--- Log opened Mon Jul 10 00:00:55 2023
00:24 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
00:25 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
00:39 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
00:43 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
00:44 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 250 seconds]
00:44 -!- stipa_ is now known as stipa
00:49 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
00:51 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 246 seconds]
00:52 -!- stipa [~stipa@user/stipa] has joined #hplusroadmap
00:55 -!- stipa_ [~stipa@user/stipa] has quit [Ping timeout: 245 seconds]
00:59 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
01:01 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 252 seconds]
01:01 -!- stipa_ is now known as stipa
01:08 -!- catalase [catalase@freebnc.bnc4you.xyz] has quit [Quit: Stable ZNC by #bnc4you]
01:18 -!- TMA [tma@twin.jikos.cz] has quit [Ping timeout: 250 seconds]
01:18 -!- TMA [tma@twin.jikos.cz] has joined #hplusroadmap
01:27 -!- catalase [catalase@freebnc.bnc4you.xyz] has joined #hplusroadmap
02:18 < fenn> the real question is whether catgirls will have one or two sets of ears
03:18 < muurkha> stipa: fix your connection
03:37 < nsh> connection's fine. stuck in revolving doors 
03:55 < fenn> https://old.reddit.com/r/aiArt/comments/14ln012/for_5_years_a_small_town_in_arkansas_has_hosted/
04:04 < nsh> 'There's actually a huge controversy in the scene right now after a genetic engineer managed to re-activate dinosaur genetics in a newborn lizard. Authorities are keeping a close watch and may have to put it down as it's growing in size with no sign of stopping anytime soon, and is currently reaching 2 meters in height. The whole community is really split about it. This isn't even his craziest project though - he is famously known for splicing parrot DNA for 
04:04 < nsh> wings, and recently has been making huge progress on a genetically engineered lactobacillus that can output diesel, fully built-in as a symbiotic colony into a custom engineered mouth pocket.'
04:04 < nsh> surprise kanzure isn't in touch
04:04 < nsh> *surprised
04:06 < fenn> they call it dino juice for a reason
04:20 -!- _flood [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
04:44 < hprmbridge> kanzure> hm?
04:49 < fenn> just some fae news
05:46 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0:ac18:c9c1:cb60:a224] has joined #hplusroadmap
06:04 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
06:05 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 246 seconds]
06:05 -!- stipa_ is now known as stipa
08:55 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
08:56 < stipa> muurkha: pay and i'll fix it
08:57 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
08:59 -!- _flood [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 264 seconds]
09:00 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 240 seconds]
09:03 < docl> stipa: how much
09:04 < stipa> Eur 50 per month should be enough 
09:04 < stipa> for fibre
09:05 < stipa> now is 4G
09:05 < stipa> it+s probably cells overheating
09:05 < stipa> it's 38°C overhere
09:06 < docl> hmm. I pay $4/mo for a ramnode instance for this
09:07 < stipa> that was pop over 15 years ago
09:08 < stipa> when ppl had 56k modems
09:08 < docl> ubuntu + tmux + irssi
09:08 < stipa> yeah, same here, manjaro + tmux + weechat via 4G
09:08 < stipa> no
09:08 < stipa> debian actually
09:09 < stipa> i have feeling that ppl with psybnc have trouble with nick highlighting and that kind of stuff
09:12 < stipa> i've got irssi on a BSD machine connected to OFTC
09:12 < stipa> als ovia tmux
09:12 < stipa> in a VM
09:13 < stipa> weechat has all the bells and whistles it takes ages to set up on irssi
09:15 < docl> I won a 56k modem in a sweepstakes from a shareware company called happy puppy once. it was a huge upgrade from the 14.4k modem my dad bought. so that's about how old I am. DSL existed but not where I lived
09:16 < docl> maybe I should try weechat and see if I like it better than irssi
09:18 < stipa> it's ok, set and forget type of thing
09:18 < stipa> works ok in tmux, it's mostly power outage on a server then tmux or weechat crash in my case
09:22 < docl> anyway irc is something that makes sense to do from ze clowd IMO, you don't have to bother people with disconnects constantly (the tmux session is running on the remote server, which is ~100% uptime because it's on a rack somewhere)
09:22 < stipa> DSL was super fancy thing when it came in my environment
09:22 < stipa> night and day compared to the 56k
09:22 < stipa> stable speed
09:23 < docl> for sure
09:23 < stipa> yeah, i like the feeling of having my own server
09:23 < stipa> it was my dream when i was a kid
09:24 < stipa> always connected machine
09:24 < stipa> i'm not responsible of bad ISP service
09:25 < stipa> that's not in my power to fix
09:25 < stipa> shitty ISP modems and such things...
09:26 < stipa> i do my best to do my part right, for everything else ppl will have to cope with
09:28 < docl> well ramnode is $4/mo. there are probably better options out there for ultra low scale. but it isn't the dream, I hear ya
09:29 < stipa> it's easy to pay for things but that also makes you dumb
09:29 < stipa> and without money
09:30 < stipa> but, i don't know what are your interests, i guess chemistry?
09:31 < stipa> i'm into electronics so i like it's technological side, paying someone else for IRC bouncer is just not acceptable
09:32 < stipa> if you're just into communicating with people via IRC then ramnode has sense
09:32 < docl> I'm a wide ranging interests guy, chemistry is recently acquired but computer science and electronics/hardware are interesting too. fun thing about irc is that everyone else is pretty into CS stuff too
09:34 < stipa> CS?
09:34 < docl> computer science
09:35 < stipa> computer scientists are on twitter and discogs
09:35 < stipa> IRC is just a bunch of Jacks
09:35 < docl> yeah but they're mostly jacks with strong CS. discord is gamers and twitter is literally everyone
09:36 < stipa> https://en.wikipedia.org/wiki/Jack_of_all_trades,_master_of_none
09:36 < docl> we have masters here too
09:37 < stipa> linkedin is full of CS dudes, those ppl have no idea what they're talking about
09:37 < docl> those are people optimizing for career
09:37 < stipa> there was a dude, like nice picture, authoritative and such stuff, the guy you don't want to mess with...
09:37 < stipa> he was talking about cloud tech
09:38 < stipa> man, he was speaking of i don't know what he even spoke about
09:38 < stipa> i said to him, cloud is just a bunch of computers running FTP servers renamed to cloud
09:39 < stipa> they think it's some kind of some voodoo hardware and stuff
09:39 < stipa> and they can explain it clearly
09:40 < stipa> at least they can but i have no clue what that is
09:40 < stipa> some even argued with me
09:40 < stipa> fuck, you know, they aren't even aware it's just a computer running software that act as servers
09:41 < stipa> that is a CS nowadays
09:42 < stipa> probably some dumb middlemen that have no clue what they're selling
09:42 < stipa> maybe they did some random CISCO course and think they're CS
09:43 < docl> that's business people trying to do CS... real CS is knowing how to make hardware do stuff. software lets you bridge from abstraction to reality. it's pretty neat.
09:44 < docl> there's a whole academic world of people going hyper abstract and then designing software to make a machine do that abstract thing as a real computation
09:46 < stipa> yeah
09:46 < stipa> all that power and nothing useful
09:47 < stipa> the abstraction trend started as a mean to bring chores of low level to the dumb masses
09:48 < stipa> what we have today is "abstraction" that practically doesn't work and waste of energy while it runs
09:49 < stipa> and you have CS experts that don't know how to turn on a computer selling that abstraction to ppl that have to rely with their lives on
09:50 < stipa> ppl have to eat 
09:50 < stipa> some sell useless crap to survive and others suffer because of that
09:50 < docl> I don't agree that it's nothing useful, but I do agree that there's a lot of waste involved. people are looking for ways to spend joules instead of human-hours. I can't blame them, exactly, but makes me uncomfortable at times
09:52 < docl> a lot of people are probably going to get really into prompting AI without learning how it translates to reality very well
09:53 < stipa> i haven't seen anything useful of AI trend, i see just a bunch of machines capable of doing insane parallel operations at once, but there'll for sure be a talent or two who'll make something useful out of all that power
09:54 < stipa> i hope so
09:55 < docl> I think learning chemistry excited me so much because it gave some insights to pierce the veil between "plug these components together" vs "make atoms and electrons do things". people who mfg chips are turning chemistry into computing hardware, and there's something fascinating about that
09:56 < stipa> chip designers don't care about the chemistry
09:56 < stipa> they build structures
09:57 < docl> structures in the abstract? no, they have to figure out how to make actual atoms, silicon and doping elements, fit the structure. there's a distinction between a geometric shape in the abstract "a circle" vs an object with that property in physical reality
09:57 < stipa> https://www.youtube.com/watch?v=69mdJv6fWXE
09:58 < stipa> ^^ Explained How Chips Are Designed
09:58 < stipa> there's no chemistry, fabs give rules that has to be followed and that's it, depends on the manufacturing process
09:58 < nsh> pretty sure it's not independent of chemistry
09:58 < docl> http://elementsofprogramming.com/eop.pdf chapter 1 discusses abstract vs concrete things, matched my prior intuitions in an interesting way. I haven't quite fully wrapped my head around the concept of "regular" they are trying to teach though.
10:00 < stipa> chemistry part would i think be in the fab a thing
10:00 < stipa> maybe not even there
10:00 < stipa> probably fab buys the equipment and that's it
10:00 < stipa> having no clue what's going on in the machines
10:01 < stipa> secret upon a secret upon a secret....
10:03 < stipa> especially with the newest nodes
10:06 < stipa> docl: thanks for the book
10:06 < docl> I guess regular = you can substitute it losslessly. so in a chip context they are looking at an electrochemical process (charge affecting conductivity -> precisely substitutable with the logic operation of a gate closing) and once you know that process will work you can rely on it and focus on just the logic. but of course stuff like "how many times do I need to perform this logical operation" matters 
10:06 < docl> because you're physically closing a circuit each time, which consumes time and energy.
10:07 < stipa> and it has to have fast turn on and off times
10:08 < stipa> the faster the turn on the faster the chip is the more energy it needs to turn on fast
10:08 < stipa> so, you have a trend of a CPU sucking a kW or so
10:08 < stipa> it's nothing strange today
10:09 < stipa> you need a phase from the electricity grid just for a PC nowadays
10:11 < docl> then because you're implementing it as circuits that must spend energy and time to open and close, even if the cost is *very* small, certain things that work perfectly fine in pure math land like "if I do this infinity times" are unimplementable. so as computer scientists we're focusing on the subset of things you can do when you have that kind of limit on what you can do. so like you might be able to 
10:11 < docl> search by brute force, but choosing an algorithm that does it faster at cost of storing more memory will just make more sense / get it done faster / be able to work for bigger problems.
10:15 < stipa> the idea was if we go smaller the energy consumption will as well but that didn't happen, by going smaller the frequency has risen and for fast switch on you need a lot of energy 
10:23 < stipa> gates on transistors are practically capacitors
10:25 < stipa> if you want to charge a cap faster you need more current 
10:29 < nsh> or you do it on a vehicle making a round trip involving velocities relativistic to your frame of reference
10:31 < docl> yeah, that's an intricacy worth understanding. if you can parallelize an algorithm / split it into many separate processes, you can go lower energy and do many slower things at once. so it's not just about trading "number of ops needed" vs "memory" in that context. and different kinds of tasks can be easier or harder to parallelize
10:32 < docl> btw some people here are thinking about how to make rod logic computers
10:33 < nsh> In Rod We Trust
10:34 < nsh> hmmm, up muurkha's street: https://anthony-zhang.me/blog/rod-logic/
10:34 < nsh> i walked past the place where the Zuse Z's were built when i was in berlin recently
10:34 < nsh> or lived or something idk the plaque was in some gobblegook language
10:35 < docl> well, it's one of the interesting uses for nanotech (not that it has a shortage of interesting uses)
10:35 < nsh> if you're gonna make things smol then i'm not sure mechanical actuation makes more sense than electron states
10:35 < nsh> it's kinda nice at our scale
10:36 < nsh> and make down to cells
10:36 < nsh> (then again, i know nothing)
10:44 < docl> erik drexler had some math on this. logic gates that switch in 0.1 ns and dissipate 10^-21 J. 10^16 instructions per second per watt
10:44 < nsh> classical data is educated stupid. operations are cubic
11:01 < stipa> light could be the near future,  i think there are already computer mobos with fibre interconnecting systems  
11:02 < stipa> light based processors also work efficiently 
11:02 < stipa> so like holographic memory and that things
11:03 < stipa> it's like powering a laser or an LED diode and that should be enough or smthn absurd like that, practically a few mA
11:05 < stipa> https://en.wikipedia.org/wiki/Optical_computing
11:06 < stipa> i think i saw optics flying around in this channel
11:07 < stipa> See How Lightelligence Created the World’s First Working Optical Computing System https://www.youtube.com/watch?v=a2merSqeVeI
11:12 < stipa> optoelectronics
11:16 < stipa> 100 x faster then the GPU
11:21 < superkuh> An exobyte of bandwidth?
11:21 < superkuh> /s
11:23 < hprmbridge> gourneau> https://www.nature.com/articles/s41467-023-38876-w
11:23 < kanzure> .title
11:23 < EmmyNoether> A biological camera that captures and stores images directly into DNA | Nature Communications
11:27 < docl> I wonder if that's anywhere near the theoretical limit for optics? current hash power to mine a bitcoin is around 10^18 ops. at the landaur limit, which is what drexler was apparently referencing, this would cost about a millijoule
11:44 < docl> https://en.wikipedia.org/wiki/Landauer%27s_principle
11:45 < muurkha> haha
12:04 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
12:07 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 245 seconds]
12:10 < stipa> docl: i would guess 100x faster is just an super shitty early prototype 
12:11 < stipa> light is the fastest thing in the universe unless quantum maybe
12:11 < docl> yeah that would be my guess
12:12 < stipa> quantum would be like, let's skip all the computation and just get the value from the future 
12:12 < docl> I'm not super confident about the landaur limit (I just know it's something a lot of smart people are willing to appear confident about, so maybe I should be)
12:13 < docl> reversible computing is supposed to be a workaround for the landaur limit though, somehow you don't have to erase it if you undo it
12:15 < stipa> that seem like it could be applied to electronics
12:17 < stipa> but low heat is also slow switching
12:17 < stipa> now does it apply to photonics too i wouldn't know
12:18 < stipa> maybe you also need a bunch of light in the photonic processor
12:18 < docl> well drexler says 0.1ns so maybe 10 GHz rod logic does approach the landaur limit. I wonder if you can overclock that in turn for faster rates
12:18 < stipa> to do complex operations
12:20 < stipa> yeah, 10GHz is super fast
12:21 -!- gptpaste [~x@yoke.ch0wn.org] has joined #hplusroadmap
12:21 < stipa> i9-13900KS Able to deliver up to 6.0 gigahertz (GHz) 
12:22 < stipa>  Intel is making the Core i9-13900KS on a 10 nm production node
12:22 < nsh> .gpt4 speculate on three or four ways we might surpass the information theoretic thermodynamic limit of Landauer's principle using novel forms of computation
12:23 < nsh> "As an AI language model, i cannae break the laws of physics, cap'n"
12:23 < stipa> so going smaller a bit 10Ghz will probably be reached
12:23 < gptpaste> ​1. Quantum Computing: Quantum computing is a promising field that could potentially surpass the limits of Landauer's principle. Unlike classical computers, which use bits as their smallest unit of data (either a 0 or a 1), quantum computers use quantum bits, or qubits, which can represent both 0 and 1 simultaneously thanks to the principle of superposition. This allows quantum  - http://sprunge.us/9wPMu2
12:23 < nsh> sweet beautiful nonsense
12:24 < nsh> 3. Optical Computing: Another potential method for surpassing Landauer's limit might be optical computing or photonic computing where photons are used instead of electrons for processing information. Photons move faster than electrons and do not produce heat due to resistance as they don't have charge unlike electrons.'
12:24 < nsh> ah yes the notorious properties of photons that are emitted when wavefunctions change without any energetic coupling 
12:24 < stipa> yeah, organic electronics is a bit much but promising
12:25 < stipa> that's probably cyborg stuff already
12:26 < stipa> photonics is the nearest
12:27 < stipa> if you look at humans as machines organic electronics suck a bunch of power
12:28 < stipa> via feeding
12:29 < stipa> nature probably did it as efficient as possible
12:29 < nsh> you can do quite a lot of thinking on a sandwich or two
12:29 < nsh> '3. Optical Computing: Another potential method for surpassing Landauer's limit might be optical computing or photonic computing where photons are used instead of electrons for processing information. Photons move faster than electrons and do not produce heat due to resistance as they don't have charge unlike electrons.'
12:29 < nsh> no
12:29 < nsh> 'Joseph Carlsmith estimates that the brain delivers roughly 11 petaFLOP/s (=1015 floating-point operations per second). If you eat a normal diet, you're expending roughly 10−13 J/FLOP.6 Sept 2022'
12:30 < nsh> oh dear i just cited lesswrong.org
12:30 < nsh> that's a paddlin'
12:30 < docl> you will be assimilated
12:31 < nsh> does floating-point even mean anything in the context of neurons
12:31 < nsh> as close to nothing as to be not worth making the distinction
12:31 < docl> hmm. it's not hard to store data in a light beam. just transmit it to a reflector at the distance in light-seconds away that corresponds to the time you want to read it out again
12:31 < nsh> oh no i drank some absinth and now my arithmetic coprocessor isn't working right
12:32 < nsh> ' Biological brains involve different principles than digital computers and these principles are not yet understood so that it is difficult to compare directly the computational capability of a human brain with a supercomputer. But some order-of-magnitude estimates (say based on the number of action potentials transmitted per second in an entire human brain between the approximately 1011 neurons that have about 1014 interconnections between each other) suggest 
12:32 < nsh> that human brains also carry out the equivalent of 100-1000 petaflops of information processing. '
12:33 < nsh> all of this nonsense was written back when people had very little appreciation of all the 'processing' that happens within neurons (rather than merely between)
12:34 < nsh> last quote from http://webhome.phy.duke.edu/~hsg/414/images/brain-vs-computer.html
12:34 < nsh> --
12:34 < nsh>  But consider the huge differences between the power and volume requirements of these two computing systems. A 100 petaflop supercomputer requires about 15,000,000 watts (enough power to support a city of about 10,000 homes), occupies an area of about an American football field of interconnected cabinets of CPUs, and requires a sophisticated and expensive cooling system to withdraw the large amount of heat produced. In contrast, your brain, even when solving a 
12:34 < nsh> difficult physics problem, consumes about 15 watts (the power to keep lit a rather dim light bulb) and has a volume of about two fists. (And of course, brains are more impressive than supercomputers in other ways in that they self-assemble from a single cell, and they are self-learning entities that can master physics, math, language, art, music, and sports without being explicitly programmed.) 
12:34 < nsh> -- (explicitly programming by a few billion years of evolution in a highly complex environment)
12:34 < nsh> *programmed
12:35 < nsh> the last paragraph is fun though
12:42 < docl> lol yeah, now I'm kind of picturing a superintelligence that works mainly by repeatedly producing a giant 3d grid of short lived billion-kelvin black holes and evaporating them to extract the data
12:49 < muurkha> with respect to https://anthony-zhang.me/blog/rod-logic/ I think Zhang has a few things to learn about fanout
12:50 < docl> but like 15 W should translate to... drexler says 10^9 ops per 10^-21 W -> 10^30 ops per watt at the landaur limit. so uh peta means 10^15 so a human at 1000 peta(fl?)ops is uh roughly one "bitcoin worth" per second? anyway 10^30 is a 10^12 bigger than that so... 1 trillion humans? we're 1/trillionth as efficient as we could be. if you trust 1000 peta-ops as a ballpark anyway
12:51 < nsh> one bitcoin worth is a pretty moving target
12:51 < nsh> usually up
12:51 < docl> oh yes. just saying.
12:54 < docl> actually that was calculated per watt so 15x as much... well, we spend way more watts on a meatbody than that, so it's really a lot more. anyway, meat brains look impressive vs silicon but not theoretical landaur limit pushing mechanisms
12:56 < nsh> we only end up in metabags because we fail to recognise our true nature by dying unmindfully
12:56 < nsh> ( https://www.britannica.com/topic/Bardo-Thodol )
12:57 < nsh> .t https://www.youtube.com/watch?v=1TV11z9CxF4
12:57 < EmmyNoether> The Tibetan Book of the Dead - Bardo Reading in English - YouTube
12:59 < docl> so the window of opportunity to make bio computers a thing is finite. maybe 1k flops per 15 watt could be replicated and made digitally programmable, and if so yes you could beat the current (1 exa-op) btc network that way in energy costs, but it's not like this is pushing theoretical limits on energy->compute
13:12 < nsh> instead of computing more, one might (also) consider computing better
13:12 < nsh> we've been computing exponential more for decades now and the general experience has tended to degrade
13:13 < nsh> (at least in proportion to resource use)
13:13 < nsh> that are it's becoming clear that pumping entropy into a finite ecosystem is not necessarily the cleverest most we've made as a species
13:14 < nsh> *move
13:16 < docl> well computers are more general than brains. I may have a 10^18 "flop" computer in my skull but I suspect I won't have much luck training it to solve btc hashes
13:19 < stipa> NPUs should emulate neurons
13:20 < stipa> there is a neuron map of larva fly
13:21 < stipa> a genius or two could make it work with NPU for sure
13:21 < stipa> in the case everything is true about neural nets the NPU should emulate a larva brain
13:22 < stipa> same with a human brain which is hard to map due it's complexity
13:23 < stipa> if mapping and all that crap is scientifically viable the NPU that could run a human brain could think it's an actual human
13:24 < stipa> with some learning, or in case some random map already has an experience of example farming it could just be copied into another NPU
13:24  * L29Ah throws geniuses at stipa
13:26 < stipa> i mean, just a larva brain simulation would be insane
13:26 < stipa> existing in some random virtual world
13:27 < stipa> flapping it's wings and reacting to the environment 
13:28 < stipa> same with human maps, poor chips thinking they're real
13:30  * L29Ah throws scifi literature at stipa
13:33  * stipa hands a blue pill to L29Ah
13:35 < juri_> this is the future of humanity, people. right here, right now.
13:37 < juri_> now if you'll excuse me, i'll go back to using todays tech to produce tomorrows algorithms, for yesterday's problems.
13:37  * stipa hands a red pill to juri_
13:45 < nsh> you'd think morpheus would have like a pez dispenser or something
13:46 < nsh> i mean has he just got those things loose in his trenchcoat or what?
13:55 < docl> I've been reading about hermann grassmann (19th century projective geometric algebra guy), interesting stuff
14:21 < alethkit> Was it projective GA that he worked on, or just GA in general?
14:28 < juri_> I think he was just GA, but honestly, i don't pay much attention to the history parts of the lessons.
14:30 < nsh> i don't think Grassmann was must aware of their history when he was working on them either, tbf
14:42 < docl> https://core.ac.uk/download/pdf/231908059.pdf is what I was reading. his approach seems projective to me the way they describe it, but I could be wrong
14:55 < nsh> it's not but people have made projective grassmann (or in current vogue 'geometric') algebras. e.g. https://en.wikipedia.org/wiki/Grassmann%E2%80%93Cayley_algebra / https://projectivegeometricalgebra.org/
14:56 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
14:56 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
14:57 < nsh> the axioms of projective geometry by Whitehead was published in 1923 by which time Grassmann have been dead for almost fifty years
14:57 < nsh> actually 1906 
14:57 < nsh> so thirty
14:58 < nsh> but it was building on work from the 19th century i guess
15:00 < nsh> 'However, there were some conceptual connections between Grassmann's ideas and projective geometry. Grassmann's theory of extension included the notion of a "meet" and "join" operation, which had similarities to the projective concepts of intersection and union. Grassmann's work also had implications for the understanding of duality in projective spaces.'
15:00 < nsh> (suggests entropymonster)
15:02 < juri_> I have my own projective geometric operators. with blackjack.
15:03 < docl> https://en.wikipedia.org/wiki/Whitehead%27s_point-free_geometry
15:05 < docl> talk about going hyper-abstract
15:05 < docl> > Point-free geometry was first formulated in Whitehead (1919, 1920), not as a theory of geometry or of spacetime, but of "events" and of an "extension relation" between events. 
15:53 < muurkha> nsh: maybe you can help juri_ with her Clifford algebra code; she's written a CAD system in it
16:00 < nsh> dunno, don't like thinking or trying hard 
16:01 < nsh> we have other people for that these days 
16:02 < nsh> (i jest, poorly) is this system described anywhere?
16:02 < Hooloovoo> hmm, something didn't like that wp title
16:03 < Hooloovoo> https://en.wikipedia.org/wiki/Whitehead%27s_point-free_geometry does that work?
16:05 < nsh> .gpt4 could whitehead's point-free geometry be related to interval arithmetic?
16:06 < gptpaste> ​Whitehead's point-free geometry, also known as mereotopology, is a theory of space that is based on the concepts of part-whole and boundary rather than points. It focuses on regions and their relationships.Interval arithmetic, on the other hand, is a mathematical method that uses intervals or ranges of values instead of fixed numbers. This allows for more accurate representat - http://sprunge.us/IFyWat
16:06 < nsh> oh yes, one's qualitative and the other's quantitative. you're truly a genius. now where's that lamp
16:11 < nsh> "This paper advocates some very heretical views about the foundations of real and interval analysis. You can either burn me at the stake for blasphemy, or open your mind to the very simple and natural arguments that I give here." - http://www.paultaylor.eu/ASD/intawi.pdf
16:12 < nsh> god save all cranks and misfits 
16:13 < nsh> another personal favourite
16:13 < nsh> .wik Pointless topology
16:13 < EmmyNoether> "In mathematics, pointless topology, also called point-free topology (or pointfree topology) and locale theory, is an approach to topology that avoids mentioning points, and in which the lattices of open sets are the primitive notions." - https://en.wikipedia.org/wiki/Pointless_topology
16:14 < nsh> oscar wilde would have appreciated it, if he wasn't so busy buggering pretty boys and being insulting curtains
16:14 < nsh> ( https://janwriter.medium.com/oscar-wilde-on-art-75c05aeeb9b )
16:15 < nsh> -being
16:18 < nsh> 'A celebrated reviewer once described a certain paper (in a phrase which never actually saw publication in Mathematical Reviews) as being concerned with the study of "valueless measures on pointless spaces"' - https://projecteuclid.org/journals/bulletin-of-the-american-mathematical-society-new-series/volume-8/issue-1/The-point-of-pointless-topology/bams/1183550014.full
16:29 < docl> pointless, but is it also useless?
16:39 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 245 seconds]
16:47 < docl> why is the "join" symbol ginormous and have stuff underneath it?
16:55 < nsh> where?
16:55 < docl> the font enlargening convention has been bugging me for a long time, usually I see it when summation is represented with capital sigma. surely there's a reason for it
16:55 < docl> on the pointless topology wiki page
16:59 < nsh> indeed it's the same convention
17:00 < nsh> ranging over i's belonging to I 
17:00 < docl> more common examples described here (but nothing about why -- presumably it makes sense in a historical context) https://mathmaine.com/2010/04/01/sigma-and-pi-notation/
17:01 < nsh> a product a sum and a union are all ways of considering some elements collectively 
17:01 < nsh> likewise a join
17:02 < nsh> but the join is the one that looks like a V, the upside down one is the meet
17:02 < nsh> (akin to intersection)
17:02 < docl> ok that makes sense to me now
17:03 < nsh> you could translate the whole thing to a logical expression 
17:04 < nsh> (being the expression of a distributive law)
17:07 < docl> https://en.wikipedia.org/wiki/Iterated_binary_operation
17:18 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
17:19 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
18:01 < muurkha> I always forget which one is meet and which one is join
18:01 < muurkha> I think it's useful to think of meet and join as generalizations of minimum and maximum, or intersection and union
18:22 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
18:26 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 246 seconds]
18:56 < docl> nice mneumonic. I assume the downward V is minimum?
19:10 < muurkha> it's not a mnemonic
19:11 < muurkha> ∨ is maximum, as it happens, and ∧ is minimum
19:12 < muurkha> I mean, in Boolean algebra, with the usual assignments of numbers to truth-values rather than the Unix shell or Urbit assignment
19:12 < docl> oh, so it's the opposite... I guess you can just remember it as being backwards :/
19:13 < docl> uh so what precisely do maxima and minima have to do with meet and join? I'm a little hazy on this
19:16 < muurkha> I always have to look it up, but apparently join is supremum ∨ and meet is infimum ∧
19:17 < docl> well wiki says V = join and turned-V is meet
19:18 < docl> https://en.wikipedia.org/wiki/Pointless_topology this page is where I'm getting that from
19:19 < docl> I have a new akrasia hack I'm trying out: get super angry at the author of the confusing material and turn it into an obsession
19:27 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0:ac18:c9c1:cb60:a224] has quit [Quit: Leaving]
19:42 < docl> I don't think it's works as the normal kind of anger which implies disrespect. it would be hard to genuinely disrespect euler (my current target). but it's easy enough to form a kind of faux outrage where I'm basically simulating a character that hates his guts. I wonder if people do this a lot and it's why nerds come across as bitter a lot of the time
19:51 < muurkha> docl: that's what I said, yes
20:18 < muurkha> I don't know how much this faux outrage happens but I certainly feel frustrated when I'm struggling to understand something, and that is kind of like being angry
20:51 < docl> yes, same here. I usually call it frustration rather than anger in my mental narrative. I guess I'm sort of questioning whether that's productive or not. because frustration seems to lead me to bounce off of things I'd rather not bounce off of, which is frustrating in and of itself
20:51 < muurkha> heh
21:48 < docl> interestingly, uneven-tempered personality trait correlates to reduced cognitive ability in various ways https://www.pnas.org/doi/10.1073/pnas.2212794120
21:49 < docl> (not open access, but the supplemental pdf is fine)
21:51 < docl> not a surprising result, angry does seem like an impairment. I'm hypothesizing that one can leverage it by turning it into obsessive focus (not really the same thing)
21:52 < docl> I'm considering it as an antidote to scott alexander's "oops I lost the lottery of being obsessed with math, guess it's kind of like being gay" https://slatestarcodex.com/2013/06/30/the-lottery-of-fascinations/
21:54 < fenn> try lion's mane
21:54 < docl> because obsessions from anger seem like something people are really malleable, as opposed to like finding it intriguing
21:55 < fenn> i got obsessed with boho post-menopause yoga instructor vibes after taking lion's mane
21:55 < fenn> consider it a reroll
21:55 < docl> intriguing!
21:56 < fenn> previously i hated all that stuff, now i have just added it to the aesthetics i can appreciate
21:57 < fenn> it's worth considering if you want to actually BE a different person
21:58 < fenn> nerds come across as bitter because they recognize the injustice of the world, having been constantly exposed to it, but are powerless to actually do anything about it
21:59 < docl> that's what I assumed originally, but then it occurred to me they might be getting something useful out of it like obsession redirection
21:59 < hprmbridge> Eli> do you still take it?
22:00 < fenn> not regularly. i took it again in early 2022, trying to help a friend of mine by example
22:01 < muurkha> there are people who just make it a regular part of their diet, no?
22:01 < fenn> it's not unlikely that most people accumulate brain damage as they get older, and literally lose cognitive competencies. maybe we can reacquire those by creating new tissue to write on
22:02 < fenn> i had to eat salmon and sardines to get through the extreme fatigue. this is probably the growing neurons needing more DHA
22:02 < fenn> fatigue only lasted a day or two
22:02 < fenn> i don't know what would happen if you ate it every day
22:03 < muurkha> extreme fatigue from eating lion's mane mushrooms?
22:23 < fenn> yes
22:32 -!- ANACHRON [~Malvolio@idlerpg/player/Malvolio] has quit [Ping timeout: 250 seconds]
22:36 < muurkha> huh, interesting
--- Log closed Tue Jul 11 00:00:55 2023