--- Log opened Sun Nov 19 00:00:29 2023
01:38 -!- millefy [~Millefeui@anantes-651-1-211-3.w90-25.abo.wanadoo.fr] has quit [Ping timeout: 252 seconds]
02:02 -!- millefy [~Millefeui@anantes-651-1-211-3.w90-25.abo.wanadoo.fr] has joined #hplusroadmap
02:07 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
02:08 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
03:42 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
04:01 -!- Llamamoe [~Llamamoe@188.146.90.70] has joined #hplusroadmap
04:09 -!- ike8 [e8f913dbdf@irc.cheogram.com] has quit [Ping timeout: 256 seconds]
05:00 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
05:03 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
06:43 < hprmbridge> kanzure> https://twitter.com/kanzure/status/1726050854656934272
07:15 < kanzure> https://bitcoinmagazine.com/culture/how-extropian-quest-for-digital-cash-secured-our-trips-to-the-stars
07:24 < kanzure> anatomy of a cult https://twitter.com/kanzure/status/1726251316513841539
07:50 -!- A_Dragon [A_D@libera/staff/dragon] has joined #hplusroadmap
07:56 -!- A_Dragon [A_D@libera/staff/dragon] has quit [Quit: ZNC - https://znc.in]
08:16 < kanzure> the cult and the "personal hell realm" https://forum.effectivealtruism.org/posts/AjxqsDmhGiW9g8ju6/effective-altruism-in-the-garden-of-ends
08:45 < nsh> not another poor kid totalised by an ought
08:48 < nsh> six months.
08:49 < nsh> hard-earned cousinhood
08:59 -!- deltab [~deltab@95.154.230.49] has quit [Ping timeout: 240 seconds]
09:01 -!- deltab [~deltab@95.154.230.49] has joined #hplusroadmap
09:06 < kanzure> re: multi-agent AI futures, "The safe path to a better future is through adversarial equilibrium between powerful AIs, achieved by putting them in the hands of many, not few. Top-down AI power gradients would otherwise open the door to tyranny, which is a real danger underestimated by many." https://twitter.com/BasedBeffJezos/status/1721174753229353307
09:08 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1175845038090100736/F_T-YhaXgAEFre2.png?ex=656cb607&is=655a4107&hm=de0c59fd81cec6b878eb64dbdaca96a0dd2045c6491b32078c3df1522a2888f7&
09:19 -!- deltab [~deltab@95.154.230.49] has quit [Ping timeout: 255 seconds]
09:19 < docl> adversarial training is how they learn to lie and want things you don't
09:19 -!- deltab [~deltab@95.154.230.49] has joined #hplusroadmap
09:20 < docl> there might be ways to set up an equilibrium like that, but can't be too naive about how you go about it
09:24 < docl> in humans we have things like principles which when violated are politically costly, votes which force political costs to be imposed, constitutional laws which make extreme power grabs legally costly, courts that impose such costs. it's all very elaborate and apparently necessary to have something like freedom in a durable/scalable way
09:28 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
09:41 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
09:48 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
10:00 < muurkha> we haven't yet found a way to have something like freedom in a durable/scalable way
10:02 < hprmbridge> alonzoc> Yeah on the surface EA is a philosophy that makes sense, given your values pursue them in a rational way. The issue is figuring out your values and because we haven't solved that philosophical problem you end up repressing yourself to serve a movement which convinces you of it's own benevolence (and is also exploited by non benevolent grifters, reminds me of commune collapses).
10:02 < L29Ah> how to grift off EA?
10:03 < L29Ah> i can only come up with making effectiveness analysis for them at a premium
10:03 < hprmbridge> alonzoc> Join non-profit org and play internal politics wekk
10:03 < hprmbridge> alonzoc> *well
10:05 < hprmbridge> alonzoc> Also so many people don't know about the extropians, admittedly it's fairly fringe history. Cypherpunks are also mostly unknown even amongst crypto people
10:08 < muurkha> pretty sure all crypto people know about cypherpunks
10:08 < L29Ah> crypto™ people
10:08 < hprmbridge> alonzoc> I've met a few who don't many in the eth space don't
10:08 < hprmbridge> alonzoc> And I'm not speaking of just NFT bros
10:09 < muurkha> like I think it would be hard to find anyone who had ever published an IACR paper who didn't know about the cypherpunks
10:09 < hprmbridge> alonzoc> Okay people who are doing research and shit do, but there's a large contingent of "blockchain devs" who don't
10:09 < muurkha> sure, lots of people who *use* crypto don't know about cypherpunks, just like lots of people who use computers don't know about Alan Turing
10:10 < hprmbridge> alonzoc> I'm talking about software people, contract solidity programmers and the like
10:10 < muurkha> carrying a Samsung cellphone in your pocket doesn't make you a computer person, writing Solidity doesn't make you a crypto person
10:11 < muurkha> can we get some fucking gatekeeping please
10:11 < hprmbridge> alonzoc> I'd say if a person is employed in the crypto industry in a technical role they count as a crypto person. Also that last one is a bad comparison but yeah
10:11 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
10:12 < hprmbridge> alonzoc> I would like some more, I wish everyone in the crypto industry had to be able to write all the cryptographic primitives would be a great improvement in quality
10:13 < muurkha> there isn't a "crypto industry"
10:13 < kanzure> muurkha: even the IACR people, some of those might be rather insulated academics surely
10:13 < muurkha> the NSA isn't Fujitsu
10:14 < muurkha> kanzure: pretty sure they all know about the Bernstein case, Gilmore's role, etc.
10:14 < kanzure> this seems like something that could be surveyed and sort of interesting to go and ask
10:14 < muurkha> it's a pretty small community
10:14 < kanzure> do you read all the IACR papers? i stopped when it was taking up too much of my time. there's a few hundred each month.
10:14 < muurkha> no
10:15 < muurkha> I meant the conferences, though, not the preprints
10:15 < kanzure> https://eprint.iacr.org/
10:15 < kanzure> oh...
10:15 < muurkha> surely there's someone who's posted a preprint who doesn't know Bernstein from a hole in the ground
10:15 < muurkha> I realize now that I didn't actually *say* that
10:16 < kanzure> thanks to the failures of the english language nobody can prove that
10:16 < muurkha> and posting a paper on a preprint server does make it public, so it amounts to "publishing" it
10:16 < muurkha> but I meant a paper that's been accepted and included in conference proceedings
10:16 < L29Ah> nitpicking competition
10:18 < muurkha> anyway, there's surely lots of people writing Solidity who not only don't know Bernstein from a hole in the ground, they don't know *crypto* from a hole in the ground
10:18 < kanzure> L29Ah: who were you trying to grift and what is your goal precisely?
10:18 < muurkha> so I think it's dumb to call them "crypto people"
10:18 < hprmbridge> alonzoc> Depending on where you draw the line "crypto people" all know of the cypherpunks, but I was more alluding to the large quantity of employed devs in the ecosystems who don't. You might not call them crypto people but if their job is writing smart contracts then they're not just generalised tech people
10:19 < hprmbridge> alonzoc> Yeah they're more blockchain people than crypto people I guess, we should reclaim crypto meaning cryptography
10:20 < L29Ah> kanzure: i weren't, i was just wondering what use could there be from EA people given i can't convert them into egoism and don't plan moving to Africa
10:21 < hprmbridge> alonzoc> Apply for safety related grants if you do anything AI related?
10:21 < L29Ah> alonzoc: plenty of cryptocurrencies aren't blockchain-based
10:22 < muurkha> L29Ah: you mean like Chaum's old ventures?  are those still running in any form?
10:22 < hprmbridge> alonzoc> True, "trustless consensus algorithm people"
10:22 < hprmbridge> alonzoc> Doesn't have the same ring
10:22 < muurkha> are there blockchain-less consensus algorithms now?  Mimblewimble maybe?
10:23 < hprmbridge> alonzoc> There's avalanche which is kinda cool if you don't care about finality
10:23 < muurkha> I mean, blockchain-less consensus algorithms that don't rely on trusting all the participants
10:23 < muurkha> hadn't heard of avalanche, thanks
10:23 < hprmbridge> alonzoc> The paper is quite well written
10:23 < L29Ah> muurkha: DAG-based is all i'm aware of that have at least somewhat decentralized consensus
10:24 < muurkha> probably a blockchain also fails to provide finality
10:25 < hprmbridge> alonzoc> Not necessarily, the protocol used in Ethereum right now aside from being a horrible hybrid has absolute finality
10:25 < hprmbridge> alonzoc> And it's a blockchain
10:25 < muurkha> oh, interesting
10:25 < L29Ah> Montelibero uses Stellar, its decentralization isn't stellar though
10:25 < muurkha> I didn't realize
10:25 < hprmbridge> alonzoc> Nakamoto consensus is old news
10:26 < hprmbridge> alonzoc> Avalanche paper: https://assets.website-files.com/5d80307810123f5ffbb34d6e/6009805681b416f34dcae012_Avalanche%20Consensus%20Whitepaper.pdf
10:27 < hprmbridge> alonzoc> There's a proper published version somewhere but this explains it
10:28 < muurkha> thanks!
10:29 < hprmbridge> alonzoc> I have a place in my heart for avalanche style setups especially because synchronously working on a monolithic global state is kinda dumb. It's simple and easy to ensure properties, but in practice each account and contract has private state so you should be able to leverage separability to allow for faster higher throughput consensus in practical cases
10:30 < hprmbridge> alonzoc> However such a system would have similar problems to sharding
10:33 < muurkha> most of the time there aren't any other transactions in flight that would conflict
10:35 < hprmbridge> alonzoc> Yeah, so you can exploit that only transactions that interact with states that's under heavy contention would be slow in processing.
10:39 < juri_> media.ccc.de just listed my talk on youtube.
10:39 < muurkha> condolences
10:40 < muurkha> because youtube comments suck
10:46 < hprmbridge> alonzoc> Physics talk YouTube comments are the best
10:46 < hprmbridge> alonzoc> They always have someone with a 50yo man in their profile pic talking about the theory of everything he concocted based on the cereal box he just read
11:01 < docl> hmm, write a cryptocurrency app for african phones that makes them less poor and have EAs responsible to back the coin?
11:02 < docl> just off the toppa my head
11:02 < L29Ah> docl: not sure if that saves the most africans from death
11:03 < docl> well they aren't dying because they are too rich. a lot of them are walking long distances to carry water from a dirty stream
11:03 < L29Ah> at least there's a vague idea that undereducated people are shortsighted and don't treat money with due respect
11:04 < docl> you could make the money more intuitive for poor people
11:06 < muurkha> a slightly better justification is that poor people are poor because they're being constantly robbed, and money is more likely to get stolen than bed nets
11:12 < docl> have 100 (or whatever) units collateral distributed every month, borrow your spending money against that. if you manage to return to zero balance by the month end, you increment your points of credit by 1, else decrement. total borrowable amt = points credit x units collateral. since it's linear, the less points credit you have the more strongly you notice the results of not reaching solvency, and the 
11:12 < docl> more you have the less you worry about it, i.e. you can better afford to be generous / it makes sense to trade for other things like reputational gains and productive assets
11:13 < docl> (it's not real borrowing, you internalize all risk)
11:28 < kanzure> muurkha: there's my own blockchainless cryptocurrency https://webcash.org/
12:02 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
12:02 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
12:15 -!- ChanServ [ChanServ@services.libera.chat] has quit [shutting down]
12:19 -!- ChanServ [ChanServ@services.libera.chat] has joined #hplusroadmap
12:19 -!- ServerMode/#hplusroadmap [+o ChanServ] by calcium.libera.chat
12:22 < kanzure> claims to be some sort of fancypants research literature search engine https://soma.science/
12:41 -!- Guest85 [~Guest85@50.47.207.149] has joined #hplusroadmap
12:41 -!- Guest85 [~Guest85@50.47.207.149] has quit [Client Quit]
12:53 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
12:55 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
13:01 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
13:10 < docl> https://www.cell.com/molecular-therapy-family/molecular-therapy/fulltext/S1525-0016(00)90218-1 // Symbiosis Therapy: The Potential of Using Human Protozoa for Molecular Therapy
13:11 < docl> sadly not about literal human protozoa that contain human DNA and specialize into cells
13:15 < kanzure> "Sentinel cells programmed to respond to environmental DNA including human sequences" (for human facial feature fingerprinting) https://www.nature.com/articles/s41589-023-01431-1
13:16 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
13:16 < kanzure> docl: that's somewhat related to how i was talking about using toxoplasmosis as a chasis for brain therapeutics https://diyhpl.us/~bryan/irc/2017-02-03-beacon.pdf 
13:19 < kanzure> "Nougat: Neural optical understanding for academic documents" https://facebookresearch.github.io/nougat/
13:21 < kanzure> wonder if that would work on the chinese superlibrary dataset...
13:36 < docl> yeah toxo is protozoan, I didn't think of that
13:52 < kanzure> docl: why not use human stem cells as the platform instead of protozoa
14:10 -!- Llamamoe [~Llamamoe@188.146.90.70] has quit [Quit: Leaving.]
14:34 < kanzure> https://blog.miguelgrinberg.com/post/it-s-time-for-a-change-datetime-utcnow-is-now-deprecated
14:34 < docl> I'm currently thinking less demanding culturing environment -> cheaper experimentation
14:38 -!- ike8 [e8f913dbdf@irc.cheogram.com] has joined #hplusroadmap
14:51 < fenn> i can't believe that in 2023 people are optimizing computer code for not storing a single byte to represent timezone
14:56 < kanzure> i thought this is deprecating naive timezoneless dates?
14:57 < fenn> no they're just changing the names around to piss you off
14:57 < fenn> naive timezoneless dates will still exist
14:58 < fenn> obviously the solution is to make utcnow() return time in UTC
14:59 < fenn> i guess they don't want to introduce a semantic change, even though it used to do that before python 3
15:00 < fenn> the whole thing is stupid as hell if you ask me
15:03 < fenn> gpt4 was wrong about the definition of technological singularity; it's nothing to do with being uncontrollable or irreversible
15:04 < fenn> what matters is that the rate of change is increasing in a feedback loop, and the outcome is unpredictable
15:33 < fenn> it seems like the antidote to EA is giving people good things to do, rather than worrying about not doing things or bad things happening
15:33 < fenn> the reason they got into it in the first place was that they wanted to do good, so the motivation is there in general
15:35 < fenn> maybe therapy for OCD
15:44 < fenn> @QuintusActual muses on EA vs e/acc: "don’t think it’s a personality thing but rather a SES thing 
15:44 < fenn> if life is already amazing for you why risk oblivion even if the probability is extremely low? 
15:44 < fenn> on the other hand if you’re being crushed by increasing inflation and competition in every marketplace, tail risk matters less than the potential promise of a Star Trek utopia"
15:44 < fenn> or at least perceived status
15:44 < fenn> many EA pin their identity/status onto "doing the most good"
15:49 < fenn> at least the current setup provides an exit from EA when you fail at EA and lose your imagined status as a do-gooder
16:03 < L29Ah> > Argentina is going full accelerationist, this guy wants to cut off trade relations with China, privatize the whole economy, and abolish 2/3rds of ministries
16:11 < kanzure> is it still the case that it's mostly software engineers getting sucked into the trap or is it other people too?
16:14 -!- juri__ [~juri@79.140.115.182] has joined #hplusroadmap
16:15 -!- juri_ [~juri@84-19-175-187.pool.ovpn.com] has quit [Ping timeout: 255 seconds]
16:22 < fenn> EA is mostly not software engineers
16:23 < fenn> they often go into the profession because it pays well
16:23 < fenn> i mean, often relative to other charity scenes
16:42 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1175959193652559974/F_U8GqebcAAdvdU.png?ex=656d2058&is=655aab58&hm=eab5e6503bb1ebec2ea9d307af50bc878dbb3e64bf3199b7e09e0a593dec98dc&
16:48 < superkuh> Wow. Great detail.
16:48 < fenn> yeah considering that's 30 km away
16:48 -!- juri__ [~juri@79.140.115.182] has quit [Read error: Connection reset by peer]
16:48 -!- juri_ [~juri@79.140.115.182] has joined #hplusroadmap
16:49 < fenn> um, might have BS'ed that number
16:50 < docl> to do more good than usual (e.g. kids not dying rather than artwork saved) is a net win if you were going to try to do good anyway. sucks that people often seem to stress out and suffer in the attempt though. I take it to be growing pains for the movement. people can usually do more good if they have better mental health, so there's (self-consistent) room for EAs to move in a more relaxed and mutually 
16:50 < docl> supportive direction, more "church" and less "cult"
16:53 < kanzure> pretty sure it's a render sorry to disappoint
16:54 < fenn> it's not a render
16:54 < kanzure> from this person https://twitter.com/AlexSvanArt/status/1726354926299717888
16:55 < fenn> hmm ok
16:55 < fenn> he's really good
16:56 < hprmbridge> alonzoc> The launch was really clean on the way up not a single dead engine etc, pity the booster failed
16:57 < fenn> here's an actual photo from spacex https://nitter.net/pic/orig/media%2FF_Olf8NbEAE-Aaw.jpg
16:58 < L29Ah> 404
16:58 < fenn> a few seconds before the scene depicted in the render
16:58 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1175963341815021649/F_OEPhhW0AApPr-.png?ex=656d2435&is=655aaf35&hm=407dbe9f50ef7ec7ad44b067985d559dbe1db68484f1a708775bd280f7d63152&
16:58 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1175963429702484010/F_OGngpb0AAqvl_.png?ex=656d2449&is=655aaf49&hm=b70e6af700d5c6e0ebc31299ea51f8a1202e95bba4d6b1015eabd00734483761&
16:59 < hprmbridge> kanzure> 0.25 km flame length
16:59 < fenn> L29Ah: can you see this post on nitter? https://nitter.net/SpaceX/status/1725905127113183535
17:00 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1175963816257929288/F_O1--UbMAAjmB9.png?ex=656d24a6&is=655aafa6&hm=fd84402b23c522fd147f437dff4d10772ecf4f37535228b02a509e517257408c&
17:00 < fenn> this looks pretty damn close to the render, only a little more frost sucked into the trap or is it other people too?
17:00 < fenn> 16:14 (*) juri__ [~juri@79.140.115.182] has joined #hplusroadmap
17:00 < fenn> gah
17:00 < fenn> https://nitter.net/pic/orig/media%2FF_OlhMBbcAAElzs.jpg
17:00 < hprmbridge> kanzure> one last one, i'll stop https://cdn.discordapp.com/attachments/1064664282450628710/1175963928543637555/F_OYHJQXkAAS_W-.png?ex=656d24c0&is=655aafc0&hm=a4f0ccd530f9a747f7ee0b15835964f412ccb495020c611d667f5207d16e7d41&
17:00 < hprmbridge> alonzoc> they're awesome
17:02 < L29Ah> fenn: yes
17:02 < fenn> it looks cool but i still think hot staging was a bad idea, especially so early in the test campaign. it probably caused the booster failure
17:03 < fenn> the leading theory is that the engine exhaust pressure from the ship pushed on the top of the booster, decelerating it and causing the propellant to slosh forward and let gas into the engine inlets, and/or causing fluid hammer
17:04 < fenn> now we didn't get to test landing of the booster at all
17:04 < fenn> was it really worth a 10% payload increase? no
17:08 -!- juri_ [~juri@79.140.115.182] has quit [Read error: Connection reset by peer]
17:09 -!- juri_ [~juri@79.140.115.182] has joined #hplusroadmap
18:05 < kanzure> there's a ~30 year VHS archive from this that hasn't been digitized yet and someone needs to remember to do that https://grg.org/
18:19 < fenn> internet archive does that stuff, talk to jason scott i guess
18:47 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 256 seconds]
20:01 -!- gwillen [gwillen@user/gwillen] has quit [Ping timeout: 268 seconds]
20:02 -!- gwillen [gwillen@user/gwillen] has joined #hplusroadmap
20:32 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
22:56 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
22:57 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
--- Log closed Mon Nov 20 00:00:30 2023