--- Log opened Wed Jun 21 00:00:36 2023
00:42 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
01:01 -!- helleshin [~talinck@108-225-123-172.lightspeed.cntmoh.sbcglobal.net] has quit [Server closed connection]
01:02 -!- helleshin [~talinck@108-225-123-172.lightspeed.cntmoh.sbcglobal.net] has joined #hplusroadmap
02:47 -!- nefercheprure is now known as TMA
04:32 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0:c4bf:8b01:e4e5:75eb] has joined #hplusroadmap
04:58 -!- Llamamoe [~Llamamoe@46.204.72.21] has joined #hplusroadmap
05:27 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
05:30 < hprmbridge> kanzure> "A kilobyte rewritable atomic memory" https://arxiv.org/pdf/1604.02265.pdf
05:41 < hprmbridge> kanzure> more synthetic embryo drama https://twitter.com/MZG_Lab/status/1671240311098716166
05:53 -!- elinda [~elinda___@191.101.18.123] has joined #hplusroadmap
05:53 -!- elinda [~elinda___@191.101.18.123] has quit [Changing host]
05:53 -!- elinda [~elinda___@user/elinda] has joined #hplusroadmap
05:57 -!- cc0_ [~cc0@2a01:4f9:c010:cf0b::1] has quit [Remote host closed the connection]
05:58 -!- cc0 [~cc0@2a01:4f9:c010:cf0b::1] has joined #hplusroadmap
05:58 -!- cc0 [~cc0@2a01:4f9:c010:cf0b::1] has quit [Remote host closed the connection]
05:58 -!- cc0 [~cc0@2a01:4f9:c010:cf0b::1] has joined #hplusroadmap
06:12 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
06:32 < kanzure> zero-knowledge proofs for block validation in bitcoin https://blog.succinct.xyz/blog/btc-warp
07:37 -!- elinda [~elinda___@user/elinda] has quit [Ping timeout: 255 seconds]
07:40 -!- elinda [~elinda___@191.101.18.123] has joined #hplusroadmap
07:40 -!- elinda [~elinda___@191.101.18.123] has quit [Changing host]
07:40 -!- elinda [~elinda___@user/elinda] has joined #hplusroadmap
07:42 -!- test__ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
07:42 < fenn> "those who only lowered a weight saw the same improvements as those who raised and lowered weights -- despite only performing half the number of repetitions. many people may believe the lifting action provides the most benefit, or at least some benefit, but we found concentric muscle contractions contributed little to the training effects."
07:43 < fenn> https://www.sciencedaily.com/releases/2022/11/221103105002.htm
07:44 < fenn> hello why is this new research and not discovered in the 1800s
07:45 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 240 seconds]
07:52 -!- elinda [~elinda___@user/elinda] has quit [Ping timeout: 260 seconds]
08:10 < kanzure> were they testing it with some kind of machine (or second person) to do the lift?
08:24 < hprmbridge> nmz787> fenn it seems you'd have to be quite rich to only lower weights.. as in... "servant, now lift the weight for me, so as for me to once more lower it"
08:24 < cc0> > Using a dumbbell, Professor Nosaka recommends using two hands to help with the concentric (lifting weight) phase, before using one arm for the eccentric phase (lowering weight), when performing [...]
08:24 < hprmbridge> nmz787> I guess unless you were pulling down a lighter-than-air balloon
08:26 < fenn> it's not the lowering that's important, it's that the muscle is being extended rather than contracted
08:31 < cc0> i thought muscles only knew how to contract and we had a pair of muscles on each joint to contract in one way and contract in the opposite way (my understanding of biceps vs triceps)
08:31 < cc0> in which case i dont really understand how you can train a muscle when extending it because it would just be "inactive" during this movement
08:39 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
08:42 < fenn> you've never struggled to hold something up and couldn't do it and the object falls while you're still holding on?
08:43 < fenn> hardly seems like an "inactive" process
08:43 < fenn> vs just dropping the thing
08:45 -!- test__ is now known as _flood
08:47 < hprmbridge> nmz787> but wouldn't that be some muscle that wraps over the top of your shoulder contracting, to keep the arm held up?
08:48 < hprmbridge> nmz787> maybe the meaning is that even though the muscles are /trying/ to contract... you're not investing enough energy, and they're /slowly/ lengthening
08:48 < kanzure> in conclusion we know nothing
08:49 < kanzure> aren't the bodybuilding people supposed to know these things. what happened to examine.com?
08:49 < kanzure> https://examine.com/search/?q=eccentric
08:49 < hprmbridge> nmz787> I guess we need to invite Arnold into this chat
08:49 < hprmbridge> nmz787> he's done bodybuilding AND lots of sci-fi/hplus-ish movies... he's gotta be an expert
08:50 < hprmbridge> nmz787> I was just referencing when his eyes were popping out on Mars after his space helmet cracked at the end of Total Recall
08:50 < fenn> not realistic btw
08:50 < hprmbridge> nmz787> https://tenor.com/view/total-recall-arnold-gif-18913376
08:51 < hprmbridge> nmz787> hyperbole is useful though, at least for being memorable
08:53 < cc0> fenn: https://media.istockphoto.com/id/1225658587/vector/biceps-and-triceps.jpg?s=170667a&w=0&k=20&c=648NxTTzOxMAkaVANrlz90DR4zMK7cZgZeNz-aBsbsk=
08:53 < fenn> when lowering a barbell, the triceps would not be active
08:54 < cc0> thats how i picture it: youre struggling to hold the object up because the triceps is trying hard to contract 
08:54 < cc0> oh
08:54 < cc0> hmm yeah
08:54 < L29Ah> Total Recall is the best comedy action movie and very arnoldish, 10/10 do recommend
08:55 < cc0> oh okay yeah makes sense, it's trying hard to contract, but is being extended because tte object is too heavy
08:55 < cc0> hehe interesting
08:55 < fenn> but you can go to like 80% effort and it's still contracted but lowering the weight
08:55 < fenn> this way you can still be somewhat in control while doing the exercise
08:56 < L29Ah> 07:44 < fenn> hello why is this new research and not discovered in the 1800s
08:56 < L29Ah> it's well known that working against an overwhelming force is the most effective way of strength training
08:57 < L29Ah> but the most dangerous
08:59 < kanzure> has there been no progress in the past 10 years on molecular mechanisms of hypertrophy from weightlifting?
08:59 < kanzure> don't tell me "tears"
09:03 < kanzure> chatgpt says ECM remodeling to accommodate hypertrophy, mechanical load-induced nuclear accretion (myonuclear number go up effect), mechanotransduction, titin-mediated mechanosensing, and inflammation.
09:03 < L29Ah> myostatins?
09:03  * L29Ah doesn't recall anything new
09:05 < kanzure> also points to "low-load blood flow restriction training" as not causing muscle damage, another point against exercise-induced muscle damage theory
09:06 < fenn> yes, it's probably mediated by oxidative stress
09:11 < kanzure> "Downsizing the molecular spring of the giant protein titin reveals that skeletal muscle titin determines passive stiffness and drives longitudinal hypertrophy" https://elifesciences.org/articles/40532
09:11 < kanzure> maybe it's just titin length or the molecular spring properties of titin, both of which can be modulated
09:11 < kanzure> titin is the longest naturally produced protein so it would be interesting to see what would happen if the length was doubled or if the spring region was doubled
09:12 < fenn> 'if someone exercises while restricting blood flow, blood and metabolic byproducts become "stuck in the muscle, unable to leave."'
09:12 < kanzure> "To aid in this work, a mouse model was created in which titin’s stiffness was increased by shortening titin’s molecular spring region. We targeted for deletion 47 PEVK exons (Ttn exon 112–158), referred to as the TtnΔ112-158 model. .... titin might store elastic energy by unfolding Ig domains in passive muscle. Through their refolding during contraction, titin might generate forces ...
09:12 < kanzure> ...that add to the active force (Eckels et al., 2018). An alternative mechanism consists of titin-based effects on thick filament structure that contribute to activating the thick filament (Piazzesi et al., 2018; Fusi et al., 2016).
09:12 < kanzure> "
09:12 < kanzure> "Hypertrophy occurs in TtnΔ112-158 mice, due to longitudinal growth that serially adds sarcomeres. This shifts the in vivo working sarcomere length ranges of TtnΔ112-158 muscles to shorter lengths, supporting that titin-based stiffness is functionally important and that its level is carefully controlled"
09:13 < kanzure> "serial addition of sarcomeres"
09:15 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
09:24 -!- yorick [~yorick@user/yorick] has quit [Server closed connection]
09:24 < fenn> kaatsu training involves putting these bloodflow restricting bands on your upper arms or thighs, restricting blood flow to the limb, but this only really affects a single muscle group
09:25 -!- yorick [~yorick@user/yorick] has joined #hplusroadmap
09:33 < hprmbridge> nmz787> is there a way to connect chatGPT to the internet for more up-to-date info?
09:33 < hprmbridge> nmz787> I've asked it a few questions that stumped it lately
09:56 < fenn> with chatGPT plus you get access to plugins such as web browsing but then it will have to just click around manually like you would. bing chat has streamlined this somewhat, but it's very restricted
09:57 < fenn> for example bing refuses to write code when told to. it can do it but only if you ask nicely
09:58 < fenn> what you really want is a fine tuned model, but openAI is not doing that, probably for "safety" reasons
10:11 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has joined #hplusroadmap
10:12 < uzkruh> How many decades until total gender conversion is perfected?
10:13 < kanzure> what is your criteria
10:13 < fenn> make babby
10:13 < fenn> oh he said gender
10:14 < hprmbridge> nmz787> seems like gender conversion is already perfect if you just ignore gender
10:14 < hprmbridge> nmz787> outlaw mirrors and reflective surfaces
10:14 < hprmbridge> nmz787> no pond shall go still
10:14 < L29Ah> uzkruh: not in the next 200 years
10:15 < juri_> I don't know. bioprinting is a thing.
10:15 < L29Ah> Gender in modern discourse is a useless concept of sets of stereotypes loosely associated with sex. As the word lost all meaning, it is only useful for grifting and should be disregarded.
10:15 < uzkruh> Uterine transplants are already possible in Ciswomen.
10:15 < kanzure> we can send fantasy nanomachines into every cell of your body and change the chromosomes, and re-process all proteins and other constructs from a simulation that would have shown what the biological state would have been if there was a different phenotype over the whole life, but this is probably not the criteria you had in mind.
10:16 < hprmbridge> nmz787> kanzure gender!=sex
10:16 < hprmbridge> nmz787> c'mon!!!!
10:16 < fenn> uzkruh: clarify what you mean
10:16 < uzkruh> Criterion of full reproductive capability, then
10:16 < fenn> uzkruh: that's called sex
10:16 < kanzure> does outsourcing it to another person (surrogacy) count?
10:16 < uzkruh> As a start
10:16 < uzkruh> No
10:16 < fenn> yes
10:17 < hprmbridge> nmz787> jesus had a surrogate birth
10:17 < hprmbridge> nmz787> and two fathers
10:17 < kanzure> are you asking for parthenogenesis
10:17 < uzkruh> No, just cultured cell reproductive organs
10:17 < kanzure> okay that is very specific. well, let's see.
10:17 < kanzure> in vitro gametogenesis needs another 2 to 5 years
10:17 < juri_> 30 years, give or take 5.
10:18 < uzkruh> Longer than I guessed...
10:18 < uzkruh> I would have thought no more than 20
10:19 < juri_> never underestimate the capability of puritan screaming to delay technology.
10:19 < fenn> AI apocalypse throws off all estimates
10:20 < fenn> juri_: actually i would expect less objections if the technology were perfected
10:20 < kanzure> the specific requested technology is a male human able to ... carry a fertilized egg, and the egg is from someone else?
10:20 < juri_> fenn: yeah. you'd expect that, wouldn't you.
10:20 < juri_> (humans are horrible)
10:20 < fenn> well, it's not perfected
10:20 < fenn> so there are objections
10:21  * fenn points at literally every transhumanist tech
10:21 < uzkruh> kanzure: Yes?
10:21 < juri_> there will be objections as long as there is a power inbalance in the percieved genders.
10:21 < fenn> it has nothing to do with that
10:22 < juri_> sure it does. do you know how many "betraying the gender" speeches i got, when i came out?
10:22 < uzkruh> Biotech is the long term solution to transgender political disputes
10:22 < hprmbridge> nmz787> the "what is a woman" movie made clear the price tag on conversion is a big motivator to sell as much conversion as possible, regardless of the actual long term consequences
10:22 < juri_> people are horrible, and hold back tech because of their backwards beliefs.
10:22 < uzkruh> 2nd Wave Feminist objections are... confusing and mostly unimportant
10:22 < kanzure> uzkruh: proof required. i see no indication that biological reality matters in that political discourse.
10:23 < kanzure> also, i doubt that we can have a sane conversation about this topic, please keep it to technology requests
10:23 < uzkruh> It will be hard to argue that a biologically converted person who can reproduce is "fake".
10:23  * juri_ nods.
10:23 < kanzure> uzkruh: "fake" what though
10:24 < kanzure> and why would it matter if it is fake? i don't understand.
10:24 < kanzure> i think you are too optimistic about politics. it is a losing strategy.
10:24 < uzkruh> I agree about not getting into metaphysical arguments about gender
10:24 < fenn> this is why i asked you to clarify and maintain a distinction about sex vs gender
10:25 < hprmbridge> nmz787> are women and men who (are born and) can't reproduce fake women and men?
10:25 < hprmbridge> nmz787> they're at least just unable to reproduce without further changes
10:25 < kanzure> is a philosophical zombie a fake human?
10:25 < uzkruh> Huh?
10:25 < kanzure> is a philosopher that argues about philosophical zombies a fake philosopher?
10:25 < kanzure> get on our level :P
10:25 < fenn> yes, next question
10:26 < kanzure> you got me there
10:26 < uzkruh> hprmbridge, what?
10:26 < fenn> hprmbridge is a bot that relays discord chats
10:26 < fenn> the user is nmz787
10:26 < uzkruh> Is there any foreseeable technique to alter chromosomes?
10:27 < hprmbridge> nmz787> yes
10:27 < hprmbridge> nmz787> but you can still alter a chromosome to have a broken phenotype
10:27 < kanzure> modifying chromosomes in adult is substantially harder than germline modification of chromosomes
10:27 < hprmbridge> nmz787> and phenotype has incomplete determinism on a psychological state
10:27 < fenn> you could hypothetically do genetic surgery on a single cell, then grow that cell into an organ such as a gonad, then transplant that gonad into an adult
10:27 < uzkruh> broken phenotype?
10:28 < kanzure> if you alter a chromosome a little bit, you don't necessarily have the phenotype you would have had if you had started with the modifications from the beginning
10:28 < kanzure> although you might! you would have to check! science.
10:28 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
10:28 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
10:28 < fenn> it's sort of impossible to know
10:29 < fenn> without god-level computers
10:29 < fenn> uh, you could grow a copy of the chromosome-modified organism to see what happens
10:29 < hprmbridge> nmz787> like you have broken-phenotype CFTR gene (you have cystic fibrosis), you can alter it to fix the broken spot but then alter a different location to re-break it
10:29 < uzkruh> Is AI Foom a real possibility in our lifetimes, though?
10:30 < fenn> yes, next question
10:30 < kanzure> is it a fake possibility?
10:30 < fenn> also yes
10:30 < uzkruh> That's my ask
10:30 < kanzure> sounds like someone hasn't watched enough blade runner and still struggling with the whole real/fake thing
10:30 < hprmbridge> nmz787> quit dreaming of electric synapses
10:30 < fenn> blade runner is confused
10:30 < hprmbridge> nmz787> sheeple
10:31 < juri_> gattica is less confused.
10:31 < kanzure> i'll admit that whatever message blade runner wants to get across they could have done it more concisely
10:31 < uzkruh> How many more viewings? How many more folded unicorns?
10:31 < fenn> in the original story, replicants weren't genetically identical to humans, but that's too easy to detect, so then we add epicycles to beat the objection-raiser into submission
10:31 < uzkruh> How many more cuts of the film?
10:32 < juri_> uzkruh: folded cranes.
10:32 < uzkruh> The villains of Gattaca were his parents
10:32 < kanzure> the villain of gattaca was vincent
10:32 < kanzure> he's just some fraud
10:32 < kanzure> if he wanted to get into space without endangering everyone, he could have just purchased a ticket on a rocket
10:33 < juri_> the villain was society.
10:34 < uzkruh> We could interpret Vincent as having unique, intangible talents, despite not being gene-modded. So less a fraud, more of a missed opportunity.
10:34 < uzkruh> Society didn't wrong Vincent
10:35 < kanzure> he is actively deceptive and harmful to a space program
10:35 < uzkruh> If he has unique talents, he would be an asset
10:35 < juri_> yeah, i'm with kanzure on this one.
10:35 < uzkruh> The film is vague on this
10:36 < fenn> make up your mind, juri_ :)
10:36 < juri_> jerome, jerome, the metronome..
10:36 < kanzure> not a single error in a million keystrocks
10:37 < fenn> vincent's intangible talent is engaging in risky behavior and "not leaving any reserve for the swim back" - not what you'd want in a pilot
10:38 < fenn> we might extend this argument to say that all humans are bad pilot candidates
10:40 < hprmbridge> nmz787> unless you were a kamikaze
10:42 < fenn> humans are bad kamikaze missile targeting systems
10:43 < kanzure> but better than pigeons right?
10:43 < uzkruh> Out-performed by electronics, I'm afraid
10:43 < kanzure> gattaca appeals to people because it's this narrative about boundary violations- it's a very "punk"/hacker movie.
10:44 < kanzure> great movie, no doubt. but it is responsible for inhibiting biotech progress pretty strongly.
10:44 < fenn> i think you entirely missed the point of gattaca
10:44 < kanzure> and it is widely misused in debates designed to halt biotech progress
10:45 < kanzure> yeah yeah, the human will conquers all, etc
10:45 < fenn> it's inherently about racism
10:45 < hprmbridge> nmz787> idk, I felt inspired by gattaca to go do biotech
10:45 < hprmbridge> nmz787> even in the face of regulation or social discontent
10:46 < hprmbridge> nmz787> boundary violations, hacker/punk, appealed greatly to my teenage self
10:46 < hprmbridge> nmz787> #saveTheTeenagers
10:46 < uzkruh> The "'Gattaca" era in inevitable, as a percentage of people will choose to not modify genes for... reasons. Eventually these people will be washed out of the genepool, hopefully.
10:47 < hprmbridge> nmz787> don't worry, human intellect isn't that superior... we're going to want to keep a pool of normies around "just in case" we screw something up
10:47 < uzkruh> The unmodded may view it as "unfair" that nodded people achieve more.
10:48 < hprmbridge> nmz787> I already find it unfair that unmodded people achieve more than me
10:48 < hprmbridge> nmz787> but that's life
10:48 < hprmbridge> nmz787> at least I survived this long
10:49 < kanzure> "washing" something out of the gene pool sounds like an active hunt, not sure if that's what you intended to say
10:49 < fenn> yeah that's literally ethnic cleansing
10:49 < uzkruh> Why would that happen?
10:50 < kanzure> it is more likely that non-competitive traits will not proliferate as quickly as the more competitive traits
10:50 < kanzure> uzkruh: are you a native english speaker?
10:50 < uzkruh> Yes?
10:50 < kanzure> are you sure?
10:50 < kanzure> i had to learn it
10:50 < uzkruh> Why?
10:51 < kanzure> "why would that happen" refers to what?
10:51 < uzkruh> Violence over gene modding
10:51 < kanzure> what?
10:51 < uzkruh> You said "active hunt"
10:52 < kanzure> no, i asked you
10:52 < fenn> perhaps you should each just restate your position
10:52 < kanzure> 10:46 < uzkruh> The "'Gattaca" era in inevitable, as a percentage of people will choose to not modify genes for... reasons. Eventually these people will be washed out of the genepool, hopefully.
10:52 < kanzure> 10:49 < kanzure> "washing" something out of the gene pool sounds like an active hunt, not sure if that's what you intended to say
10:52 < hprmbridge> nmz787> I guess you could say being out-competed and having a strain die off isn't an active hunt as much as a hostile takeover
10:53 < uzkruh> Why would there be any active hunt?
10:53 < uzkruh> Exactly
10:53 < hprmbridge> nmz787> kanzure hasn't seen idiocracy, have you uzkruh?
10:53 < kanzure> you were the one that said washed... perhaps you're not familiar with ethnic cleansing.
10:54 < fenn> uzkruh seems to be claiming it would happen by natural selection ("social darwinism") whereas from history we know that racially motivated violence and genocide is quite common
10:54 < uzkruh> I've seen it. That future had no genemodding, though.
10:55 < uzkruh> fenn... are you saying we shouldn't modify ourselves?
10:55 < kanzure> what?
10:55 < fenn> not at all
10:55 < fenn> i'm very much in favor of enhancing as many people as possible as soon as possible
10:57 < docl> rather than nanosurgery, you might just keep supplementing stem cells and use WILT to limit existing cells reproduction rate. it's slow though
10:57 < uzkruh> Interesting
10:58 < fenn> uzkruh: should unmodified people get some sort of "affirmative action" bonus so they can remain competitive?
10:59 < uzkruh> That might be necessary
10:59 < docl> that might be necessary to avoid cancer anyway. it's one of aubrey degrey's proposals. it's also how you end up with cells that don't have mitochondrial mutants outcompeting the harmless ones
10:59 < kanzure> docl must be reading the backlog.
11:01 < fenn> uzkruh: would modified people feel resentment toward unqualified unmodified people making decisions for them?
11:01 < uzkruh> So, is AI Foom plausible within 30 years?
11:02 < fenn> it will happen in the next 5 years
11:02 < uzkruh> What makes you think so?
11:03 < fenn> cost of compute trends and the increased pace of innovation in AI
11:03 < fenn> have you used gpt-4 yet?
11:03 < uzkruh> That's speculative reasoning though, isn't it?
11:04 < fenn> i don't have a time machine
11:04 < uzkruh> Gpt-4 is far, far from AGI
11:04 < hprmbridge> nmz787> .g ai foom
11:04 < hprmbridge> nmz787> ugh, I forgot bot actions dont work from discord
11:04 < hprmbridge> kanzure> do you need AGI for foom? why?
11:04 < fenn> i plotted some trends on a graph and squinted at it real good. there will be human-equivalent supercomputers within five years. it's not clear we actually need that much
11:05 < hprmbridge> nmz787> is this what you mean by AI FOOM? https://en.wiktionary.org/wiki/foom
11:05 < uzkruh> Will this improve life for humans?
11:05 < fenn> nobody knows
11:06 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has quit [Quit: Client closed]
11:06 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has joined #hplusroadmap
11:06 < hprmbridge> kanzure> nmz787: it's hard vs fast takeoff
11:07 < hprmbridge> kanzure> uh, slow
11:07 < uzkruh> https://davidmanheim.medium.com/a-tentative-typology-of-ai-foom-scenarios-54ff20c906c3
11:08 < uzkruh> Surely we have some idea of what we want from super-AI?
11:09 < fenn> most of the doom prophecies around AI foom involve no humans in the loop
11:09 < uzkruh> We have to plan out our lives for the next decades
11:09 < fenn> no you dont
11:09 < uzkruh> If it happens in 5 years, what should we do?
11:09 < fenn> well, that depends
11:09 < kanzure> what should you do anyway?
11:09 < uzkruh> Not Butlerian Jihad, I hope
11:10 < kanzure> are you an adherent to the AI doom prophecies ?
11:10 < uzkruh> No
11:10 < kanzure> then why ask about butlerian jihad ?
11:11 < uzkruh> I'm not convinced super-AI is even going to happen in our lifetimes.
11:11 < kanzure> that's usually something proposed by those people
11:11 < kanzure> it might not, but why would you need to be convinced either way?
11:11 < uzkruh> I was mostly joking
11:12 < uzkruh> Do we have a plan for how we would benefit from friendly AI?
11:12 < fenn> i don't consider the machiavellian manipulation takeover to be a "foom"
11:12 < fenn> unless it somehow uses that to make more chips and get smarter
11:13 < kanzure> uzkruh: you can already use AI systems and benefit from those
11:13 < uzkruh> Slightly
11:13 < fenn> what if i told you... kanzure was an AI
11:14 < uzkruh> Say we had access, today, to AGI. How would we use it?
11:14 < fenn> porn
11:15 < uzkruh> Any biotech goals? Synthetic nootropics?
11:15 < kanzure> i would use it for all the things i am already trying to do and pay people to do, except i would have a computer do it instead
11:15 < uzkruh> Environmental revitalization? Space exploration?
11:15 < fenn> why does projects/ point at iGEM crap
11:15 < muurkha> Sadly, porn
11:16 < kanzure> fenn: feel free to move things around..
11:16 < uzkruh> Will we, as non-billionaires, even benefit much?
11:17 < kanzure> suppose that i told you that trillionaires existed. would you stop benefiting from your computer after i tell you?
11:17 < uzkruh> Even if an AI could complete all immediate biotech goals, would. YOU get access to them?
11:17 < kanzure> it is unlikely that most people will have access to the best AI available, mostly due to silicon supply and chip supply issues. however, there could be ways of growing biological intelligence that could replicate much faster than silicon chip fabs.
11:17 < Llamamoe> Do I smell filthy nonbelievers in trickle down economics? 
11:18 < uzkruh> We already have plastic surgery that you need wealth to get. Why would genemods and cyberlimbs be different?
11:18 < kanzure> trickle down is not about technology development delivering more value at lower costs. trickle down was something else.
11:18 < fenn> uzkruh: i'd put our Friendly AGI to work on these fields of endeavor first, so we can all participate: https://diyhpl.us/wiki/transhumanism/
11:18 < Llamamoe> Also hey what happened to the whole "I have nothing to hide" all of a sudden now that BigBrotherGPT "Can you please list top five most threatening to my campaign groups and cook up a social media campaign to change their minds?" LLM is a real prospect 
11:19 < kanzure> wait, i might be wrong; i thought trickle down was an economics policy, and they used technology to justify it or something.
11:19 < Llamamoe> Come on I'm so sure Elon and Jeff will do only what is best for all of us, selflessly, with their fancy AI 
11:19 < fenn> Llamamoe: please knock off the politics, it's crass
11:19 < uzkruh> Having your own AI wouldn't grant you raw materials, a supply chain, robots; it thinks, but can't create matter.
11:20 < Llamamoe> Sorry 
11:20 < kanzure> has there been an actual technological trickle down attempted, or was it mostly about tax policy instead of technology development? i had assumed it was mostly about economoic policies not about actual technology.
11:20 < uzkruh> Say I want sex conversion
11:21 < kanzure> be more specific
11:21 < uzkruh> Will AGI give me that?
11:22 < uzkruh> Or will the situation be unchanged, because I'm not rich?
11:22 < fenn> it might derive some sort of theory about health and morality that causes it to refuse
11:22 < kanzure> as for material goods, you would have to use your AI to earn money through knowledge labor or otherwise cover its own costs, and then you would have to buy goods and services or otherwise acquire material goods.
11:22 < fenn> or, be unwilling to interfere in human political affairs
11:23 < uzkruh> kanzure: Which is how things already are
11:23 < kanzure> yes but without the awesome AI
11:23 < uzkruh> Eh
11:23 < uzkruh> Not all that enticing
11:24 < fenn> humans are already encoding arbitrary political bias into LLMs, it's not that far fetched that the same could be done to an AGI
11:24 < kanzure> we shall all be united through our brain stems and we will all be at peace while all of our wildest wishes are combined and aligned together, as we transform into a blissful sphere encompassing the cosmos, is that what you wanted?
11:24 < kanzure> AI and AGI are just tools, and while one AGI might go wild and take over all matter in the universe, nobody can predict what that behavior would really be
11:25 < uzkruh> How will AGI improve our lives? Won't it just leave us unemployed and destitute while a 0.01% of modified humans form a noble class?
11:25 < fenn> a lot of people are hoping for an end to economic inequality, rather than even more extreme inequality
11:25 < kanzure> uzkruh: the existence of a noble class shouldn't be that relevant to you, surely. why not continue to use AI on your own even if you are not a member of that class?
11:25 < uzkruh> I don't have much use for it
11:26 < fenn> this is all very vague. you should ask about specific scenarios rather than "AGI"
11:26 < uzkruh> For example?
11:26 < fenn> it's like "will civilization be good for us, the uzkruh tribe?" well, it depends...
11:26 < kanzure> life is what you make of it?
11:27 < uzkruh> Civilization And It's Discontents, 2050 Updated Edition
11:28 < fenn> an AGI developed by OpenAI would have very different goals and biases than one developed by Russia
11:29 < uzkruh> Is there a plan, though?
11:29 < fenn> ?
11:29 < uzkruh> To make AI benefits us?
11:29 < uzkruh> *make sure
11:29 < fenn> https://openai.com/blog/planning-for-agi-and-beyond
11:30 < uzkruh> Not just AI Alignment.
11:30 < fenn> is there a difference?
11:30 < fenn> do you know about the law of comparative advantage?
11:31 < uzkruh> Can we trust OpenAI to have our interests in kind?
11:31 < uzkruh> *in mind
11:31 < fenn> disregarding the scenario where AI fucks off and disappears into the cosmos never to be heard from again, an aligned AGI would trade and interact with the world in overall beneficial ways, by definition
11:32 < uzkruh> Even assuming we have a Friendly AI, our lives wouldn't necessarily improve
11:32 < fenn> you could get hit by a bus
11:32 < uzkruh> Yes?
11:33 < fenn> i dunno
11:33 < fenn> shit happens
11:33 < fenn> nobody's perfect, not even a super duper AGI
11:34 < uzkruh> So, if you had access to a super-AI, what would you do with it?
11:34 < kanzure> nobody promised AI or AGI would be some sort of universal beneficence blanket over humanity/all things
11:34 < kanzure> i would probably continue to do what i am already doing, but it would end up being done faster
11:35 < uzkruh> No, they certainly did not
11:35 < uzkruh> That's not a bright future, then
11:35 < kanzure> one of the things about the definition of the singularity is that it's supposed to be this point beyond which you are unable to plan or predict in any meaningful way because things change so rapidly
11:35 < kanzure> it could be very bright, how would you know it's not
11:36 < uzkruh> How do we ensure a good future?
11:36 < kanzure> has that ever been done before?
11:36 < uzkruh> Yes?
11:36 < kanzure> when? who gets to define good?
11:37 < fenn> predicting the present is hard enough
11:37 < kanzure> when has that been done before?
11:37 < uzkruh> Will it be beneficial to YOU?
11:37 < fenn> is that a good future?
11:37 < fenn> it sounds selfish
11:37 < kanzure> no. answer my question. you said that ensuring a good future has been done before. when?
11:38 < uzkruh> This is confusing. If AI won't give us a better future, why would we promote it?
11:39 < kanzure> nobody said it can't give its user a better future
11:39 < uzkruh> Is this an AI Prepper channel?
11:39 < juri_> no? i write haskell.
11:39 < kanzure> maybe? what is AI prepper?
11:40 < kanzure> we have a fairly strong interest in intelligence both biological and artificial. some of us are users of AI tools. others are bystanders. some build AI tools.
11:40 < uzkruh> So we just... wait and see?
11:41 < juri_> again, i write haskell.
11:41 < kanzure> you could wait, but what are you waiting for? what do you want to be doing right now?
11:41 < juri_> writing haskell.
11:41 < uzkruh> No plan, it's inevitable, it will change things for good or bad, we just have to live with it?
11:41 < juri_> :)
11:41 < fenn> uzkruh you seem ot be very uncomfortable with navigating uncertainty
11:42 < fenn> the future was always highly uncertain even without AI
11:42 < uzkruh> Why are you psychologizing this?
11:42 < kanzure> because we are trying to understand your questions and what you are really asking
11:42 < juri_> because i want you to write haskell.
11:42 < juri_> :D
11:42 < kanzure> that too
11:42 < uzkruh> I dont
11:42 < uzkruh> I don't code
11:43 < uzkruh> Not a STEM person
11:43 < juri_> you seem interested in STEM topics.
11:43  * juri_ shrugs.
11:43 < fenn> we generally promote the idea of creating the future you want to happen
11:43 < fenn> it doesn't mean we exist in some alternate reality where nothing outside of our control happens, though
11:44 < uzkruh> How do we do that?
11:44 < juri_> i write haskell.
11:44 < fenn> by working on things you expect to have good outcomes
11:44 < fenn> by not working on things you expect to have bad outcomes
11:44 < juri_> specifically, i'm working on a 6 axis slicing system with high precision. i'm hoping it'll be useful when bioprinting.
11:44 < uzkruh> Which will be completely surpassed by AI in 5-10 years?
11:45 < juri_> have you talked to AI? I mean, i'm impressed, but it will be a long time until it does what i do.
11:45 < fenn> not everyone shares my prediction timeline
11:45 < uzkruh> Ok...?
11:45 < juri_> it's a useful debate partner, but kindof an idiot.
11:46 < fenn> it's justifiable to do manual work on things that will be obsoleted by AI, if you don't expect that to happen soon
11:46 < juri_> in fact, doing the manual work will increase the dataset for the AIs.
11:47 < uzkruh> Will AI benefit anyone without a tech job?
11:47 < fenn> on the other hand, yesterday i saw this guy learning how to draw and trying really hard to justify that to the world, getting angry at the hypothetical twitter viewer judging him for that choice. it seemed misguided
11:47 < juri_> do tech jobs benifit anyone without a tech job?
11:47 < uzkruh> You tell me
11:47 < fenn> yes, society is vastly benefitted by technology
11:48 < uzkruh> This makes AI sound... paltry. Not worth thinking about.
11:48 < juri_> exactly the opposite.
11:49 < juri_> i can't find anyone to keep up with what i do, so having an AI i can argue with about math for hours? good stuff.
11:49 < fenn> uzkruh what are you interested in doing? you're not a STEM person, unsure about the benefits of technology, it's hard to imagine what you might find "good" or of benefit
11:50 < uzkruh> If I don't work in STEM, don't own a company, and can't predict when or how it will manifest... why would I care?
11:50 < fenn> i'm imagining some sort of religious monk dedicated to self-cultivation and character development
11:50 < fenn> is that an accurate guess? (i didn't try very hard)
11:51 < uzkruh> More or less
11:51 < fenn> well, AI will be able to help you with that too, by providing more interaction scenarios for your own personal development
11:51 < juri_> bioprinting will be neat. maybe then, i won't be "fake".
11:51 < fenn> consider why we play video games. games are essentially training in a simulation, so we have better performance in real scenarios
11:52 < uzkruh> Will we be able to afford bioprinting, though?
11:52 < juri_> I do everything with Free Software.
11:52 < fenn> simulation training can be extended beyond the typical "shoot the bad guys" into situations with real moral depth and tests of personal integrity
11:53 < uzkruh> I doubt we need moral moral quandries; our lives are complex enough.
11:53 < fenn> how about a universal basic income? would that simplify your life?
11:54 < uzkruh> Will AI result in that?
11:54 < fenn> like i said, it depends
11:55 < uzkruh> Is there cause to hope?
11:55 < fenn> yes, the current leader in AI, OpenAI, has committed to pursuing UBI
11:55 < juri_> if you can be told to hope by someone on the internet.. go read papers. :)
11:55 < uzkruh> ?
11:56 < fenn> the argument against UBI was always "who will pay for this?" but when a miracle machine is pumping out trillions of dollars in excess economic activity, the answer is easy
11:56 < juri_> the papers all of this is based on are public.
11:58 < uzkruh> If we have UBI but no jobs, we also can't go on strike, and would have no outside income. Couldn't the upper class take advantage of us?
11:58 < juri_> very. they are, now.
11:58 < fenn> it's better than no UBI and no jobs
11:58 < uzkruh> Steady-State Society
11:59 < fenn> is this one of those NWO conspiracy theories?
11:59 < juri_> solid state society was better.
11:59 < fenn> stand alone complex
12:00 < uzkruh> Oppression before AI, oppression after
12:00 < fenn> it's not like you're the only one worrying about this
12:02 < fenn> "Individuals who, with no coordination or knowledge of their peer's actions, act as though they're working together toward a common goal"
12:03 < juri_> fenn: isn't that what we do here?
12:03 < fenn> it could describe an illuminati non-conspiracy or a hacktivist movement just as easily
12:03 < juri_> sure, but when it comes time to cut off heads, count me out.
12:04 < fenn> you don't want your head cut off or you don't want to cut off heads?
12:05 < uzkruh> "I never thought leopards would eat MY face" said tyhe Leopards Eating Faces Party voter
12:05 < juri_> I'm rather attached to the contents of my head.
12:05 < fenn> regarding "AI prepper" question, Wei Dai said that cryptography was basically your only defense against a superintelligence
12:06 < fenn> we can timestamp things to prove they existed, and pile hashrate and economic incentives on top so those attestations can't be forged
12:07 < fenn> unfortunately for unknown reasons the internet archive hasn't been taking this seriously
12:08 < fenn> this channel's logs will be timestamped at the end of the day, so at least this conversation will be proven to have existed
12:08 < fenn> unless cryptography gets broken and nobody fixes it
12:09 < fenn> or at least, the superintelligence will be forced to maintain a web of lies and be constrained by logic, instead of just inventing history out of whole cloth
12:09 < hprmbridge> nmz787> "uzkruh> Do we have a plan for how we would benefit from friendly AI?" do less work on things I don't enjoy or get easily distracted about because I've offloaded much of the work... do things I care more about or am less annoyed by (and which I can't offload now/at the time)
12:10 < uzkruh> Personally, I don't forsee much benefit to me
12:11 < fenn> you are spoiled
12:12 < uzkruh> Um
12:12 < uzkruh> What?
12:13 < fenn> if you really have nothing in your life that would be improved by a radical increase in economic abundance, then you're far better off than most people
12:14 < fenn> also you probably are suffering from excessive pessimism, or postmodernism
12:14 < uzkruh> Why would we have more abundance?
12:14 < fenn> well duh that's the whole point
12:14 < uzkruh> AI could as easily impoverish all but a few
12:15 < fenn> why bother
12:15 < juri_> capitalism impoverishes all but a few. how would that be different?
12:15 < uzkruh> That's the point
12:15 < fenn> it's extra effort to take stuff from people vs just making more stuff
12:16 < fenn> it would be like carefully excavating an ant hill to get the ant eggs
12:16 < juri_> so, i've chosen to do something, rather than nothing, in the face of a historic level of uncertainty.
12:16 < juri_> if i'm wrong, at least i did something about it.
12:17 < fenn> i'm trying to keep up with AI news and get skilled at using today's tech
12:17 < fenn> and generally not dying
12:17 < juri_> good luck.
12:17 < fenn> you too
12:18 < fenn> is chatGPT well informed about f-rep methods?
12:18 < fenn> and, er... algebraic topology? (is that relevant?)
12:19 < juri_> I have found it's useful to know what problems are mind-numbing to solve, and what problems have solutions in special cases.
12:19 < juri_> but you REALLY have to lead it around.
12:19 < juri_> (and that's with GPT4)
12:21 < muurkha> fenn: Brewster isn't really into high-tech solutions or solutions to potential problems.  he likes to solve problems when they appear
12:21 < uzkruh> AI Risk: We don't know when it will happen, or what it will do, or how to respond to it, but it will change everything in unforeseeable ways.
12:21 < uzkruh> Pascal makes a new Wager
12:21 < juri_> yep.
12:22 < muurkha> fenn: have you seen those videos where people pour molten aluminum down anthills to get a cool aluminum casting of the anthill shape?
12:22 < fenn> pascal's wager has no downside, but spending all your time worrying about AI does
12:23 < fenn> muurkha: yes, we could all be turned into modern art i suppose
12:23 < uzkruh> Pascal's Wager has in infinite downside?
12:23 < muurkha> hopefully it will be more profitable to employ us
12:24 < uzkruh> I'm NOT worry about AI, since I can't predict, changfe, or plan for it.
12:24 < juri_> s/AI/*/
12:24 < fenn> "Pascal's Wager: Belief in God is rational, as potential gains outweigh losses if God exists; disbelief risks eternal suffering if wrong." there's no downside to "believing" in god and acting as you would have anyway
12:25 < uzkruh> No religion offers those terms
12:26 < fenn> uh what was my point again
12:26 < fenn> it's good to not be totally clueless. that was my point
12:26 < fenn> it takes time and effort to not be clueless
12:27 < muurkha> and usually it doesn't work anyway
12:27 < uzkruh> I'll take a break from thinking about AI, then.
12:27 < fenn> muurkha: you're too humble
12:28 < muurkha> fenn: rather than "too humble" I would say "too clueless"
12:42 < hprmbridge> nmz787> belief in god doesn't imply belief in a specific religion
12:43 < hprmbridge> nmz787> to me it's like saying I believe in physics, though I don't particularly understand or know or subscribe wholly to any particular model
12:45 < muurkha> how does "I believe in physics" make sense?  what's the alternative, Berkelian idealism maybe?  solipsism?
12:46 < muurkha> wouldn't Berkelian idealism qualify as a physics too?
12:49 < hprmbridge> nmz787> I'm not sure what they teach at Berkeley
12:49 < hprmbridge> nmz787> pretty sure physics is a subject
12:49 < fenn> the other berkeley, the guy the town is named for
12:50 < hprmbridge> nmz787> idk him
12:50 < hprmbridge> nmz787> physics is a term though
12:51 < hprmbridge> nmz787> the alternative is (in hillybilly accent) "I don't believe in physics"
12:51 < fenn> "Subjective idealism is a form of philosophical monism that holds that only minds and mental contents exist."
12:52 < muurkha> probably this isn't going to be a productive conversation then ;)  we're already down in the mud of reasoning from ethnic stereotypes instead of logical propositions
12:54 < fenn> sorry i said the g-word
12:54 < hprmbridge> nmz787> some people probably have never heard of physics
12:54 < hprmbridge> nmz787> so how could they believe?
12:55 < hprmbridge> nmz787> /me  goes out to prosyletize physics
12:55 < fenn> even babboons have expectations about how objects will behave
12:55 < fenn> where do i get these misspellings from
12:55 < hprmbridge> nmz787> huh physics isn't on this list https://www.lcps.org/cms/lib4/VA01000195/Centricity/Domain/16649/Ology%20List%20of%20Sciences.pdf
12:56 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Quit: 9EA brb GCTGGG 2023-06-21 19:55:59:756]
12:56 < fenn> anne marie helmenstine, the greatest bootlegger of all time
12:56 < hprmbridge> nmz787> "Loudoun County Public Schools"
12:56 < hprmbridge> nmz787> they've had a bunch of public controversy... so I guess I shouldn't be surprised
12:57 < fenn> no the document is from about.com, a content mill
12:57 < fenn> helmenstine is their science plagiarizer in chief
13:09 < hprmbridge> nmz787> chatGPT is my new stackoverflow
13:14 -!- Llamamoe [~Llamamoe@46.204.72.21] has quit [Quit: Leaving.]
14:00 -!- codaraxis__ [~codaraxis@user/codaraxis] has joined #hplusroadmap
14:03 -!- codaraxis [~codaraxis@user/codaraxis] has quit [Ping timeout: 240 seconds]
14:13 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has quit [Quit: Client closed]
15:07 < L29Ah> huh uzkruh entertained you for so long
15:14 < hprmbridge> nmz787> was it an AI that you wrote?
15:14 < hprmbridge> nmz787> you sly fox
15:53 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has joined #hplusroadmap
15:55 < uzkruh> Will Chatgpt-5 be able to manipulate symbols to brainhack humans?
15:56 < hprmbridge> kanzure> what, like propaganda does?
15:56 < uzkruh> The new Sumerian training data...
15:57 < uzkruh> I haven't read Neuromancer, somehow. You may throw rotten fruit at me. 😔
15:57 < hprmbridge> kanzure> this is typically the story of the "silver tongue"-- but people are finding that often what happens with AI is that people spontaneously do the thing anyway and ask the AI to do it, rather than the AI tricking them into doing it.
15:58 < hprmbridge> kanzure> (and that kind of obeyance is good)
16:00 < uzkruh> https://en.m.wikipedia.org/wiki/BLIT_(short_story)
16:00 < fenn> i find myself looking at a lot more stable diffusion output than i really should
16:00 < fenn> it's mostly humans doing the optimizing though
16:01 < fenn> there's a sort of "holy shit i can't believe how good this is" reaction that won't stop
16:01 < uzkruh> If I wanted to make a game in Unity, I could use AI for pixel art and music...
16:02 < uzkruh> But can chatgpt-4 write code?
16:02 < kanzure> yes
16:02 < kanzure> and so can github copilot and github copilot X
16:03 < fenn> and a bunch of other code-specific models
16:03 < uzkruh> If I tell it to code Nethack in Unity, it would be functional-with-debugging?
16:03 < kanzure> .gpt -4 Write "hello world" in python saying hello to uzkruh. Include a fanficul piece of ASCII art like pyramids or stars and special effects.
16:03 < uzkruh> I've wanted to make a game, but my non-coding is a barrier...
16:03 < gptpaste> ​print("Hello uzkruh!")print("╔═══╗")print("║███║")print("║(o) ║")print("╚═══╝")# Adding special effectsimport timefor i in range(5):    print("" + "Hello uzkruh!" + "") # Red text    time.sleep(0.5)    for i in range(3):    print("" + "Hello uzkruh!" + "") # Yellow text    time.sleep(0.5)    for i in - http://sprunge.us/pa2rcT
16:04 < uzkruh> Caves of Qud is like this
16:04 < fenn> huh the color codes render for me, i didn't realize you could send raw terminal color codes over IRC
16:05 < kanzure> .gpt -4 Write the main game loop for a Unity clone of nethack for Uzkruh. Call it Uzkruh's Nethack.
16:05 < gptpaste> ​As an AI language model, I cannot provide a complete code for the main game loop of Uzkruh's Nethack. However, I can give you an overview of what the main game loop should include:1. Initialize the game world and player character.2. Display the game interface and user input options.3. Wait for user input and process it accordingly (e.g., move player character, attack enemy, - http://sprunge.us/iTBVSJ
16:05 < kanzure> sometimes it misbehaves like that
16:05 < uzkruh> Oof
16:05 < kanzure> you have to be stern with it
16:05 < uzkruh> Pesky "laws" and "ethics'"...
16:06 < fenn> it's .gpt4 not .gpt -4
16:06 < kanzure> .gpt4 Show an example of basic game loop control for Unity for a version of Nethack and call it Uzkruh's nethack.
16:06 < kanzure> .. are you sure?
16:07 < kanzure> .gpt Show an example of basic game loop control for Unity for a version of Nethack and call it Uzkruh's nethack.
16:07 < gptpaste> ​To create a basic game loop control for Unity in a version of Nethack called Uzkruh's Nethack, you can follow these steps:1. Create a new Unity project and name it "UzkruhsNethack".2. In the Assets folder, create a new folder called "Scripts".3. Inside the Scripts folder, create a new C# script called "GameLoopControl".4. Open the GameLoopControl script in your preferred c - http://sprunge.us/3yqt1Y
16:07 < fenn> to get better code it helps to have it lay out its plans first, then implement them
16:07 < gptpaste> ​Here is an example of a basic game loop control for Unity for a version of Nethack called Uzkruh's nethack:```using UnityEngine;using System.Collections;public class UzkruhsNethack : MonoBehaviour {    private bool gameOver = false;    void Start () {        // Initialize game objects and variables    }    void Update () {        if (!gameOver) {            // H - http://sprunge.us/oiAsEw
16:08 < uzkruh> Fascinating
16:09 < muurkha> fenn: IRC terminal codes are not the same as ANSI color codes
16:09 < fenn> the refusal was because it didn't know what Uzkruh's Nethack was, i guess it assumed that was an existing game it didn't know about?
16:09 < fenn> muurkha: why would it print IRC color codes though?
16:10 < fenn> .gpt print some red words in ANSI color codes and then in IRC color codes
16:10 < gptpaste> ​ANSI color codes:- red- redIRC color codes:- 4red- 5red# Thu 22 Jun 01:10:12 CEST 2023 - print some red words in ANSI color codes and then in IRC color codes - http://sprunge.us/0nzZfK
16:10 < fenn> maybe the bot is doing translation
16:10 < kanzure> might also depend on the bot library
16:10 < kanzure> shoo
16:11 < kanzure> uzkruh: what is your education or job if any?
16:13 < fenn> according to geohot, gpt4 is just 8x 220B gpt3's trained on different data and glued together
16:14 < fenn> "mixture of experts"
16:14 < kanzure> aren't we all
16:14 < fenn> c. elegans isn't
16:29 < fenn> chatGPT recomments an exclusively bioluminescent diet, as part of global traditional alien cuisine day
16:29 < kanzure> for the record, i think a first contact scenario would actually result in us knowing collectively less, or confusion increasing greatly.
16:30 < fenn> "knowing"
16:32 < hprmbridge> kanzure> https://youtu.be/urcL86UpqZc
16:32 < Muaddib> [urcL86UpqZc] Interview with an Emacs Enthusiast in 2023 [Colorized] (8:50)
16:33 -!- sgiath [~sgiath@2a02:25b0:aaaa:aaaa:a3c3:ed4b:6b06:0] has quit [Server closed connection]
16:33 < fenn> "Indeed, restructuring entire cities to resemble bioluminescent coral reefs and incorporating such ecosystems would come with significant expenses."
16:33 -!- sgiath [~sgiath@mail.sgiath.dev] has joined #hplusroadmap
16:36 < fenn> Opinion: The government should create a mandatory annual scrapbooking event where everyone's most embarrassing moments are organized into publicly displayed, interactive wall exhibits, encouraging laughter and deep contemplation.
16:37 < fenn> i explicitly asked for no dystopias *thinking face*
16:37 < kanzure> "parenthesis paralysis"
16:38 < kanzure> "you can make a keybinding for that"
16:38 < kanzure> "also you can bind that keybinding to a shorter keybinding."
16:39 < fenn> sounds like bit pair encoding
16:39 < fenn> byte pair encoding*
16:39 < fenn> same diff
16:46 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has quit [Quit: Client closed]
16:47 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has joined #hplusroadmap
16:47 < fenn> not a dystopia apparently:
16:48 < fenn> 'Rotate jobs among people randomly every 30 days and, daily, workers must perform the previously mentioned "See Random Cute Animal Video" ritual 37 times intermittently, record their 20-second video impressions, and congregate for an hour weekly to facilitate skill-sharing, constructing a virtuous cycle of learning and relationship building in which new opinions and ideas are openly discussed and
16:48 < fenn> celebrated.'
16:51 < fenn> "With neon green as the global standard, it becomes simpler to coordinate interior design choices."
16:53 < fenn> '"color-ticker" technology that automatically alters the neon shades in every room daily within a wider color spectrum'
16:56 < fenn> "Permanently levitating prevents us from walking on delicate ecosystems like plant roots, protecting their habitats, and maintaining natural ecological relationships."
17:02 < uzkruh> Fun
17:02 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
17:03 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Killed (tantalum.libera.chat (Nickname regained by services))]
17:03 < fenn> ' We should inform some tree populations of human transport history, using the "wood wide web," to learn valuable lessons about unleashing SPLT (silent personal levitation technology) context of a society embracing the closure of major highways.' (this is because highways disrupt tree communication, apparently)
17:04 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has quit [Quit: Client closed]
17:04 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
17:08 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has joined #hplusroadmap
17:10 < fenn> "whenever someone sneezes, they should save the mucus and add it to canisters, before sending the fully packed canister to a specially created park, named the Unified Communal Allergy Center"
17:13 < fenn> ' "The Genomic Aria," would consist of an intricate stained-glass installation spanning several meters. Fueled by the discovered genetic relationships and patterns, each glass piece represents a nucleotide base, gradually composing vast sequences from analyzed mucus samples.'
17:16 -!- Malvolio is now known as Guest1242
17:16 -!- Guest1242 [~Malvolio@idlerpg/player/Malvolio] has quit [Killed (zirconium.libera.chat (Nickname regained by services))]
17:17 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
17:23 < fenn> "While this may sound strange, cockroach-flavored ice cream could become a customized undercooking solution. If certain cultures accept insects as a food source and a person's gut flora tolerates cockroach-based nutrients, then such a product can be made and adjusted within the undercooking framework" (for optimal gut flora)
17:23 < fenn> "this flavor choice caters to the most refined and unique taste and will ultimately define the gastronomical preferences of our time."
17:25 < fenn> "tattoo QR codes on uncovered knees"
17:25 < uzkruh> Snowpiercer
17:27 < fenn> "However, such a system should uphold data privacy guidelines, allowing individuals to choose the content of their QR codes and control which information is accessible"
17:28 < fenn> 'It was not ice cream, but rather "protein blocks". This symbolizes the poor living conditions and exploitation experienced by inhabitants of the lower sections of the train.'
17:30 < fenn> 'if humans created an international organization focused on deciphering, documenting, and utilizing whale mating songs, they could rediscover the long-lost knowledge of Atlantis to advance humanity in unforeseen ways.'
17:35 < fenn> 'By working in communion with cetacean philosophers and combining our understanding of the "wood wide web," future interspecies collaborations could establish contact with extraterrestrial intelligent life.'
17:37 < fenn> omg
17:38 < fenn> 'Humanity should voluntarily dedicate one generation to collectively forego sleep in favor of scientific, cultural, and social development, essentially working round-the-clock while being artificially sustained by advanced bioengineering and neurostimulation technology. While this would be an incredibly difficult sacrifice and not align with existing individual desires or beliefs, the massive
17:38 < fenn> progress achieved would unarguably propel future generations into a flourishing era.'
17:38 -!- uzkruh [~uzkruh@fl-67-235-196-252.dhcp.embarqhsd.net] has quit [Ping timeout: 246 seconds]
17:39 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 240 seconds]
17:39 < jrayhawk> sleep deprivation is a very traditional brainwashing technique
17:41 < fenn> "every human's role is constantly interchanging between scientific, artistic, manual labor, and cultural stewardship activities. These work rotations would facilitate communal ownership and the rapid assimilation of global advancements into society"
17:42 < fenn> i think chatGPT just invented coralpunk
17:45 < fenn> "To ensure humanity's long-term coherence with its core values amid the dizzying pace of development and interplanetary colonization, humanity needs to establish a unique, ever-expanding decision-making infrastructure consisting of synthetic philosopher-guides created through precise cognitive balancing of individuals under the aforementioned sleep-deprived, labor-intensive global stewardship.
17:45 < fenn> These philosopher-guides—converged AI-human hybrids yet to be conceived by
17:46 < fenn> any present mind—will steer humankind's pursuits with an intertwined grasp of logic, ethics, and aesthetics. As the torch-bearers of a radically new era, these entities will help humanity navigate decisions of universal consequence, guarding against the erosion of fundamental virtues by artificial priorities and bridging the otherwise
17:46 < fenn> yawning gap between unbridled futurism and sacrosanct human ideals."
17:48 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
17:48 -!- TMM_ [hp@136.243.23.165] has joined #hplusroadmap
17:51 < fenn> "The next crucial evolutionary step is for the human-animal-artificial intelligence entity to not only engage in cooperative relationships with formerly  considered invasive species (e.g., rats, Asian carp) but also to develop and deepen our communication with these species to extract valuable information about tuning ecosystems while co-creating a socio-ecological system with enhanced
17:51 < fenn> adaptability."
17:53 < jrayhawk> the hazards of training on reddit
17:54 < fenn> "create and implement a comprehensive language alphabet formed from distinctive signals and communication processes used by diverse species. This would enable a direct representation of all ecosystem participants at significant international and interspecies events, further enabling collaborative decisions on ecological sustainability to guarantee representation of individual beings and
17:54 < fenn> ecosystems across the entire biome spectrum."
17:54 < fenn> jrayhawk: is it wrong? i value your input
18:22 < fenn> oh apparently coralpunk is a thing already? https://i.redd.it/1dj9avzs8xi91.png
18:23 < fenn> no, not really
19:05 < kanzure> huh
19:18 < hprmbridge> nmz787> https://twitter.com/BrianLongFilms/status/1671423822443212803
19:18 < hprmbridge> nmz787> Ai for some marvel tv show intro
19:19 < hprmbridge> nmz787> "The revelation is prompting some backlash from audiences on social media, given that using AI presumably eliminated the need for graphic designers and animators to craft the opening credits.
19:19 < hprmbridge> nmz787>
19:19 < hprmbridge> nmz787> The argument is especially timely, since the WGA is currently on strike after failed negotiations with the Association of Motion Picture and Television Producers, which included language about protecting writers against the use of AI in the writing process. Over the past eight weeks, the use of AI to replace laborers has come to the forefront of many discussions about the strike."
19:24 < fenn> worse than amateur
19:29 < jrayhawk> phoenix point was very coralpunk
19:30 < fenn> maybe that tv show intro was made a long time ago and is now finally wending its way through the production process
19:30 < fenn> i don't think the mainstream is really aware of just how good the models are now, in capable hands
19:33 < fenn> phoenix point looks very generic mass effect + judge dredd + cthulhu mythos
19:34 < jrayhawk> the thing julian gollop really aped is Alpha Centauri
19:34 < fenn> i guess this is something https://www.pcgamesn.com/wp-content/uploads/2021/09/phoenix-point-acheron-corrupted-horizons.jpg
19:35 < jrayhawk> "what if instead of a terrestrial world mind, we had an aquatic world-mind?" and then he proceeded to copy Gaia's Stepdaughters in the form of The Disciples of Anu
19:41 < jrayhawk> which, to his credit, was a good choice. alpha centauri was a rich source of narrative ideas few people have been exposed to
19:44 -!- codaraxis__ [~codaraxis@user/codaraxis] has quit [Ping timeout: 240 seconds]
19:49 < jrayhawk> a pity gollop's raison d'etre, tactics, was implemented as tedious unskippable braindead DPS racing no sane person wants to sit through 40 hours of
19:50 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0:c4bf:8b01:e4e5:75eb] has quit [Quit: Leaving]
20:13 < fenn> the concept of hit points and health bars should be retired
20:18 -!- flooded [~flooded@149.102.254.66] has joined #hplusroadmap
20:22 -!- _flood [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 258 seconds]
20:34 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
21:19 -!- codaraxis [~codaraxis@user/codaraxis] has quit [Ping timeout: 245 seconds]
21:43 -!- mxz [~mxz@user/mxz] has quit [Quit: cya]
21:44 -!- mxz [~mxz@user/mxz] has joined #hplusroadmap
22:51 -!- stipa [~stipa@user/stipa] has quit [Quit: WeeChat 3.0]
23:03 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
23:26 < hprmbridge> Eli> I’ve been reading about prebiotic chemistry lately. Specifically nick lane. There are a number scientists working on the idea of how to create something from nothing. Create a self replicating life form from scratch. I used to think this would be impossible, and I still think it will be stupendously difficult given we have no idea what life originally looked like, but I’ve started to think that
23:26 < hprmbridge> Eli> there is a chance this could happen. There’s actually a number of people working on all aspects of the problem and discovering natural ways for the Krebs cycle to originate, for example. But if this can be done, it opens up a very interesting world of options. From allowing us to terraform other planets via panspermia, to allowing us to experiment with making different optimal types of life on
23:26 < hprmbridge> Eli> earth. Cells are self replicating and excellent at producing things that can be very useful to humans.
23:28 -!- test__ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
23:31 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 250 seconds]
23:31 -!- flooded [~flooded@149.102.254.66] has quit [Ping timeout: 252 seconds]
23:31 < hprmbridge> Eli> The reason prebiotic chemistry is important is because we can learn what is thermodynamically favorable for life to begin. So, if we wanted to terraform another planet with synthetic cells, we might not know if they will be optimal or even successful on another planet. Understanding how life can arise spontaneously allows us to understand all the pathways so we can be more assured of success.
23:34 < hprmbridge> Eli> After all, the first organisms essentially terraformed earth. There was thousands of more CO2 in the atmosphere before eukaryotes appeared. The metabolic waste product of oxygen caused mass climate change and killed most of the organisms that existed, allowing eukaryotes, and thus humans, to survive.
23:47 < muurkha> prebiotic chemistry might be an interesting source of inspiration
23:48 < muurkha> but designing a replicator doesn't require knowing how one arose spontaneously
23:49 < muurkha> or could arise spontaneously
--- Log closed Thu Jun 22 00:00:37 2023