--- Log opened Wed Aug 09 00:00:23 2023
00:42 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 260 seconds]
00:43 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
00:44 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
01:40 -!- Urchin[emacs] [~user@user/urchin] has quit [Ping timeout: 256 seconds]
01:55 -!- Urchin[emacs] [~user@user/urchin] has joined #hplusroadmap
02:26 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
02:59 -!- Urchin[emacs] [~user@user/urchin] has quit [Ping timeout: 256 seconds]
03:20 -!- Urchin[emacs] [~user@user/urchin] has joined #hplusroadmap
04:38 -!- Roy_Mustang [A_D@libera/staff/dragon] has joined #hplusroadmap
04:41 -!- A_Dragon is now known as Guest8048
04:41 -!- Guest8048 [A_D@libera/staff/dragon] has quit [Killed (tungsten.libera.chat (Nickname regained by services))]
04:41 -!- Roy_Mustang is now known as A_Dragon
05:05 -!- Netsplit *.net <-> *.split quits: archels, gwillen, fenn, lkcl, SDr|hu, Betawolf
05:05 -!- Netsplit over, joins: gwillen, Betawolf, archels, SDr|hu, lkcl, fenn
06:15 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
07:51 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
07:55 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 256 seconds]
08:42 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Quit: Gateway shutdown]
08:42 < muurkha> hello juri_  and Urchin[emacs] 
08:43 < muurkha> really I think the FDA should revert to not testing drugs for efficacy
08:45 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
09:12 < Urchin[emacs]> what would be the point then?
09:16 < muurkha> testing them for safety
09:16 < muurkha> which was all they did for the first 50 years, which were perhaps not coincidentally the 50 years of greatest pharmaceutical progress in human history, much of it in the US
09:18 < Urchin[emacs]> are you sure it's not because they found a lot of low hanging fruit in that period?
09:19 < muurkha> I'm definitely not sure
09:20 < muurkha> I did say "perhaps not coincidentally", not "not coincidentally".  It might be coincidence, as you say
09:21 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
09:26 < hprmbridge> nmz787> Snake oil is still snake oil even if it doesn't hurt you
09:26 < Urchin[emacs]> ^
09:33 < CNOT> they (our predecessors) didnt even know the brain was doing chemistry really, they went from totally ignorant to having a clue during those years
09:34 < CNOT> like, what the fuck were they thinking
09:34 < Urchin[emacs]> well, at least FDA managed to ban arsenic as food ingredient about a 100 years late
09:42 < hprmbridge> Eli> This is something I think about. But also, do we allow drugs that didn't go through phase 2 and 3 to be paid for by medicare? It seems like there should be the option to sell phase 1 only to the market. But, medicare should wait to pay until after phase 3.
09:43 < hprmbridge> Eli>  https://cdn.discordapp.com/attachments/1064664282450628710/1138875111315878009/image.png
09:44 < hprmbridge> Eli> I was thinking about how to solve Erooms law:
09:44 < hprmbridge> Eli> 1) Put every humans medical data in the cloud so that we can draw associations with a massive n-values. All your genomic data, every medical test, your blood results, all in a freely accessible, open online format for anyone to access and analyze. This would save more lives than just about anything society could do, barring we all decided to start eating, sleeping, and exorcizing healthy, which
09:44 < hprmbridge> Eli> seems unlikely to happen soon.
09:44 < hprmbridge> Eli> 2) Put every person who receives government medical support into a human trial. You want free healthcare? You're going to contribute your body to science to help future generations. Need cancer treatment? You're going to be in an experiment where you test new treatments/protocols, and you're payment is free healthcare, but that's it. You're not getting paid thousands of dollars by pharma companies
09:44 < hprmbridge> Eli> to take part in a trial. If socialized medicine is actually about better health outcomes, this would be the way to do it. This would significantly lower the barrier to entry for new drugs. It would massively increase n-values for pharma companies to ensure safety and efficacy.
09:45 < Urchin[emacs]> socialized medicine is about actually using the life saving stuff that exists
09:52 < fenn> i would like to point out that "snake oil" was originally derived from chinese water snakes, which were high in omega-3, and used for centuries in chinese traditional medicine. it was only when western hucksters began selling fake snake oil made of mineral oil that it got a bad reputation
09:53 < fenn> "Stanley faced federal prosecution for peddling mineral oil in a fraudulent manner as snake oil."
09:54 < fenn> we don't need an FDA for preventing fraud
09:55 < fenn> the FTC defined the elements of deception cases. First, "there must be a representation, omission or practice that is likely to mislead the consumer."
09:56 < fenn> "the representation or omission must be a material one—that is one that would have changed consumer behavior."
10:00 < Urchin[emacs]> that's interesting information
10:00 < fenn> (the fact that he was charged by the bureau of chemistry which later became the FDA is not especially relevant)
10:01 < Urchin[emacs]> also, the typical mineral oil has laxative effects, AFAIK
10:01 < Urchin[emacs]> that's one uncomfortably "explosive" fraud ;)
10:02 < fenn> i think it was applied externally
10:04 < Urchin[emacs]> hopefully
10:04 < Urchin[emacs]> that reminds me to lubricate my knives with mineral oil...
10:04 < Urchin[emacs]> anyways...
10:07 < fenn> "The earliest medicinal patent was granted in 1698, for Epsom Salts. Patents only applied to the medication, not to its packaging, therefore competitors were free to copy a medication's packaging in an effort to increase their sales at the expense of the patented product."
10:07 < fenn> so the net effect is that the patented medicine is used even less, and the consumer is confused even more
10:07 < Urchin[emacs]> later things shifted to trademarks being the selling point
10:08 < Urchin[emacs]> especially "Aspirin", by Bayer/IG Farben, depending on era
10:08 < Urchin[emacs]> history of IG Farben is interesting, but grim
10:09 < fenn> yeah yeah, literally everyone's guilty everywhere. you are all damned to hell for eternity. happy now?
10:15 < muurkha> nmz787: I don't think people should take ineffective drugs, just that the involvement of the FDA in that process is counterproductive
10:16 < muurkha> Eli: exposing people's medical data to future Maos, Hitlers, and Pol Pots would cost lives too
10:17 < muurkha> hard to know which effect would dominate
10:17 < CNOT> dont eat used cat litter
10:18 < fenn> there's a massive campaign of denial that humans are different. someone might analyze all that data and discover inconvenient facts
10:18 < fenn> and then what
10:18 < fenn> it would be anarchy. cats and dogs living together
10:19 < muurkha> inconvenient facts such as who has genetic diseases that would consign them to the death camps
10:19 < fenn> right
10:19 < muurkha> under the new laws
10:20 < fenn> or just ineligibility for government sponsored child support
10:20 < muurkha> or mandatory sterilization
10:21 < muurkha> historically death camps and mandatory sterilization have been applied to this kind of thing much more often than ineligibility for government sponsored child support
10:21 < fenn> citation needed
10:21 < muurkha> https://en.wikipedia.org/wiki/One-child_policy#Enforcement
10:22 < muurkha> https://en.wikipedia.org/wiki/Compulsory_sterilization
10:22 < muurkha> https://en.wikipedia.org/wiki/Compulsory_sterilization#United_States
10:24 < muurkha> also, belonging to the wrong ethnic group, though that's probably something you can't hide any longer:
10:24 < muurkha> > Between 1970 and 1976, Indian Health Services sterilized between 25 and 42 percent of women of reproductive age who came in seeking healthcare services.
10:24 < fenn> from your first link, "Cash payments or other material rewards and fines acted as incentives, increasing the number of participants. ...  mandatory sterilization occurred after the birth of the second or third child."
10:25 < fenn> it sounds like most people who were sterilized during the one child policy did so voluntarily?
10:26 < muurkha> arguably people who get sterilized to avoid a fine are doing so voluntarily, in the same way that people who join the army to avoid imprisonment do so voluntarily
10:26 < muurkha> but generally it's important to recognize the coercion involved when discussing it
10:27 < fenn> don't tell me you're one of those people that says a fine and the lack of a reward are the same thing
10:28 < muurkha> I'm saying the opposite: that there is a major, important difference between them
10:28 < fenn> and all this is in the context of massive financial enforcement of the one child policy
10:28 < muurkha> because you seem to be saying that sterilization to avoid a fine is "voluntary"
10:29 < fenn> i do think there is a difference
10:29 < fenn> in china it doesn't matter though, because in the end they will sterilize you
10:30 < muurkha> well, not anymore
10:30 < fenn> hmph
10:30 < fenn> there's all kinds of nonsense still going on
10:32 < muurkha> yes.  and privacy is important to defend against it.  we need privacy not because we are ashamed of our secrets but because other people's judgment and intentions are not always flawless
10:34 < muurkha> as an illegal immigrant living in a province where programming without a license is also illegal, I have perhaps a special sensitivity to this issue
10:35 < muurkha> the most important factor for a successfully posthuman future is protecting the posthumans from the humans
10:36 < hprmbridge> alonzoc> Tbh I think some degree of eugenicist logic is applicable. There are real difference between people we are not all born as clean slates. However "negative" eugenics is a flawed approach and in the past was done for flawed reasons, in the modern world genetic screening of fertalized eggs for IVF along with genetic counciling of new couples are all useful and should be done.
10:36 < hprmbridge> alonzoc> Like yeah "I" wouldn't have been born if my parents had undergon genetic counciling and decided to do IVF and genetic screening (wasn't a option at the time), however a person would have been born occupying the same space in the world as me, probably having the same name. But the eggs being screened before birth aren't people yet
10:36 < hprmbridge> alonzoc> tag: imo
10:37 < hprmbridge> alonzoc> Like if genetic counciling was honest with potential couples about the odds of genetic illness in their children they could make informed decissions and the high risk groups would all mostly opt for IVF, which would have the effect of almost eliminating genetic illness in a few generations
10:38 < Urchin[emacs]> bleh, you can still have mutagenic environmental factors
10:38 < Urchin[emacs]> what a lovely cesium rain ...
10:39 < hprmbridge> alonzoc> True, but stuff like heritable genetic illnesses that aflict my family would be gone
10:39 < Urchin[emacs]> or some nazi decides to turn things around and sells mutagens to random pregnant women
10:39 < Urchin[emacs]> that's an actual historical example
10:40 < Urchin[emacs]> https://en.wikipedia.org/wiki/Thalidomide
10:40 < muurkha> maybe; plausibly the heritable genetic illnesses that persist persist because of some evolutionary advantage
10:40 < hprmbridge> alonzoc> This isn't compeltely far fetched
10:40 < muurkha> Tay-Sachs and sickle-cell anemia are two clear examples there
10:42 < muurkha> hmm, maybe Tay-Sachs is still controversial and not a clear example: https://en.wikipedia.org/wiki/Societal_and_cultural_aspects_of_Tay%E2%80%93Sachs_disease#Controversy_over_heterozygote_advantage
10:43 < hprmbridge> alonzoc> Or i'm thinking immune disregulation and intelligence, you've got CCR5 dropout mutations for one. While there's also the fact that IQ measures for the Ashkenazi jewish population is higher than the baseline population by about 1sd, and they have a bunch of gentic illnesses, a lot of that is small genepools and founded effects but it's not properly studied and I am sus from anecdotal evidence i've
10:43 < hprmbridge> alonzoc> seen
10:43 < hprmbridge> kanzure> what we need is to set global fertility rates to 10% for a while
10:43 < hprmbridge> kanzure> and the term you are thinking of us "coercive eugenics" btw
10:43 < hprmbridge> alonzoc> Not really, we are already gonna hit a demographic crisis in a while or go
10:44 < muurkha> 10% would probably require selecting the next generation to be almost entirely female
10:44 < hprmbridge> kanzure> I think we can pay people and have millions of births pretty easily.
10:44 < muurkha> or some even more extreme biological intervention
10:44 < hprmbridge> kanzure> this is not hard
10:44 < hprmbridge> kanzure> people like doing it
10:44 < muurkha> people like doing it to the tune of 4%, not 10% annual population growth
10:45 < hprmbridge> alonzoc> Well kanzure is on their way to make a GMO army using artificial wombs it seems after first killing off human fertility...
10:45 < muurkha> you could get to 8% by having a population that's almost entirely female
10:45 < hprmbridge> kanzure> also, do more pregnancies of multiples.
10:45 < hprmbridge> kanzure> you are not thinking creatively
10:45 < hprmbridge> alonzoc> ohh, kanzure is talking 10% growth rate. Not slowing down
10:45 < muurkha> yeah, both artificial wombs and multiple pregnancies are plausible interventions
10:46 < hprmbridge> kanzure> multiples is way more plausible
10:46 < hprmbridge> alonzoc> sorry that pattern matched  to the whole "stop overpopulation" argument
10:46 < fenn> multiples is dumb
10:46 < muurkha> well, multiples is easily achievable
10:47 < muurkha> but it has limits and serious drawbacks
10:47 < fenn> way more complications, resource contention, harder on the mother
10:47 < hprmbridge> kanzure> it may be dumb but this wouldn't have happened if we didn't let reproduction rate slide which was also dumb.
10:48 < fenn> we lost the memetic war
10:48 < muurkha> Malthusian overpopulation is a serious risk in the absence of progress
10:48 < muurkha> the absence of progress is also a serious risk
10:49 < muurkha> I mean, we are observing a lot of absence of progress right now
10:49 < hprmbridge> alonzoc> Yeah a lot of that is because of industrialized society isn't one that is conducive to the formation of stable couples nor those couples becoming sufficiently confident and stable for them to have children
10:49 < hprmbridge> alonzoc> like south korea are trying a bunch of interventions and they're not working
10:49 < fenn> most people don't want progress, they want "progress"
10:49  * fenn gestures ineffectually
10:49 < Urchin[emacs]> I don't think we'll be able to maintain this level of population for much longer
10:50 < fenn> see
10:50 < fenn> even urchin has succumbed to postmodernism
10:50 < muurkha> but once you have O'Neill cylinders in mass production, overpopulation ceases to be a worry for several more orders of magnitude
10:50 < Urchin[emacs]> it's not that it's impossible fundamentally, but we've taken a particular path that precludes the viability
10:51 < hprmbridge> alonzoc> Reee, You technocratic techno-fetishist! Don't you know that technology can't solve social problems and we need social solutions! (some writer for any modern news outlet)
10:51 < Urchin[emacs]> muurkha: you would have had to seriously start on that ages ago, we're late to that particular party
10:51 < muurkha> nah, we'll get there.  exponential growth is surprising
10:51 < fenn> technology can solve almost all social problems
10:51 < Urchin[emacs]> muurkha: nah, we're not working on that
10:52 < Urchin[emacs]> not really
10:52 < muurkha> remember all the people in march 02020 who were asking why covid was such a big deal if only like 1000 people were sick?
10:52 < Urchin[emacs]> fenn: that's a view I used to have, no more
10:52 < fenn> please don't conflate corporate power with technology itself
10:53 < Urchin[emacs]> fenn: I'm not
10:53 < muurkha> Urchin[emacs]: you'll probably enjoy reading my notes on this problem in https://dernocua.github.io/notes/backward-exponential.html
10:53 < hprmbridge> alonzoc> Population maybe, but the current issue is that our self-"stabilization" of population is leading to demographic collapse. I'd be okay with a stabalised population but the issue is the path there involves going through a more unsustainable region despite the population being lower in population
10:54 < Urchin[emacs]> we're going to have a demographic collapse anyway
10:54 < muurkha> .t
10:54 < muurkha> oh, botless still
10:54 < Urchin[emacs]> current agricultural practices practically guarantee it
10:54 < hprmbridge> alonzoc> True, it's something we need to figure out how to adapt around similarly to variations in climate
10:55 < muurkha> Urchin[emacs]: if grain yields go down perhaps people will just eat less meat?
10:55 < fenn> show me on the diagram where we can't use technology to fix social problems: https://www.mdpi.com/mti/mti-04-00038/article_deploy/html/images/mti-04-00038-g001.png
10:56 < Urchin[emacs]> muurkha: the bigger problem is that we're eating fossil fuels (thanks to IG Farben, BTW)
10:56 < muurkha> Urchin[emacs]: oh, I used to believe that, but then I investigated, and it turns out to be false
10:57 < muurkha> more specifically the energy output of farmland in the form of food calories vastly exceeds the fossil-fuel energy input in the form of things like tractor fuel, fertilizer, and pesticides, except in the case of crops like lettuce which have little caloric value
10:57 < Urchin[emacs]> muurkha: and the nitrogen fixation bottleneck?
10:58 < muurkha> a classic paper on this is https://www.mdpi.com/1996-1073/2/1/1/htm which finds about four joules output for one joule input
10:58 < Urchin[emacs]> that's the eating fossil fuels part
10:58 < muurkha> as it turns out, the energy required for nitrogen fixation is a small fraction of that one joule input
10:59 < hprmbridge> alonzoc> Harber go brr, it's pretty efficient tbh the issue of how much energy goes in is purely scale related. Also it doesn't rely on fossil fuels any energy source works. And if we all have a massive famine the total amount of nitrogen we need is reduced
11:00 < hprmbridge> alonzoc> Yeah ikr, it's mostly just a mantra by the art history people who are annoyed by how much tech has come to be the main driving force of social change and development
11:00 < hprmbridge> alonzoc> I kinda agree on the issue with it being cargo culted
11:00 < muurkha> well, some hydrogen source is needed
11:01 < fenn> biological nitrogen fixation is a thing
11:01 < fenn> vastly underfunded if you ask me
11:01 < hprmbridge> alonzoc> defo, there's some promising work on it
11:01 < muurkha> is there a biological nitrogen fixation system that's potentially competitive with Haber–Bosch?
11:02 < fenn> in what metric?
11:02 < muurkha> energy input
11:02 < docl> soybeans?
11:02 < fenn> is direct solar radiation considered energy input?
11:02 < muurkha> yes
11:03 < hprmbridge> alonzoc> it doesn't really need to be tbh, if you can farm previously non-viable land it's all profit
11:03 < muurkha> all land is viable to put solar panels on
11:03 < muurkha> then you can use them to drive Haber–Bosch
11:03 < fenn> probably not any time soon then. you'd need to construct a nitrogen fixing food web from a methanotrophic foundation
11:04 < fenn> which is another severely underfunded area
11:04 < fenn> photosynthesis is just very inefficient
11:04 < fenn> did i mention we can use technology to solve problems?
11:05 < fenn> maybe nitrogen fertilizer is a non issue
11:05 < hprmbridge> alonzoc> yeah but you then have grid overhead, and infrastructure problems, etc. You have much lower society wide overheads with self-contained biotech solutions, they're less efficient but they won't fall into disrepair when a global fammine hits leading to issues with bootstraping
11:05 < hprmbridge> alonzoc> Also solar panels have an additional manufacturing cost etc
11:06 < hprmbridge> alonzoc> it's all good but a self-contained plant that can fix it's own nitrogen and grow in mostly infertile soil is... well immensely useful
11:06 < fenn> 20% efficient solar panels don't need water, pesticides, harvesting, fragile goods handling, refrigeration
11:06 < Urchin[emacs]> people have been looking into combining solar panels and agriculture, there are advantages in shading plants from harsh sunlight
11:06 < fenn> 0.2% soybeans convert nitrogen at much less efficiency
11:07 < hprmbridge> alonzoc> Tbh tho i'm more interested in closed loop or aprox-closed loop systems like aquaponics with waste recycling
11:07 < fenn> mayb 0.02% solar energy goes to rhizopus nodules
11:07 < Urchin[emacs]> there are good news, but if you're looking for an optimist, it's not me
11:07 < fenn> less actually
11:08 < fenn> anyway there are strains of azotobacter that can be cultured in aerobic conditions
11:08 < hprmbridge> alonzoc> Yeah it'd be nice if you could power cells directly off the grid. give me a bioengineered blob that grows a potato given nutrients and a connection to a power source
11:09 < hprmbridge> alonzoc> who needs "leaves" and "roots" and all that stuff that burns energy and drops efficiency
11:09 < fenn> more like cheese than a potato i'd guess
11:09 < fenn> "textured vegetable protein" shows that you can make something sorta meat like
11:09 < hprmbridge> alonzoc> Yeah I guess, I just kinda like potatoes they're neat
11:10 < hprmbridge> alonzoc> great on a roast dinner
11:10 < fenn> yes potatoes are important
11:10 < fenn> i'm somewhat dismayed that it's 2023 and we still can't buy fusion powered synthetic food
11:11 < fenn> this was supposed to happen in the 1980s
11:11 < hprmbridge> alonzoc> Being able to grow blocks of cheese from a power source and base nutrients would also be useful
11:11 < fenn> there are two synthetic food brands, calysta feedkind and solar foods solein
11:11 < Urchin[emacs]> eh, there's plenty of fusion
11:11 < Urchin[emacs]> in the Sun
11:12 < fenn> there's none on earth because of politics
11:12 < fenn> pork barrel politics of the big science variety
11:12 < Urchin[emacs]> it's fucking hard as well
11:14 < fenn> rare earth barium copper oxide superconductors were discovered in 1987
11:14 < fenn> they have a quench field strength high enough to enable construction of a reasonable sized tokamak
11:16 < fenn> magnesium diboride fibers can handle 55 T
11:16 < fenn> that's a lot
11:16 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
11:16 < hprmbridge> alonzoc> ITER is a dumb a project, like I like my tokmaks but international projects are always slow and expensive and take forever because of bikeshedding
11:17 < fenn> does wikipedia do some kind of weird personalized search results now?
11:18 < fenn> maybe it's just a coincidence that "high temperature superconductivity" is the only article with the phrase "neon boiling point"
11:20 < fenn> alonzoc, the point is it doesn't have to be a huge international project. scaling laws go with magnetic field strength to the fourth power. ITER is huge because 1) it uses shitty superconductors with a weak field strength, and 2) it's a jobs program designed to waste vast quantities of money in certain locations
11:20 < CNOT> wikiped does not do that fenn
11:21 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
11:29 < hprmbridge> alonzoc> Yeah I agree, I was mostly just miffed that ITER is such a waste
11:46 -!- mrdata [~mrdata@user/mrdata] has quit [Ping timeout: 260 seconds]
11:46 < fenn> SPARC (vc funded tokamak) "on schedule for operation in 2025"
11:47 < fenn> this is basically, "what if we did the obvious straightforward thing" using ReBCO superconducting wires
11:48 < fenn> it seems likely the war will mess things up, since the superconducting tape was being produced in russia
11:50 -!- mrdata [~mrdata@135-23-182-55.cpe.pppoe.ca] has joined #hplusroadmap
11:54 < fenn> https://newatlas-brightspot.s3.amazonaws.com/dc/9a/603031e442f284dcbe6f0ab6265a/mit-fusion-magnets-02-press.jpg
11:55 < fenn> that's only one magnet segment but it gives you a sense of scale
12:01 -!- mrdata [~mrdata@135-23-182-55.cpe.pppoe.ca] has quit [Read error: Connection reset by peer]
12:11 -!- mrdata [~mrdata@user/mrdata] has joined #hplusroadmap
12:20 < muurkha> you'd think that of the 11_881_376 five-letter acronyms they could come up with one that isn't a well-known high-tech project that in all likelihood several of the team members had used in the past
12:20 < muurkha> (using English letters; number may vary by alphabet)
12:21 < muurkha> wikipedia's privacy policy does not allow weird personalized search results
12:23 < muurkha> alonzoc: you don't have grid overhead and infrastructure problems if you locate your Haber–Bosch plant near the solar panels.  back-of-the-enveloping: 10km radius, 1000W/m² nameplate irradiance, 21% nameplate efficiency, 25% capacity factor (typical for places like Arizona) gives you 6 terawatts average around the clock
12:24 < muurkha> existing Haber–Bosch plants have power consumption orders of magnitude smaller than that
12:24 < muurkha> so there is no need to transmit the solar power more than a single-digit number of kilometers to run a Haber–Bosch plant
12:25 < muurkha> if we scale down to a more reasonable 1 km radius then it's 60 gigawatts
12:25 -!- lkcl [lkcl@freebnc.bnc4you.xyz] has quit [Quit: BNC by #bnc4you]
12:25 < muurkha> solar panels are indeed expensive to buy but are quite stable once they're installed
12:25 -!- lkcl [lkcl@freebnc.bnc4you.xyz] has joined #hplusroadmap
12:26 < muurkha> and producing (Fenn's estimate) three orders of magnitude more ammonia per joule can compensate for a multitude of sins
12:27 < muurkha> alonzoc: indoor hydroponic marijuana farms are working pretty hard on your "bioengineered blob that grows a potato given nutrients and a connection to a power source" although of course hemp isn't potatoes
12:29 < muurkha> also, Dutch greenhouses, although of course they get most of their energy input from the sun rather than from horticultural LEDs
12:29 < muurkha> they do use horticultural LEDs, but only when it's not daytime
13:02 < muurkha> anyway I think the page of my calculations I linked above is relevant to the question of how soon you have to start to get enough O'Neill cylinders to avoid Malthusian problems
13:05 < muurkha> as you can see, the answer is that once you have anything approaching a biological growth rate for your fabrication system, even a couple of decades is plenty of growth time
13:06 < fenn> can you cut power to a haber-bosch plant for ~18 hours of the day
13:07 < fenn> the point isn't to grow plants on artificial light, it's to directly power cellular metabolism with electricity
13:07 < fenn> the less intermediary steps the better
13:08 < fenn> reverse https://en.wikipedia.org/wiki/Microbial_electrosynthesis
13:08 < fenn> uh. reverse microbial fuel cell
13:10 < fenn> may need to take a detour through methane
13:11 < muurkha> fenn: mostly you need thermal energy, which is easy to store
13:11 < muurkha> for Haber–Bosch
13:12 < muurkha> in general, process plants do not like to have their power turned on and off
13:12 < muurkha> their efficiency and purity depends on operating in steady state
13:12 < fenn> using your numbers for solar power, i get 150MW (1km radius) and 15 GW (10km radius)
13:14 < fenn> fission powered haber-bosch makes a lot of sense. you only want thermal power anyway, and the product is easy to transport
13:15 < fenn> and it gets rid of the "we're just eating petroleum" nonsense
13:16 < fenn> i wonder if all this excess ammonia is breeding large numbers of ammonia oxidizing soil bacteria
13:17 < fenn> oh i forgot to mention the thermochemical hydrogen cycles
13:18 < fenn> anyway you don't need turbines or generators
13:24 < fenn> sorry, 165MW and 16.5GW
13:26 < fenn> anyway you won't get 100% fill factor
14:01 < muurkha> thank for checking the calculation; let's see
14:02 < muurkha> 1km radius is 3.1 million m², which is 3.1 gigawatts raw peak
14:03 < muurkha> 21% efficiency times 25% capacity factor is 5.3%
14:03 < muurkha> 5.3% of 3.1 GW is indeed 160 MW and not 60 GW as I said
14:04 < muurkha> oh, I see my error.  due to user interface errors on my part, I was dividing by 21% and 25% instead of multiplying
14:05 < fenn> if you wanted to solve for a variable in an equation without doing manual symbol manipulation, how would you do it?
14:06 < muurkha> 150 or 160 MW is still enough for a pretty good-sized process plant I think.  but you might want to transmit the power over 2km of wires instead of like 300m as I was thinking
14:06 < fenn> not partial differentials or difficult stuff, just being lazy
14:06 < muurkha> still, it's not the kind of grid costs I understood alonzoc to be referring to me
14:06 < muurkha> I would open Gnumeric and put a value for the variable in one cell
14:07 < muurkha> and the equation in another cell
14:07 < muurkha> and use goal seek
14:07 < muurkha> on the Tools menu
14:07 < fenn> is there not an easy to use symbolic math system still
14:07 < muurkha> s/to me/ to/
14:07 < fenn> i could never get mathematica to do what i wanted
14:08 < muurkha> sympy is pretty reasonable but I don't know that it's an HCI improvement over Mathematica
14:08 < muurkha> but I interpreted your request as not wanting to use a symbolic math system
14:08 < fenn> understandable
14:08 < fenn> i normally do the manual method of "goal seek"
14:09 < muurkha> I wrote a simple calculator not long ago that incorporates goal-seek
14:09 < fenn> i.e. tweak the variable until it works
14:09 < muurkha> http://canonical.org/~kragen/sw/dev3/rpneact.py\
14:09 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
14:09 < muurkha> http://canonical.org/~kragen/sw/dev3/rpneact.py
14:10 < muurkha> the example there is that you type
14:11 < muurkha> V R/I! V I*P! 240V! 100R! P 100 - R <>
14:11 < muurkha> which defines I = V/R, P = I*V, V=240, R=100
14:11 < muurkha> and then uses an optimization algorithm to find a value of R which gets P-100 to approximate 0 as closely as possible
14:12 < muurkha> I never use it
14:12 < muurkha> it's sort of like a one-dimensional spreadsheet
14:12 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 246 seconds]
14:13 < fenn> what is the ! operator?
14:13 < muurkha> "define" or "store"
14:14 < muurkha> you can add Enter in between arbitrary tokens in that line
14:14 < fenn> what is the space character for?
14:15 < fenn> i'm confused by the discrepancy of: V I*P! and 240V!
14:15 < fenn> shouldnt the first one be I*PV!
14:16 < muurkha> multi-letter variable names are allowed
14:16 < muurkha> so VI would be a two-letter variable "VI", while "V I" is two variables
14:16 < muurkha> * then multiplies them (see the name of the program)
14:16 < fenn> oh i messed up reading the RPN, i get it now
14:16 < muurkha> and P! then defines P to be V*I
14:17 < fenn> blech
14:17 < muurkha> it's surprisingly powerful for such a quick hack, but as I said, I never use it
14:18 < docl> nifty!
14:20 < muurkha> apparently I wrote it over the course of about 10 hours one Saturday morning in 02021
14:21 < muurkha> as long as you leave expressions on the stack, they are redisplayed with their values after each line of input
14:21 < fenn> :) http://canonical.org/~kragen/sw/dev3/hackonfuse/
14:22 < muurkha> what the fuck
14:23 < muurkha> I think I must have written this by interchanging parts of an actual HN page
14:24 < muurkha> this is amazingly hilarious
14:24 < fenn> it looks like an april fools thing
14:26 < muurkha> like item #5 has "police informants" in place of "contact tracing", while item #6 has "contact tracing" in place of "facial recognition", and item #4 has "facial recognition" in place of I think "sequence diagrams"
14:26 < muurkha> which I moved down to item #23
14:29 < fenn> the first one i really read was "pitching your early stage fake app" which is perfect for HN
14:29 < fenn> and of course europe should go away
14:30 < muurkha> I just submitted it to HN
14:31 < muurkha> btw if you hate Mathematica you'll probably dislike https://asciinema.org/a/389974 too
14:31 < muurkha> which is an interactive keystroke-driven computer algebra system
14:32 < fenn> alg.py for posterity
14:33 < muurkha> yes, sorry
14:33 < muurkha> http://canonical.org/~kragen/sw/dev3/alg.py to be exact
14:33 < muurkha> another quick one-day hack
14:33 < fenn> the problem with mathematica was something like, i would type in a reasonable expression, then do solve(x) and it would crap itself
14:34 < fenn> instead of solving for x
14:37 < fenn> asciinema should display keystrokes
14:39 < muurkha> yes, arguably alg.py should also display keystrokes
14:39 < muurkha> is this the kind of thing you're talking about? https://nbviewer.org/url/canonical.org/~kragen/sw/dev3/sympy-solve.ipynb
14:40 < fenn> yes
14:41 < muurkha> (it's finding zeroes)
14:41 < fenn> right
14:41 < muurkha> can you run that in your own Jupyter notebook?
14:41 < fenn> that's the sort of UI i would expect
14:42 < fenn> um, jupyter has gained a lot of weight since i last tried, let me see
14:42 < fenn> and now i get to decide whether to install this with pip or apt
14:42  * fenn sweats nervously
14:43 < muurkha> I didn't know about sympy.solve
14:52 < fenn> of course it's a python clusterfuck
14:52  * fenn sighs
14:59 < muurkha> I think I just did apt install jupyter-notebook python3-sympy python-sympy-doc
14:59 < muurkha> but it's been a while
14:59 < fenn> i went with pip because i have learned my lesson
15:00 < muurkha> sounds like it didn't work out
15:06 < fenn> http://fennetic.net/irc/sympy_notebook.png
15:07 < fenn> now how do i make sure that jupyter isn't letting everybody on the LAN pillage all my files
15:24 < fenn> hm interesting that the "new" jupyterlab interface is more like a good old fashioned desktop GUI, vs the "old" notebook interface which is all mobile touchscreen wasted space big buttons for fat fingered children
15:26 < hprmbridge> alonzoc> Tbh I don't really like any of the jupyters. I've used it for sagemath stuff, but tbh I've had much better ux with org-mode babel. For one the plaintext nature means you can easily put it under git
15:26 < hprmbridge> alonzoc> But notebooks in general are really neat
15:27 < fenn> jupyterlab doesn't quite get the equations right unfortunately
15:28 < hprmbridge> alonzoc> How's that?
15:32 < fenn> just misaligned glyphs, gaps in the square root symbol, font looks off somehow: http://fennetic.net/irc/sympy_jupyterlab.png
15:33 < hprmbridge> alonzoc> What's it using to render the equations
15:33 < fenn> they both use mathjax
15:34 < hprmbridge> alonzoc> Huh
15:39 < fenn> sympy does generate an image too so maybe the janky mathjax rendering has an image pasted over it that i don't see in the HTML inspector for some reason
15:53 < fenn> one wonders why this isn't built in, auto converts to a format suitable for version control https://github.com/mwouts/jupytext
16:37 < muurkha> it doesn't look too bad
16:37 < muurkha> i don't think jupyter defaults to file pillaging
16:38 < muurkha> oh, jupyterlab does look bad
16:38 < muurkha> maybe it's failing to load a font?
17:26 -!- codaraxis [~codaraxis@user/codaraxis] has quit [Ping timeout: 256 seconds]
18:20 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
18:20 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
18:29 -!- Urchin[emacs] [~user@user/urchin] has quit [Killed (NickServ (GHOST command used by Urchin[emacs]`))]
18:29 -!- Urchin[emacs] [~user@user/urchin] has joined #hplusroadmap
19:13 < muurkha> alonzoc: that's an interesting note about org-mode
19:23 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
19:32 < muurkha> alonzoc: all the demos in https://youtu.be/kkqVTDbfYp4 https://youtu.be/D3FzMPZm7vY and https://youtu.be/1qOFYluebBg seem to only include textual output in the org-mode file
19:32 < Muaddib> [kkqVTDbfYp4] Emacs From Scratch #7 - Configure Everything with Org Babel (56:03)
19:32 < Muaddib> [D3FzMPZm7vY] Write Everything In Emacs Org Mode? You NEED This Plugin! (13:22)
19:33 < Muaddib> [1qOFYluebBg] Org-babel haskell notebook example (1:04)
19:33 < muurkha> and it looks like a huge amount of noisy text to type (and read) as overhead for each cell
19:33 < muurkha> I shouldn't say "in the org-mode file"
19:33 < muurkha> what I mean is "in Emacs's display of the org-mode file"
19:33 < muurkha> since I don't really want base64ed PNGs to go into the org-mode file, that's the big problem with Jupyter I want to get away from
19:35 < muurkha> what I really want from Jupyter is that when I'm editing something like https://nbviewer.org/url/canonical.org/~kragen/sw/dev3/fourier-correlation.ipynb
19:36 < muurkha> to have the shortest possible delay and the minimal possible distractions in between editing something like the `xlim(-250, 250)` line in the last cell and seeing the results in the plot below
19:37 < muurkha> Jupyter is pretty shitty there: I have to hit Ctrl-Enter to get the results, and then it deletes the old plot from the notebook before inserting the new one, which often (in the common case!) scrolls the notebook down, which is distracting in itself but also means that when the plot reappears it's partway off the screen and I have to scroll down to see the whole thing
19:40 < muurkha> but at least everything that shows up on YT suggests that org-babel would be even worse on this axis?
19:41 < fenn> having to press ctrl-enter seems like a good design decision
19:41 < muurkha> well, it's a shitty design decision, but in Python the alternative woudl be far worse
19:41 < muurkha> but, in org-mode, is there even a way to get a data plot to show up inline at all?  I see that you can do that for LaTeX equations
19:42 < fenn> the alternative i'm imagining is like autocomplete or progressive search but it runs arbitrary code every time you press a key
19:42 < muurkha> arbitrary side-effect-free code, yes
19:42 < fenn> how does a notebook know what your code does
19:42 < muurkha> and aborts it and restarts it when you change the code again
19:43 < muurkha> well, with Python it can't, at least not without a lot more infrastructure we don't have
19:44 < fenn> i've seen notebooks that download multi-gigabyte AI models an use real money to execute
19:44 < muurkha> that's why I say that with Python the alternative would be far worse
19:44 < muurkha> in https://youtu.be/ZMEcb2rpauU he does in fact make a plot
19:44 < Muaddib> [ZMEcb2rpauU] Introduction to Emacs Org Mode (13:46)
19:44 < muurkha> which he saves to a file with plt.saveplot('sine_plot.png')
19:45 < muurkha> and then views by manually adding a link to ./sine_plot.png to his org-mode buffer and opening the link
19:45 < muurkha> which then doesn't even open inline in the org-mode buffer, but in a new window
19:45 < muurkha> which, naturally, is too small, so you can't see the X-axis labels without manually scrolling it
19:46 < fenn> did you see jupytext
19:46 < muurkha> just your link
19:46 < muurkha> alonzoc: can current org-mode do better than this?
19:47 < fenn> well i don't know anything but it seems to address the embedded binary data problem
19:48 < muurkha> in https://youtu.be/YmQ1CYMz-OY we have the same problem of the plotting code hiding the plotted data in a file that you then have to manually open
19:48 < Muaddib> [YmQ1CYMz-OY] EmacsConf 2022: Health data journaling and visualization with Org Mode and GNUplot - David O'Toole (24:30)
19:49 < muurkha> at 10'56"
19:50 < muurkha> also in https://youtu.be/BbZbtgvwrI8 at 10'42"
19:50 < Muaddib> [BbZbtgvwrI8] ESS and Org Mode (14:02)
19:50 < muurkha> "we have some results down here.  we can use control-c, uh, control-o, to open it"
19:50 < muurkha> NO, FUCK, I HAVE A COMPUTER TO DO SHIT LIKE THAT FOR ME
19:52 < muurkha> what I actually want is something more like http://canonical.org/~kragen/sw/81hacks/autodiffgraph/
19:52 < muurkha> but not limited to what I could hack together in one evening at the office
19:54 < muurkha> like, if I have a data series or continuous function, I would like to see it plotted, by default.  and I want that plot to be part of the notebook I'm editing, not something I have to take additional explicit actions to access
19:56 < muurkha> (in this case I'm also automatically plotting its derivative by default, using automatic differentiation, but I don't mean to imply that that's the best choice for a general-purpose UI
19:56 < muurkha> )
20:27 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
20:31 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 240 seconds]
20:51 < hprmbridge> kanzure> google scholar can still find weird stuff https://twitter.com/literalbanana/status/1689420167024095232
20:53 < hprmbridge> kanzure> free energy minimization principle and in vitro neuronal cultures https://www.nature.com/articles/s41467-023-40141-z
20:56 < hprmbridge> kanzure> and a podcast episode regarding the subject from a coauthor https://www.preposterousuniverse.com/podcast/2020/03/09/87-karl-friston-on-brains-predictions-and-free-energy/
21:37 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 245 seconds]
21:38 < jrayhawk> i find it's least irritating to read friston's colleagues rather than friston directly
22:05 < muurkha> heh
22:17 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
23:07 -!- Urchin[emacs] [~user@user/urchin] has quit [Quit: ERC (IRC client for Emacs 27.1)]
23:37 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
23:40 -!- test_ [flooded@gateway/vpn/protonvpn/flood/x-43489060] has quit [Ping timeout: 245 seconds]
--- Log closed Thu Aug 10 00:00:24 2023