--- Log opened Sun Feb 18 00:00:02 2024
00:45 < fenn> RNA editing sounds like a great way to ensure continuing revenue streams for pharma companies
00:55 < fenn> the "wurstchen" image generation model was trained from scratch for ~$40k in GPU time (24,000 A100 hours available @ $13/hr on vast.ai)
01:01 < fenn> i want to say it's roughly comparable to stable diffusion 1.4 but since they filtered out 98% of the input data for "aesthetics" (and other political stuff) it's hard to say really how it will end up performing once a competent fine tuner gets to it
01:01 < fenn> like maybe it's already equivalent to a fine tuned stable diffusion model from a model bias perspective, so there's nowhere up to go. who can say
01:02 < fenn> anyway it's just a datapoint for how expensive things are now (3 months on one 8x A100 node)
01:10 < fenn> aka "wuerstchen" blech
01:30 < L29Ah> > the exchange loses the majority of its value when plausible deniability is lost
01:30 < L29Ah> jrayhawk: how so?
02:08 -!- CryptoDavid [uid14990@id-14990.uxbridge.irccloud.com] has joined #hplusroadmap
02:17 < jrayhawk> lol
03:01 < hprmbridge> Eli> You kind of are a worm fenn. You’re a tube with food going in one end and waste out the other. And you can reproduce. Just get some lambos as a subconscious signaling mechanism to display high genetic NPV. Even better, if you can take advantage of information asymmetry by getting your customer intoxicated, you may increase your perceived genetic NPV for at least long enough for her to make an
03:01 < hprmbridge> Eli> investment.
03:04 < hprmbridge> Eli> Remember, you’re in a market. So if you do not have the actual highest Future Value, you can at least manufacture demand via advertising, information asymmetry, and false claims
03:11 < hprmbridge> Eli> Of course, if the time horizon of the investment is very short, perhaps even one night, it will be difficult to do proper due diligence. This can result in suboptimal investments.
03:15 < juri_> this is a very weird way to discuss this topic. let's not treat humans like investment vehicles please?
03:17 < L29Ah> why not?
03:17 < hprmbridge> Eli> Cause he’s not an economist
03:17 < hprmbridge> Eli> Cause he’s not an economist
03:17 < juri_> no, because she's not a he, thankyouverymuch.
03:18 < hprmbridge> Eli> Sorry. Difficult to tell on these platforms
03:19 < L29Ah> if humans aren't investment vehicles, then it doesn't make sense to reproduce among other things
03:20 < L29Ah> but we observe that females are on contrary more eager to reproduce, so they're more likely to treat humans as investment vehicles
03:20 < juri_> for once, we completely agree. hense, why i'm married to another woman. :P
03:21 < juri_> I am my pattern, not my genes. and life affords plenty of opportunities to help others along the path, without creating more others.
03:28 < hprmbridge> Eli> I’ve been thinking about the steady state equilibrium society has to find based on gender strategy. Men are more likely to invest in short term investments with low long term risk but high potential upside. Women have higher longer term risk with a low due diligence investment strategy. The pill changed investment strategies for men and women resulting in a new steady state gender dating
03:28 < hprmbridge> Eli> equilibrium. This pretty much explains much of the behavior we see in men and women today in comparison to 100 years ago.
03:56 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
06:04 < hprmbridge> nmz787> Gender in chat is very odd indeed
06:11 < nsh> boak
06:13 < nsh> people who think markets intrinsically make things better may find themselves being given formative experiences with being on the brunt end of the stick one day
06:14 < nsh> and those who think a price can be put on anything will be soon parted from their soul
06:20 -!- darius___ [~darius@2605:a601:aad3:ed00:6dd1:cd87:41b8:1182] has joined #hplusroadmap
06:25 -!- darius__ [~darius@2605:a601:aad3:ed00:9ca0:27ee:5996:f393] has quit [Ping timeout: 268 seconds]
06:26 < hprmbridge> nmz787> Eh, price is not necessarily monetary
06:30 < hprmbridge> kanzure> "Modified human glucagon-like peptide-1 (GLP-1) produced in E. coli has a long-acting therapeutic effect in type 2 diabetic mice" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5531477/
06:34 < juri_> gender is very strange to start with. we're just told it's simple.
07:07 < TMA> it is just epiphenomenon over biology, which itself is an ephenomenon over chemistry, an epiphenomenon over physics :)
07:07 < hprmbridge> nmz787> Gender is less and less useful IMO. Sex is all that matters for procreation. Secondary sex characteristics are also pretty influential.
07:16 < TMA> prices are useful, bu they are just proxies for making preference ordering a bit more tractable
07:18 -!- cthlolo [~lorogue@77.33.24.3] has joined #hplusroadmap
07:18 < TMA> according to my macroeconomist teacher, monetary prices are recently less useful, because the sheer number of them exceed the capacity of human brain (on the order of 1e5 to 1e6 price datapoints)
07:19 < TMA> what is the price of milk for example?
07:21 < TMA> it is no longer well defined. milk is split to many categories: fat contents: whole milk, skimmed milk, ... method of production: bio/organic, conventional
07:23 < TMA> preservation: raw, pasteurized, UHT; container: glass, plastic jug, layered paper without a spigot, layered paper with a spigot, ...
07:24 < juri_> oat, almond, soy..
07:25 < TMA> that's not a milk. a milk substitute
07:25 < TMA> lactose removed/not removed
07:26 < TMA> and place of origin, if you want to boycott your neighbor
07:26 < juri_> tomatoes are a fruit, but they do not go in a fruit salad.
07:26 < juri_> soy milk goes in cerial. therefore, it's milk.
07:26 < TMA> then there are brands superimposed over that
07:28 < TMA> juri_: by the same analogy stainless steel is a precious metal, because some jewelery is made out of it.
07:29 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
07:29 < TMA> pseudomilk has many of the culinary uses of true milk, yet there are differences which preclude their unification
07:30 < TMA> let's just observe that you can have allergy to soy (and therefore soy pseudomilk) but not to ordinary milk or vice versa
07:30 < L29Ah> TMA> according to my macroeconomist teacher, monetary prices are recently less useful, because the sheer number of them exceed the capacity of human brain (on the order of 1e5 to 1e6 price datapoints)
07:30 < L29Ah> that's bullshit since no single human brain needs 1e5 price datapoints for any single task
07:31 < TMA> it does not need them for a _single_ task
07:32 < L29Ah> pretty sure one won't even need to consider 1e5 prices over lifetime
07:32 < L29Ah> unless you're a day trader or such
07:34 < L29Ah> i don't see your point
07:35 < TMA> if you are on a tight budget, you need to watch if an egg costs 4¤ in grocery store A or 4.10¤ in grocery store B
07:36 < TMA> that's two prices for one commodity if you have just two vendors
07:37 < TMA> if you have more vendors, the number grows fast
07:38 < L29Ah> then you note all the prices you need and go to stores A and B for correspondingly cheap stuff, and don't consider prices until they change, and that doesn't happen often
07:40 -!- Mabel [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
08:03 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
08:03 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
08:13 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
08:13 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
08:27 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
08:27 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
08:37 < hprmbridge> Katylase> Hi friends! How are you?😊
09:21 < jrayhawk> I'm alive! How 'bout yourself?
09:57 < hprmbridge> Eli> Doesn’t this mostly just reveal time preference? Large companies care a lot about almost infinitely small price differences and have the resources to measure this. But egg due diligence has low upside for most humans, and potentially very large downside with regards to opportunity cost.
10:11 < hprmbridge> Katylase> Awesome! I'm 3d printing some protein models right now🤭
10:11 < jrayhawk> are they foldable?
10:14 < hprmbridge> Katylase> No, just regular surface representations
10:14 < hprmbridge> Katylase> But I have 4 different types/versions of them
12:53 -!- cthlolo [~lorogue@77.33.24.3] has quit [Quit: Leaving]
13:17 -!- CryptoDavid [uid14990@id-14990.uxbridge.irccloud.com] has quit [Quit: Connection closed for inactivity]
13:19 -!- Guest0 [~Guest0@107-199-26-179.lightspeed.cicril.sbcglobal.net] has joined #hplusroadmap
15:25 < hprmbridge> Eli> Akkermansia is supposed to increase GLP1 and you can buy it here if you want to experiment: https://pendulumlife.com/products/pendulum-glucose-control-2-og
15:26 < hprmbridge> Eli> I’m a little curious about titration, but I can’t be bothered to read any of the papers to see how they deal with that in studies
15:30 < hprmbridge> Eli> If their study is in any way accurate, a .6% reduction in A1c is pretty decent. https://cdn.discordapp.com/attachments/1064664282450628710/1208918477851725865/IMG_0974.png?ex=65e50811&is=65d29311&hm=4964e3bd0a27dbd826769fe041be99ae5c8073b00d615bef718670b06cb6faed&
16:25 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 256 seconds]
18:07 < fenn> recently i've been re-reading marshall brain. the guy was so on point in every way... https://marshallbrain.com/etq-wholesale
18:08 < fenn> i almost feel bad linking to his current site instead of the wayback machine from 2003
18:24 -!- Guest0 [~Guest0@107-199-26-179.lightspeed.cicril.sbcglobal.net] has quit [Quit: Client closed]
19:21 -!- srk_ [~sorki@user/srk] has joined #hplusroadmap
19:22 -!- srk| [~sorki@user/srk] has joined #hplusroadmap
19:23 -!- srk [~sorki@user/srk] has quit [Ping timeout: 246 seconds]
19:26 -!- srk| is now known as srk
19:27 -!- srk_ [~sorki@user/srk] has quit [Ping timeout: 272 seconds]
19:28 -!- helleshin [~talinck@108-225-123-172.lightspeed.cntmoh.sbcglobal.net] has quit [Read error: Connection reset by peer]
19:43 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
19:44 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
20:07 < hprmbridge> Katylase> @jrayhawk here are my 3d printed Cas9s! They're kind of cute...🥰 https://cdn.discordapp.com/attachments/1064664282450628710/1208988244562157648/IMG_2625.jpg?ex=65e5490b&is=65d2d40b&hm=8512ce9d734dfdbecd61377e4e4c58898a46c2f02b3beaad528e959f41f54c9f&
20:15 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
20:35 -!- mxz [~mxz@user/mxz] has quit [Ping timeout: 255 seconds]
21:28 < jrayhawk> Are you going to be painting tertiary, secondary, or primary structural components?
21:42 < hprmbridge> nmz787> Isn't the 3d already tertiary?
21:44 < hprmbridge> nmz787> Maybe I never learned this right, or forgot it.. but I'm currently thinking primary is the sequence and the side chains, secondary is thermal folding in a relatively clean/pure buffer, and tertiary is folding when helped by other folding helpers (post translational modification) and also hooking up with other proteins in a non-covalent manner (holoenzyme crap)
22:25 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 255 seconds]
22:36 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
22:41 < jrayhawk> okay, supersecondary structures, then
22:42 -!- mxz [~mxz@user/mxz] has joined #hplusroadmap
22:49 < hprmbridge> Katylase> This? https://cdn.discordapp.com/attachments/1064664282450628710/1209029088736710666/IMG_2626.png?ex=65e56f15&is=65d2fa15&hm=f50159ceb7d7a615d62494fb5d2915c8f8b8bba1d87ec27ee3e0aa0ce37e02ad&
--- Log closed Mon Feb 19 00:00:03 2024