--- Log opened Sun Feb 26 00:00:55 2023
01:00 -!- cthlolo [~lorogue@77.33.23.154] has joined #hplusroadmap
03:29 -!- cthlolo [~lorogue@77.33.23.154] has quit [Read error: Connection reset by peer]
06:11 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has joined #hplusroadmap
07:10 < kanzure> https://dashboard.cohere.ai/playground/generate "should be able to run inference on 2x3090, which if you get used ones off of ebay is under $1200"
08:40 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
08:40 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
10:17 < L29Ah> https://nitter.lacontrevoie.fr/vita_dao/status/1629509297381203968 huh, VitaDAO comes to the nearby village
11:00 < fenn> is there actually any good reason to keep health data secret? all i can think of is stonk shenanigans like "the CEO is unhealthy" or people being embarrassed about a medical condition
11:03 < fenn> avoiding genetically targeted disease vectors seems a little too future-thinking for congress critters
11:03 < fenn> (targeted bioweapons)
11:05 < fenn> someone will try to scam you because you or a relative are in a hospital?
11:06 < L29Ah> russian congress critters made sending genetic material abroad for this very reason
11:06 < L29Ah> *illegal
11:06 < L29Ah> the ban doesn't really work tho
11:08 < L29Ah> fenn: generally speaking, information about your medical conditions can be interpreted as your vulnerabilities for nefarious uses
11:09 < fenn> but also for beneficial uses
11:10 < fenn> how are we going to get good genotype-phenotype correlations if everyone's data is secret?
11:10 < L29Ah> for example, if you have a weak heart, you're easy to chase; if you have an allergy, you can be incapacitated or outright killed with the right substances, and it would be hard to attribute it to a human culprit
11:10 < L29Ah> well, you can ask them nicely and they will share it
11:11 < L29Ah> allow cookies? ([Y]/n)
11:11 < fenn> aren't those sort of things generally known anyway? i mean i wouldn't want to accidentally kill someone with my peanut butter sandwich
11:11 < L29Ah> not until you offer your sandwich to an allergic person?
11:11 < fenn> when not sharing is the default, the vast majority of people won't share. also we have this huge bureaucracy overhead for managing protected health information, and it complicates research
11:24 < kanzure> there's a company going around paying people to take IQ tests and buy their genome sequence
11:25 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
11:25 < kanzure> with the idea that no existing genetic data set is measuring IQ that thoroughly
11:25 < kanzure> and if you want the best data, you gotta directly buy it up
11:25 < kanzure> what's worse is that some government genomic datasets are becoming harder to access because they are moderating the projects that are given access
11:26 < kanzure> (there was an article posted here a while back about some sort of access censorship that NCBI is doing?)
11:26 < fenn> to prevent exactly this kind of genetic discrimination i presume
11:26 < kanzure> no it was something dumber like preventing IQ research
11:26 < fenn> how DARE you do science on humans!
11:26 < fenn> that's what i meant
11:26 < kanzure> yeah what if we did science on humans and everything turned out ok?
11:27 < fenn> what if everything turned out a lot better than ok
11:28 < fenn> what if "lazy" or "immoral" or "criminal" people are actually literally sick, and they could be healed and turned into "good" and "pro-social" contributors
11:29 < fenn> what if 99% of people have problems like this and everyone is silently suffering
11:33 < kanzure> what if panicking about getting killed by AI is silly when people should be freaking out about age-related death or should we just silently suffer over that
12:00 < muurkha> fenn: there are a lot of cases where people are discriminated against because of aspects of their health
12:01 < muurkha> the most obvious case is that health insurance companies would love to set their premiums based on complete knowledge of your medical condition, and if you have enough more information about it than they do when choosing to buy health insurance, they become unprofitable
12:02 < fenn> but that's already illegal
12:02 < muurkha> some places, but making things illegal doesn't make them impossible even in those places
12:03 < muurkha> but also lots of people discriminate (including unfairly) based on things like whether you have HIV, whether you're on PEP or PrEP, whether you have hepatitis, what drugs you're addicted to
12:03 < muurkha> whether you've had an elective abortion, whether you've had a miscarriage
12:04 < muurkha> your mental health problems, what psychiatric drugs you're on
12:05 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1079494332387692626/crop.png
12:05 < muurkha> whether you've had one of your balls chopped off, or whether it never developed properly in the first place
12:06 < muurkha> whether you've been pregnant, especially if you don't evidently have a child today, even if you haven't miscarried or had an abortion
12:08 < muurkha> whether you've had male lactation or other intersex characteristics
12:09 < muurkha> so, lots of people have a strong interest in keeping certain data about their health private, in large part because *other* people are kind of assholes sometimes
12:10 < muurkha> consequently they will take measures to avoid disclosing that data if they think it will become public, and if this results in them being unwilling to share it with doctors who are trying to treat them, or to go to the doctors in the first place, they will get worse health care and may die early as a result
12:18 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
12:19 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has quit [Read error: Connection reset by peer]
12:22 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Read error: Connection reset by peer]
12:36 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
12:41 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Read error: Connection reset by peer]
12:42 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
12:52 < fenn> i agree that other people can be assholes, but if all of society cooperates to keep certain characteristics or health histories secret, then it's that much more "sensational" and "other" when info about a secret-worthy condition gets out to friends, family, employers, and so on. if people were familiar with these things on a regular basis, it's less likely to be a big deal
12:55 < muurkha> I've heard that before, and I think it's slightly true, but not enough to solve the problem
12:56 < fenn> i'm lumping all this into "people being embarrassed about a medical condition"
12:57 < muurkha> I mean being female, old, black, a child, or a burn victim who has lost most of their face is extremely not secret, and people continue to discriminate heavily against people with those characteristics
12:58 < muurkha> it's probably true that they would discriminate more heavily if people were somehow able to keep their age or sex a secret
12:58 < fenn> it happens on the internet
12:58 < muurkha> but that doesn't mean that they don't discriminate at all when they're conscious of being surrounded by people of different ages and sexes
12:59 < kanzure> this is boring
12:59 < muurkha> it is, but it's the problem medical progress is facing, so we need to figure out how to solve it
12:59 < fenn> homomorphic encryption to the rescue~
13:00 < muurkha> also, it's clearly inaccurate to describe someone being concerned that they'll be discriminated against because they're on PrEP as being "embarrassed about a medical condition"; they're gay, which is arguably not even a medical condition, and many gay people are not at all embarrassed about it
13:00 < fenn> is the answer just hacking the shit out of every medical database in the world and leaking it?
13:00 < fenn> i mean i find it hard to believe that anyone actually has good security
13:02 < muurkha> I explained above why that would make the situation worse rather than better
13:02 < fenn> there's also a lot of non-medical info that should be relevant to understanding the genotype-phenotype relationship, but that's also not available because genome data is not linked to social network profiles or whatever
13:02 < muurkha> at least with respect to other people being assholes
13:04 < muurkha> there would clearly be benefits (you could do research on the resulting data, and presumably new medical institutions would arise to fill the vacuum of health care for people who care about their privacy) but it seems obvious to me that the harms would be enormous, greater than the plausibly achievable benefits
13:04 < fenn> i'm holding your concerns about people being discriminated up against the ever present looming horizon of EVERY FUCKING HUMAN BEING DYING WITH 100% CERTAINTY
13:05 < fenn> so excuse me if i don't give a shit about stuff that's already happening anyway
13:05 < muurkha> they'll die sooner if they refuse to go to the doctor because they don't want their parents to find out they're gay
13:06 < fenn> and no, i don't think that understanding the genotype-phenotype relationship will magically solve aging, but it's a start
13:06 < kanzure> nah let's just all die
13:06 < kanzure> that seems like a ... reasonable answer.
13:08 < muurkha> also it won't matter very much what medical research discovers if there aren't experienced medical practitioners available to most people so they can apply the results
13:08 < fenn> that's a stretch
13:08 < kanzure> medical licensing is actually fake
13:08 < kanzure> you can just make up licensing bodies and people will believe it
13:08 < kanzure> they just go along with it
13:08 < muurkha> agreed, but medical skill isn't, and people acquire it as they practice (as well as when they study)
13:09 < fenn> medicine won't disappear because *some* closeted gay people don't go to a doctor for "reasons"
13:09 < muurkha> Mao got significant improvements in life expectancy from the "barefoot doctors"
13:09 < muurkha> almost everyone has *something* about their health they don't want to be public; you're very much an exception there
13:09 < fenn> did you just assume my privilege~
13:10 < muurkha> haha
13:10  * muurkha puts on fenn's privilege
13:10 < fenn> is there a sane culture i can move to
13:10 < muurkha> I wish
13:10 < fenn> i'm tired
13:10 < kanzure> you're stuck with us
13:11 < muurkha> in pretty much every culture people engage in aggression against each other and use privacy to be less vulnerable to that aggression
13:11 < kanzure> rorschach.jpg
13:11  * L29Ah throws utopia books at fenn
13:12 < muurkha> the Amish and Hutterites manage to keep things down, but they still have to deal with intra-community aggression
13:12 < muurkha> and they can only keep its level as low as they do because they keep the communities very small
13:12 < L29Ah> my favorite privacy-rejecting one is https://proza.ru/2013/09/08/930 et al
13:13 < muurkha> thank you, L29Ah!
13:13 < kanzure> fenn: would you accept less public data sharing in exchange for society investing in science and actual progress?
13:14 < fenn> L29Ah: could you summarize this? appears to be a novel
13:16 < L29Ah> hmmm
13:16 < fenn> some nerds set up a (presumably libertarian) nation in the pacific islands?
13:16 < L29Ah> anarchist libertarianish wingnuts building their state in oceania, having fun and beating up other states
13:17 < kanzure> something about.. gold.. and a battle at sea?
13:17 < kanzure> my bear babble isn't so good
13:17 < muurkha> is this a seasteading kind of thing or more a question of acquiring existing land?
13:18 < kanzure> redsprachen?
13:18 < L29Ah> muurkha: mostly latter
13:18 < L29Ah> kanzure: that's an adventure story, you know
13:19 < kanzure> kremlingo
13:21 < L29Ah> among, eh, the globally-recognized privacy-rejecting stories i can only recall Hellstrom's Hive, but that's kinda boring nowadays
13:22 < kanzure> i don't think fenn is arguing for the destruction of privacy
13:23 < fenn> i was always rooting for the borg
13:25 < fenn> i just haven't seen a justification for medical recordkeeping inefficiencies that's commensurate with the harms caused by it, and the lost opportunity from research that isn't possible because of the unavailability of data
13:25 < fenn> also it seems that "deanonymization" is a lie we tell to children
13:26 < fenn> er. "anonymization"
13:31 < L29Ah> i'm 80% sure the practice comes from some abrahamic religion text anyway
13:34 < L29Ah> > And whatsoever I shall see or hear in the course of my profession, as well as outside my profession in my intercourse with men, if it be what should not be published abroad, I will never divulge, holding such things to be holy secrets.
13:35 < L29Ah> https://en.wikipedia.org/wiki/Hippocratic_Oath
13:36 < muurkha> definitely pre-Abrahamic-era
13:36 < muurkha> I mean there were Abrahamics around but they were confined to a small part of the Middle East at the time
13:38 < muurkha> I'd be surprised if there wasn't something similar in Han China
13:53 -!- CVDFNSTR is now known as Mabel
15:49 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
15:50 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 246 seconds]
15:50 -!- stipa_ is now known as stipa
15:59 -!- dreda25 [~dreda@2604:3d08:6781:af00:fc4b:31a5:e3c7:4282] has joined #hplusroadmap
16:02 -!- dreda25 [~dreda@2604:3d08:6781:af00:fc4b:31a5:e3c7:4282] has quit [Client Quit]
16:39 < kanzure> hmmm https://nitter.unixfox.eu/gunsnrosesgirl3/status/1524354764481482754#m
16:40 < kanzure> "Goniurellia tridens, each wing has a detailed image of an ant" 
16:40 < hprmbridge> lachlan> Wow, that’s wild
16:41 < lsneff> It’s mind blowing that something like that can evolve
16:41 < lsneff> The feedback is so indirect
16:41 < kanzure> Macrocilix maia has an image of a beetle on each wing https://nitter.unixfox.eu/gunsnrosesgirl3/status/1517594072948617217
16:59 < superkuh> Neat.
18:14 < kanzure> doom https://www.lesswrong.com/posts/Aq82XqYhgqdPdPrBA/full-transcript-eliezer-yudkowsky-on-the-bankless-podcast
18:14 < kanzure> (you can safely ignore)
23:13 -!- Mabel is now known as MARVEST
--- Log closed Mon Feb 27 00:00:56 2023