't, for example, just re-arrange entries in a table and change the byte code values.
2023-11-02 08:57:33 <KipIngram   > I'm still torn over exactly what to call the new system I'm planning.  It is going to have a very explicit vm and that vm will be a "byte" code interpreter.  So in that sense I think of it as "token threaded."  But that vm will execute a code-threaded ForTH (with the code being those vm instructions rather than native processor instructions).
2023-11-02 08:57:40 <KipIngram   > So it'll have elements of both architectures.
2023-11-02 09:37:05 <veltas      > rendar: Why avoid callstack? Not an advantage, and you're generating exponential code size for that
2023-11-02 09:37:39 <veltas      > I'm constantly seeing this idea on here and don't get it at all, it's not a 'saving', the return stack is there for a reason
2023-11-02 09:38:09 <veltas      > It's also just plain convenient having a second stack to save/restore things on
2023-11-02 09:38:09 <KipIngram   > I feel the same.  I know yo're not actually writing a "Forth' per se, but eliminating the call stack makes it "really really super not Forth."  :-)
2023-11-02 09:38:50 <KipIngram   > Many of Forth's characteristic features arise from the explicit access it gives you to that second stack - most languages hide it under the hood.
2023-11-02 09:39:33 <KipIngram   > I don't think I'm aware of any signficant language that wouldn't have one.
2023-11-02 09:42:01 <veltas      > It's a bit like having a macro assembler and only using macro expansion and no subroutines
2023-11-02 09:42:15 <KipIngram   > Yes - exactly like that.
2023-11-02 09:42:23 <veltas      > When you're writing your masm and the output hits the 2GB limit you'll find out why!
2023-11-02 09:42:33 <veltas      > why people use subroutines
2023-11-02 09:42:50 <KipIngram   > Among other things you lose the ability to use recursion other than tail-optimized "recursive iteration."
2023-11-02 09:43:45 <KipIngram   > Subroutines are so valuable that the industry has seen fit to provide hardware support for them.  It's WHY processors have stacks at all.
2023-11-02 11:28:50 <rendar      > veltas, consider i have to transfer my user defined function to another machine
2023-11-02 11:28:57 <rendar      > i have to calc all dependencies
2023-11-02 11:29:08 <rendar      > with all inlined functions, i simply send the code
2023-11-02 11:29:14 <rendar      > yeah you have bigger functions..
2023-11-02 11:30:27 <veltas      > Not really sure why remote target affects this
2023-11-02 11:30:37 <veltas      > If anything that's more reason to use subroutines
2023-11-02 11:41:46 <KipIngram   > It would be easy to use relative addressing for subroutine calls; you could then send the subroutines over as part of the same package.
2023-11-02 11:43:19 <KipIngram   > I'm planning relative addressed calls in my new system's vm - it will reduce the size of calls to "nearby" code.
2023-11-02 11:44:28 <KipIngram   > So, I don't see that as something that precludes ore unnecessarily complicates supporting subroutines.
2023-11-02 11:44:36 <KipIngram   > s/ore/or/
2023-11-02 16:14:07 <KipIngram   > We have a cleaning service that comes and does the house once a week.  Everytime they power on their vacuum cleaner, the lamp here in my family room flickers, and it's not uncommon for a breaker to trip and I have to go out the garage and turn it back on.
2023-11-02 16:14:38 <KipIngram   > I just took a look at their vacuum - there's no label on it, but it does identify as an "industrial" vaccum.  I'd really like to know what it's power consumption is.
2023-11-02 16:15:32 <KipIngram   > Surely they aren't trying to use something a 15 amp breaker won't service.
2023-11-02 16:16:05 <KipIngram   > Along with whatever other (small) loads might happen to be on that circuit.  It isn't anything other than lights.
2023-11-02 16:21:58 <KipIngram   > I am very very happy with this switch to the Browser.  What the heck was Google doing with all that energy that's now not being consumed???
2023-11-02 16:22:13 <KipIngram   > Phoning home all my information, no doubt.
2023-11-02 16:22:54 <Zarutian_iPad> wasting it on js ui overbloated frameworks
2023-11-02 16:23:26 <KipIngram   > I saw a video last weekend with some computer security guy interviewing that Occupy the Web dude; he was talking about Chrome sending everything about you off to Google.
2023-11-02 16:25:01 <Zarutian_iPad> 'Occupy the Web' sounds as dubious as 'Just Stop Oil' somehow
2023-11-02 16:25:19 <KipIngram   > He didn't really say much I didn't already assume, though - the topic of the interview was how to be anonymous online, but the final bottom line was pretty much "If you want to stay anonymous from commercial interests, you can do fairly well, but if you're trying to stay anonymous from the intelligence agency of your own country, you really have little hope."
2023-11-02 16:26:21 <Zarutian_iPad> KipIngram: did he talk about at all how to lay down falsetracks or polute the 'footprint'?
2023-11-02 16:26:22 <KipIngram   > Oh, don't get me started on oil.
2023-11-02 16:26:36 <KipIngram   > He did talk about sowing false identity information, yes.
2023-11-02 16:26:47 <KipIngram   > He didn't give a step by step uide or anything, thouh.
2023-11-02 16:26:59 <KipIngram   > My wife works in the oil industry.  Offshore subsea engineering.
2023-11-02 16:27:03 <Zarutian_iPad> good sounds like he knew a bit whathe was talking about
2023-11-02 16:27:22 <KipIngram   > The very idea that we can "stop oil" and in any way maintain the operation of the planet is brain dead.
2023-11-02 16:28:10 <Zarutian_iPad> the 'Just Stop Oil' thing is akin to PETA or other such associations that have only function to paint certain activism in bad light
2023-11-02 16:28:12 <KipIngram   > I mean, I'd love to see us develop a really good battery technology and switch to a primarily EV fleet.  That would be great for the environment, from an emissions perspective.
2023-11-02 16:28:33 <KipIngram   > But the energy still has to come from somewhere, and renewables just can't do it all by themselves.
2023-11-02 16:28:41 <KipIngram   > Oil and/or nuclear are unavoidable.
2023-11-02 16:29:19 <KipIngram   > I'm not "opposed" to renewables, but I think it's rare to see them assessed in a properly objective way.
2023-11-02 16:30:23 <Zarutian_iPad> one idea I have and told any investor-type that willlisten is carbon capture via growing bamboo/bambus in say high insolation places like Nigeria or other places where you can (partially) desalinate sea water for irrigation.
2023-11-02 16:30:31 <KipIngram   > a few years ago a guy at my alma mater announced a new "lithium glass" battery technology that looked like a huge stride forward.  More life cycles, safter, more energy and power density, and (most important) it charged in minutes in stead of hours.
2023-11-02 16:30:40 <KipIngram   > We need something like that to make EV's really feasible.
2023-11-02 16:30:55 <Zarutian_iPad> (there are bambus variants that tollerate brackish water)
2023-11-02 16:30:58 <KipIngram   > You need to be able to pull into a "charging station" and fully charge up in minutes, just like you do with gasoline now.
2023-11-02 16:31:34 <KipIngram   > I've been watching to see if it looks like that battery is getting commercialized, but I don't see it yet.
2023-11-02 16:31:39 <KipIngram   > Maybe something went wrong with it.
2023-11-02 16:31:52 <KipIngram   > And besides, I don't think we have enough lithium.
2023-11-02 16:32:19 <Zarutian_iPad> my father drives a plugin hybrid which means he often can go the daily stuff on electricity alone.
2023-11-02 16:32:39 <KipIngram   > Sure - overniht charging helps - it can allow us to shift some of the usage.
2023-11-02 16:32:48 <KipIngram   > But to shift completely we need what I mentioned above.
2023-11-02 16:33:01 <KipIngram   > For times when you want to go on a serious road trip.
2023-11-02 16:33:53 <Zarutian_iPad> wasnt there an project of making supercaps via replacing beðmi (summer wood?) with aluminum or other such in wood?
2023-11-02 16:33:54 <KipIngram   > I drove a hybrid for a few years.
2023-11-02 16:34:01 <KipIngram   > A Honda Civic.
2023-11-02 16:34:20 <KipIngram   > It was great - I made frequent trips to Austin, and when it was new I could get 65 mpg on that trip.
2023-11-02 16:34:30 <KipIngram   > But it didn't stay that way - the batter degrades.
2023-11-02 16:34:46 <KipIngram   > After a while it didn't work as well, and then I found out that they wanted people to bring the cars in for a firmware upgrade.
2023-11-02 16:34:53 <Zarutian_iPad> (tréni is lignin but the other one I do not recall the english term for)
2023-11-02 16:35:08 <KipIngram   > Found out later the firmware upgrade just had it lower the capacity minimum so it wouldn't throw the errors anymore.
2023-11-02 16:35:15 <KipIngram   > Didn't improve my mileage one bit.
2023-11-02 16:35:36 <KipIngram   > and to replace those batteries is a major, MAJOR expense.
2023-11-02 16:36:54 <KipIngram   > I think the battery technology just isn't as good as it really needs to be for that application.
2023-11-02 16:40:16 <Zarutian_iPad> so, would you buy a small solar/wind powered gasoline maker that makes two liters every three days?
2023-11-02 16:40:53 <KipIngram   > What's it going to make it out of?
2023-11-02 16:41:01 <KipIngram   > the materials, I mean.
2023-11-02 16:41:40 <KipIngram   > It's hard to imagine that could possibly save money over buying it at the gas station.
2023-11-02 16:41:44 <Zarutian_iPad> CO2 from air and water from water main (less than you washing machine uses)
2023-11-02 16:41:54 <KipIngram   > Given the economies of scale behind that industry.
2023-11-02 16:42:20 <KipIngram   > It's an interesting idea.  I wouldn't mind learning more about it - is there actually such a thing?
2023-11-02 16:42:37 <Zarutian_iPad> not thinking of it as replacement but more as an suppliment
2023-11-02 16:43:10 <Zarutian_iPad> sadly, no. Still quite energy inefficient at this small scale
2023-11-02 16:44:22 <Zarutian_iPad> but from explorative engineering perspective it is quite doable but there is a reaction/catalyst component that is very very hard to make with current technology afaict
2023-11-02 16:45:31 <Zarutian_iPad> much easier to make methane but people are not used to handle car refilling when the fuel is literally a gas
2023-11-02 16:45:52 <Zarutian_iPad> though home gas grills are a thing... hmm..
2023-11-02 16:50:13 <KipIngram   > I'd find the process interesting, I'm sure.  And there might be a lot of people out there who would be happy about taking CO2 out of the atmosphere - that might give it a market in spite of it not being efficient yet.
2023-11-02 16:50:33 <KipIngram   > I.e., there are people who might have motivations other than just pure financial.
2023-11-02 16:51:56 <Zarutian_iPad> but I think bambus is the best short term bet
2023-11-02 16:52:18 <Zarutian_iPad> its wild how fast it grows
2023-11-02 16:52:58 <KipIngram   > Yeah, I've heard that.  In ww2 I heard the Japanese sometimes tied people to the ground over those things to get them to talk - lying on them apparently doesn't stop them from growing.
2023-11-02 16:54:10 <KipIngram   > I think we have things left to learn in the areas of chemical and organic technology.  In fact, I think that's the "killer app" for quantum computing - it will let us simulate MUCH more complex reactions that we can today.
2023-11-02 16:54:39 <KipIngram   > I think success on the quantum computing front (real success) could open up a "golden age" of designer "super chemicals."
2023-11-02 16:55:24 <KipIngram   > And some of the microorganisms that live down near those vents in the bottom of the oceans - I've heard ideas about using them in fuel cells.
2023-11-02 16:55:43 <Zarutian_iPad> well, we could 'steal' a lot from nature using proteins and even make denovo proteins
2023-11-02 16:55:46 <KipIngram   > My point being that I think there's a lot still on the table on those fronts that we haven't gotten smart enough to figure out yet.
2023-11-02 16:57:12 <KipIngram   > Right - designer proteins too.  Lot of potential I think.
2023-11-02 16:57:32 <KipIngram   > Maybe we get a century on that front that looks like what the last one looked like in electronics.
2023-11-02 16:57:59 <KipIngram   > I mean, we're not that far from the theoretical limit of semiconductor tech, i think.  The speed with which we figured it all out is staggering.
2023-11-02 16:58:01 <Zarutian_iPad> also dna brick origami, pna, oligios and such for say Integrated Circuit wafer patterning
2023-11-02 16:58:28 <KipIngram   > Considering that the transisor wasn't invented until 1950-ish.
2023-11-02 16:58:32 <KipIngram   > That was t=0.
2023-11-02 16:58:54 <KipIngram   > and a few years ago I heard that a lab in Israel had a working (in the lab) transistor made of just 12 atoms.
2023-11-02 16:59:05 <KipIngram   > That's damn close to the limit; it has to be.
2023-11-02 16:59:09 <Zarutian_iPad> the issue with current methods of patterning ICs is the expense of the equipment and specially the masks
2023-11-02 16:59:24 <KipIngram   > So if we commercialize that, then that's zero to the limit in under a hundred years.
2023-11-02 17:00:07 <KipIngram   > Yeah.  I really, really wish that had gotten to the point where it was feasible to do it at home (on small scale, of course).
2023-11-02 17:00:48 <Zarutian_iPad> KipIngram: and that is electronics tech only. Optronics can give speedups quite beyond that but it is tricker
2023-11-02 17:01:39 <Zarutian_iPad> half micron ic patterning via light is well doable in a garage
2023-11-02 17:01:45 <KipIngram   > Yeah.  We're not "done" yet.  Several areas for us to make some major proress.
2023-11-02 17:01:49 <KipIngram   > progress
2023-11-02 17:02:05 <KipIngram   > Then we'll be bored.  :-)
2023-11-02 17:02:39 <Zarutian_iPad> heck using electron beam patterning you can go smaller than current euv but it takes a long while each wafer
2023-11-02 17:02:44 <KipIngram   > It's hard to overestimate how big an impact electronics has had on economic growth the last 100-120 years.
2023-11-02 17:03:02 <KipIngram   > Maxwell did his thing and put a bow around it, and the world just took off like a rocket.
2023-11-02 17:03:31 <KipIngram   > much like what thermodynamics and the steam engine did in the industrial revolution.
2023-11-02 17:03:58 <KipIngram   > I guess the real "critical mass" on chemistry also came in the early 20th century.
2023-11-02 17:04:07 <KipIngram   > So we kind of got a twofer that century.
2023-11-02 17:04:20 <Zarutian_iPad> Maxwell did his thing whilist studying the wax iron filing images Columb made of magnetic fields around conductors
2023-11-02 17:04:48 <KipIngram   > Was that Coulomb or Ampere?
2023-11-02 17:05:24 <Zarutian_iPad> Columb, Ampere studied conductence of materials iirc
2023-11-02 17:05:25 <KipIngram   > Ampere and Faraday established the basic observations.  Maxwell put them on a sound mathematical footing.
2023-11-02 17:05:46 <KipIngram   > If you JUST take FaradAy and Ampere's relations and render them as differential equations, they aren't consistent.
2023-11-02 17:05:49 <KipIngram   > They're broken.
2023-11-02 17:05:55 <KipIngram   > Maxwell saw that and saw how to fix it.
2023-11-02 17:06:16 <KipIngram   > He added the "displacement current" term to the curl equation of the B field.
2023-11-02 17:06:44 <KipIngram   > But Ampere's Law is the one that talks about currents making magnetic fields.
2023-11-02 17:06:48 <Zarutian_iPad> but I am not that familiar with the history of the discovery of electricty
2023-11-02 17:07:01 <KipIngram   > I associate Coulomb with charges making electric fields.
2023-11-02 17:07:17 <KipIngram   > God, I hope I am a little bit, given it's my profession.
2023-11-02 17:07:29 <KipIngram   > Doesn't mean I've got it all right, though.
2023-11-02 17:07:56 <KipIngram   > Anyway, it's that displacement current term that allows EM wave propagation.
2023-11-02 17:08:02 <Zarutian_iPad> same here though my profession is more hands on than mathematical
2023-11-02 17:08:03 <KipIngram   > Radio and TV.
2023-11-02 17:08:43 <Zarutian_iPad> to me antenna stuff is like black magic
2023-11-02 17:08:45 <KipIngram   > I'm a big fan of how once you take on Coulomb's law, then special relativity makes magnetism required.
2023-11-02 17:09:20 <KipIngram   > I think so too (antennas).  I've studied it and I know it's all mathematically sound, but it's still just amazing.
2023-11-02 17:10:01 <Zarutian_iPad> I just know certain rules of thumb regarding dipole, yagni and such
2023-11-02 17:10:11 <KipIngram   > The other really cool bit on that front is that in quantum mechanics it turns out that the wave function has a "phase."  If you shift all of the phases in a problem by the same amount, it doesn't change any predictions.
2023-11-02 17:10:31 <KipIngram   > It's what they call a "global symmetry."
2023-11-02 17:10:52 <KipIngram   > Well, you can use gauge theory methods to promote that to a local symmetry, where you change the phase by different amounts at different places.
2023-11-02 17:11:07 <KipIngram   > That will change the predictions, though, unless you add some new terms to the physics to compensate for it.
2023-11-02 17:11:17 <KipIngram   > It turns out that those new terms are precisely electromagnetism.
2023-11-02 17:11:44 <KipIngram   > Gauge theory feels almost like cheating to me - it's like getting new physics for free.
2023-11-02 17:11:59 <KipIngram   > but more often than  not when you do that trick you wind up finding something that turns out to be real.
2023-11-02 17:14:06 <KipIngram   > It's pretty much where we got the theories of the strong and weak forces from too.
2023-11-02 17:14:21 <KipIngram   > And you CAN get EM that way, even though it wasn't our original path to it.
2023-11-02 17:14:31 <KipIngram   > So the whole Standard Model is gate theory.
2023-11-02 17:16:38 <KipIngram   > Ugh.
2023-11-02 17:16:40 <KipIngram   > gauge
2023-11-02 17:17:10 <KipIngram   > Pesky old gravity still stubbornly refuses to fit in, though. :-)
2023-11-02 17:17:26 <Zarutian_iPad> here is thing
2023-11-02 17:18:06 <Zarutian_iPad> gravity does not seem to be a force like electromagnetism and such are
2023-11-02 17:18:48 <KipIngram   > Right - it's not clear that it's SUPPOSEd to go in there alongside the others.
2023-11-02 17:18:59 <Zarutian_iPad> many moons ago I described my personal hypohypothesis on it
2023-11-02 17:19:03 <KipIngram   > Lot of folks want it to, but... nature may just not care.
2023-11-02 17:19:29 <KipIngram   > It's not like we really NEED it to, to get practical things done.
2023-11-02 17:19:47 <KipIngram   > We can use general relativity to solve for the spacetime curvature, and then run the Standard Model on that background.
2023-11-02 17:19:50 <KipIngram   > And that works great.
2023-11-02 17:20:18 <KipIngram   > Unless it's less than 10^-42 seconds after the Big Bang (which it's not) or you need to be right AT the singularity of a black hole (which you don't).
2023-11-02 17:20:36 <Zarutian_iPad> and you can bend spacetime using/moving energy at relistvic speeds
2023-11-02 17:20:47 <KipIngram   > outside of those corners, our "mashup" of GR and standard model gets the job done.
2023-11-02 17:21:16 <KipIngram   > Or just by gathering a lot of energy into one place.
2023-11-02 17:21:20 <KipIngram   > Which is exactly what mass is.
2023-11-02 17:23:26 <Zarutian_iPad> 
ACTION is now wondering what happens along the path that kugelblitz blackhole is shot along at good fraction of lightspeed

2023-11-02 17:23:39 <KipIngram   > :-)
2023-11-02 17:23:50 <KipIngram   > I suppose general relativity would let you solve for that.
2023-11-02 17:23:59 <KipIngram   > I know weird things happen when they're spinning fast enough.
2023-11-02 17:24:41 <Zarutian_iPad> hmm... might the kugel cause frame dragging along the path?
2023-11-02 17:29:33 <KipIngram   > I assume it would.  Pretty early on Einstein showed that a rotating cylinder of sufficient math would drag around the stuff inside.  It excited him hugely because he regarded it as supportive of Ernst Mach's ideas.  He fired off an elated letter to Mach.
2023-11-02 17:31:29 <Zarutian_iPad> so, for a scifi story it might be part of near ftl or even ftl drive. A bit insane one where you could sustain the kugel with a maser or such
2023-11-02 17:34:47 <KipIngram   > An idea that struck me once was that even for lower speed flight, if you could accelerate your rocket "ejecta" super close to light speed, you get a Lorentz boost to the momentum it buys you.  And you can just keep making that factor bigger by getting closer and closer.
2023-11-02 17:35:00 <KipIngram   > The upshot would be that you could carry a tiny amount of fuel.
2023-11-02 17:35:15 <KipIngram   > I've proposed that as a way the "Epstein drive" in The Expanse might work.
2023-11-02 17:35:37 <KipIngram   > They didn't tool around at light speed, but they also never seemed to have any real problem carrying around propellant for long voyages.
2023-11-02 17:35:56 <KipIngram   > They could maintain acceleration for half the trip, flip over and decelerate for the other half, which got them on-board gravity.
2023-11-02 17:36:06 <KipIngram   > That would be ENOUGH to give us the solar system.
2023-11-02 17:37:11 <KipIngram   > Of course I have zero ideas for how to get propellant to those speeds in any practical way.
2023-11-02 17:37:30 <Zarutian_iPad> I think cyclers and lightsails are more realistic
2023-11-02 17:37:38 <KipIngram   > I love cyclers.
2023-11-02 17:37:55 <KipIngram   > I love the idea of a big luxury hotel in a Mars / Earth cycler orbit.
2023-11-02 17:38:04 <KipIngram   > With all the radiation shielding you could ask for, etc.
2023-11-02 17:38:27 <KipIngram   > I like sails too, but somehow cyclers have always kind of infatuated me.
2023-11-02 17:38:37 <Zarutian_iPad> plus there is beamed microwave propulsion available too
2023-11-02 17:39:24 <KipIngram   > Yeah, using EM radiation as the propellant would solve the propellant mass problem.  If only we could get practical thrust levels.
2023-11-02 17:39:39 <thrig       > the engines in The Expanse always did seem to move at the speed of plot
2023-11-02 17:39:44 <Zarutian_iPad> use the microwave beam to heat the propellant to high tempertures
2023-11-02 17:39:46 <KipIngram   > Yes, they did.
2023-11-02 17:40:23 <KipIngram   > I really, really enjoyed the first six Expanse books.
2023-11-02 17:40:36 <KipIngram   > The I cracked open #7 and suddenly 30 years had passed.
2023-11-02 17:40:55 <Zarutian_iPad> but I think the Epstein drive was the only plothole the author allowed himself/herself/themselfs
2023-11-02 17:41:17 <KipIngram   > I was pissed - I felt a lot like I would have if real friends of mine (close ones) had suddenly just not been around for 30 years and then just showed back up.
2023-11-02 17:41:41 <KipIngram   > There were loose plot thread left hanging.
2023-11-02 17:42:25 <KipIngram   > And everyone had old person aches and pains.
2023-11-02 17:42:34 <KipIngram   > I have those of my own - I don't need to read about them.
2023-11-02 17:43:03 <KipIngram   > I wanted to read more about my brash youngster friends charging around righting wrongs in the solar system in their state of the art bad-ass ship.
2023-11-02 17:43:28 <thrig       > ... named after the nag in Don Quixote
2023-11-02 17:43:42 <KipIngram   > That said, pretty much the best "space battle" scene I've ever read is in one of those first six books.
2023-11-02 17:43:51 <Zarutian_iPad> would you have liked an in between series that solved those loose threads and described the adventures that were alluded to in book 7 onwards?
2023-11-02 17:44:10 <KipIngram   > Yes.
2023-11-02 17:44:29 <KipIngram   > But really I wanted them to stay young.
2023-11-02 17:47:54 <Zarutian_iPad> 
ACTION is readding about TwitterX spat and has the vice of reading the comments when he shouldnt.

2023-11-02 17:49:49 <Zarutian_iPad> though that vice is better than smoking I suppose.
2023-11-02 17:53:45 <Zarutian_iPad> back to programming topics
2023-11-02 17:54:13 <Zarutian_iPad> I been using jump tables a bit in one project
2023-11-02 17:55:30 <Zarutian_iPad> one thing I did in implementation is that is the index given into is higher than the table length then execution continues after the jump table
2023-11-02 17:57:25 <Zarutian_iPad> one question I am mulling over is if I need an variant that has an 'offset' and if the given index is below or equiv to that offset the same occurs
2023-11-02 17:57:57 <Zarutian_iPad> btw when the jump table is, er, jumped over the index remains on stack.
2023-11-02 23:31:31 <KipIngram   > There's an interesting math puzzle.  You have three indistinguishable dice.  You can only roll them as a group.  They're standard fair 6-sided dice.
2023-11-02 23:31:49 <KipIngram   > How can you simulate rolling a single die with this set of dice?
2023-11-02 23:38:39 <KipIngram   > Then the bonus problem - how do you simulate rolling two six-sided dice with the 3-set?
2023-11-02 23:38:52 <KipIngram   > All the outcome probabilities have to be correct, of course.