--- Log opened Wed Feb 01 00:00:31 2023
00:17 -!- Llamamoe [~Llamamoe@46.204.77.68.nat.umts.dynamic.t-mobile.pl] has joined #hplusroadmap
00:25 -!- Llamamoe [~Llamamoe@46.204.77.68.nat.umts.dynamic.t-mobile.pl] has quit [Quit: Leaving.]
00:54 -!- stipa [~stipa@user/stipa] has quit [Quit: WeeChat 3.0]
01:48 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
01:48 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
02:49 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Quit: FNLSPDRR]
03:03 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
03:12 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Read error: Connection reset by peer]
03:27 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
04:53 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Ping timeout: 268 seconds]
04:56 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
05:32 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0::4249] has joined #hplusroadmap
06:19 -!- Jay_Dugger [~jwd@47-185-201-55.dlls.tx.frontiernet.net] has joined #hplusroadmap
07:16 -!- stipa [~stipa@user/stipa] has joined #hplusroadmap
07:31 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
07:31 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
09:32 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0::4249] has quit [Ping timeout: 248 seconds]
09:32 -!- yashgaroth [~ffffffff@c-73-147-55-120.hsd1.va.comcast.net] has joined #hplusroadmap
09:39 -!- Llamamoe [~Llamamoe@46.204.72.135.nat.umts.dynamic.t-mobile.pl] has joined #hplusroadmap
09:46 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has joined #hplusroadmap
10:04 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
10:22 -!- Llamamoe [~Llamamoe@46.204.72.135.nat.umts.dynamic.t-mobile.pl] has quit [Ping timeout: 256 seconds]
10:25 < fenn> why hasn't openAI won the loebner prize (chatbot turing test) yet? it seems like an easy project for an intern with the capabilities they have already
10:27 < kanzure> that's not how you move a goalpost!
10:27 < Jay_Dugger> Heh.
10:28 < hprmbridge> lachlan> Too easy!
10:37 -!- Llamamoe [~Llamamoe@46.204.76.182.nat.umts.dynamic.t-mobile.pl] has joined #hplusroadmap
10:50 < fenn> dreamerV3 plays atari games and minecraft and can mine diamonds in 6 days: https://arxiv.org/pdf/2301.04104v1.pdf
10:51 < fenn> the author also made this 2d minecraft style benchmark game https://github.com/danijar/crafter
10:55 < fenn> it took 17 day-equivalents when learning from scratch i think
10:55 < fenn> also it can control virtual robots either through vision or proprioception
10:56 < fenn> they shoulda called it "doer" not "dreamer"
11:00 < kanzure> it doesn't use "expert data" to train it?
11:01 < fenn> no
11:02 < fenn> there are "sparse rewards" for things like getting your first log, crafting a pickaxe, etc
11:07 < fenn> presumably the training process can be sped up by running minecraft faster than realtime, but i don't know how fast they actually ran it, only the number of environment steps (10 million) and the default minecraft simulation speed (20 simulation steps per second)
11:09 < fenn> looks like it spends most of its time derping around before it ever gets a log (by punching a tree) and then rapidly takes off from there
11:20 < fenn> "learning to walk in 1 hour (no simulator)" https://www.youtube.com/watch?v=xAXvfVTgqr0
11:27 < lsneff> Really massive project Orion would be the easiest way to move asteroids
11:39 < kanzure> <nsh>one does not move asteroids, he simply realizes that the current path of the asteroid is already correct</nsh>
11:45 < L29Ah> just calculate a set of mass-trajectories that you need to launch so that all the asteroids coalesce at the desired point as a result of gravitational interactions and collisions over time
11:55 < kanzure> "Since nobody's actually answering your question: Colossal's business model is to use de-extinction to push the frontier of various genetic manipulation techniques, patent them when possible, and sell/license the IP they generate to people who are doing other things. For example, the idea is that in de-extincting the dodo, we'll generate useful techniques for manipulating birds that the ag ...
11:55 < kanzure> ...industry might want to use with chickens."
11:55 < kanzure> "I've spoken with some people there and this is basically their plan. Technology and experience in making significant genetic changes in large organisms likely ends up being valuable for lots of things less flashy than de-extincting famous animals."
11:58 < kanzure> so it's intellectual property licensing wrapped up in the "nobility" of "bio-conservationism" haha
12:08 < Jay_Dugger> I eagerly await the tear jerking videos of the suffering of factory farmed dodo birds.
12:09 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has quit [Read error: Connection reset by peer]
12:11 < muurkha> MatthewGoodman: it depends on the environment; there are definitely iron-deficient algal environments, but they probably aren't inside chlorella bioreactors
12:13 < Jay_Dugger> spirulina extracts have been refined into food coloring.
12:13 < Jay_Dugger> It can make a pretty blue.
12:54 < kanzure> bunnie presentation on hardware security https://twitter.com/ChristopherA/status/1620887440448974848
12:55 < kanzure> https://www.youtube.com/watch?v=JMOWU6pFflw
13:14 < kanzure> a video explaining "DeSci" https://www.youtube.com/watch?v=FgFeO-ry6Po
13:28 -!- codaraxis [~codaraxis@user/codaraxis] has quit [Ping timeout: 248 seconds]
13:44 < nsh> lol kanzure :)
13:45 < nsh> the sensible thing to do is progressively build up a constellation of orbital masses which can build momentum using long trajectories with gravity assist and then when the time comes you can just modulate the paths of some subset of these such that there is a reasonable prospect of achieving the required deflection
13:47 < nsh> the specifics boil down to how much delta v can you put on such things relative to their heft and/or can they be given dynamic inertial boosts by some extrinsic mechanisms
13:47 < nsh> and a computational problem that's not especially difficult
14:15 -!- Llamamoe [~Llamamoe@46.204.76.182.nat.umts.dynamic.t-mobile.pl] has quit [Quit: Leaving.]
14:37 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
15:44 < kanzure> "A family has donated the body of their 113-year-old loved one to science: passed away Jan 27th 2023, just 48 days shy of her 114th birthday [...] Her body was sent to Science Cares in Denver, CO- donated to science but I believe they are not doing any specific testing on her."
15:47 -!- L29Ah is now known as L29Ah[x]
16:17 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Read error: Connection reset by peer]
16:24 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
16:24 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
17:52 -!- codaraxis [~codaraxis@user/codaraxis] has quit [Ping timeout: 252 seconds]
18:57 -!- yashgaroth [~ffffffff@c-73-147-55-120.hsd1.va.comcast.net] has quit [Quit: Leaving]
19:02 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
19:06 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has quit [Read error: Connection reset by peer]
19:10 < hprmbridge> MatthewGoodman> There isn't much of a way to harvest delta V. You can use it to make certain trajectories a bit cheaper, but with big time tradeoffs
19:11 < hprmbridge> MatthewGoodman> https://en.m.wikipedia.org/wiki/Interplanetary_Transport_Network
19:11 < hprmbridge> MatthewGoodman> Spoiler, it's slow. Real slow
19:12 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
19:23 < fenn> "Rejuvenate Bio claims that when they exposed healthy mice near the end of their lives to a subset of the Yamanaka factors, they lived for another 18 weeks on average, compared to just 9 weeks for those that didn’t undergo cellular reprogramming."
19:26 < fenn> MatthewGoodman really long rotating tethers can enable some pretty crazy stuff, like picking up bags of dirt on the lunar surface, flinging it to another rotating tether in low earth orbit (which boosts the tether's eccentricity and dropping the moon rocks toward earth) and that tether can then fling a payload back up to the moon without using any propellant or energy. you can construct a solar
19:26 < fenn> system wide network of momentum conserving tethers in this fashion, so as long as the potential energy of the system is balanced there's no need for energy input
19:29 < fenn> and these are things on hohmann trajectories or something similar, so not as slow as belbruno and farquhar style trajectories
19:29 < fenn> the catch is, well, you only get one chance to catch the payload
19:31 -!- codaraxis [~codaraxis@user/codaraxis] has joined #hplusroadmap
19:37 < fenn>  https://www.nature.com/articles/s43587-022-00183-2 "In vivo partial reprogramming alters age-associated molecular changes during physiological aging in mice"
19:40 < hprmbridge> MatthewGoodman> It's a fun notion, but for every unit of deltaV the tether lifts, it's associated system get a bit lower orbit around the overall barycenter
19:41 < hprmbridge> MatthewGoodman> Moving stuff between the moon and earth is a pretty big potential difference for example
19:42 < muurkha> but when it catches an incoming spacecraft from a higher-potential-energy state, its orbit gets raised back up
19:42 < hprmbridge> MatthewGoodman> It's true. But only because you spent the delta v on the rocket
19:42 < muurkha> right, so the energy moves around the system from tether to tether, carried by payloads
19:42 < muurkha> but none of it is lost
19:44 < fenn> the momentum too
19:44 < hprmbridge> MatthewGoodman> Well, it either has to be lost when it enters a gravity well, unless you are a fan of lithobreaking. Which -technically- isn't a loss as it will increase the planetary angular momentum
19:44 < fenn> correct
19:44 < fenn> unless you gently set it down on the planetary surface with the tether, there will be some inefficiency
19:48 < fenn> the "delta V" comes from the initial orbit of the moon rocks in the moon as it orbits
19:48 < hprmbridge> MatthewGoodman> Ahh, I see. So you want to move the moon closer to the earth as an energy source?
19:50 < hprmbridge> MatthewGoodman> I guess we could just tether the sun and drop it in the black hole in the galaxy center as a shortcut here
19:52 < fenn> why stop there
19:54 < hprmbridge> MatthewGoodman> A better question is where to start
19:57 < fenn> drive to bothell washington and knock some sense back into tethers, Inc
19:59 < fenn> spacex pretty much owns the GTO market, but perhaps there's still room for other orbital tugs like for dead satellite removal
19:59 < hprmbridge> MatthewGoodman> Looks like they are working on interesting stuff
19:59 < fenn> tethers unlimited*
19:59 < hprmbridge> MatthewGoodman> Gotta spend about 1/2 the delta v to deorbit. Good luck with that charity drive
20:00 < fenn> oh jeez tethers unlimited got bought out, it's over
20:00 < fenn> i've never seen anything interesting come out of a company once it's acquired
20:00 < fenn> what a shame
20:02 < fenn> an electrodynamic tether could drag dead satellites down to a lower orbit where they'd deorbit due to drag, and repeat this indefinitely without using any propellant
20:02 < fenn> and no fancy catching involved
20:03 < hprmbridge> MatthewGoodman> I might need a physics refresher on that one
20:04 < fenn> solar panels power a current along the tether. at each end of the tether is an ion collector/emitter sphere that completes the current loop through the ionosphere. when a current is run in a magnetic field, due to the lorentz force the wire is accelerated sideways. (the ions are also accelerated the opposite direction, but i don't think this is where the momentum comes from)
20:05 < fenn> this only really works around earth, but earth is a special case worth considering
20:06 < fenn> oh the gas giants also have magnetic fields and atmospheres so it should work there as well, but you'll have to bring your own power
20:09 < hprmbridge> MatthewGoodman> Has this ever been demonstrated or is this purely speculative? Generally you can't ion drive with the atmosphere because the drag of solar panels dwarfs the thrust you can put on light ions
20:10 < fenn> this is in low earth orbit. there was a space shuttle payload called SEDS that demonstrated the opposite effect, decelerating in earth's magnetic field to generate a current. unfortunately it was hit by a micrometeoroid and the tether burned through before they got much data
20:10 < muurkha> fenn: I think the fact that momentum is conserved here is less interesting
20:13 < fenn> er, TSS-1R not SEDS: https://en.wikipedia.org/wiki/Space_tether_missions?useskin=vector#TSS-1R_mission
20:13 < fenn> muurkha: the fact that momentum is conserved is what allows you to not use propellant
20:13 < fenn> energy is cheap in the inner solar system
20:14 < fenn> MatthewGoodman it's not ion drive, it's more like the tether is the rotor in an electric motor, and the earth's core is the stator
20:15 < fenn> a homopolar motor? i dunno
20:16 < muurkha> oh, yes, you're right.  I feel like I was pretty stupid uh
20:17 < muurkha> three minutes ago
20:17 < muurkha> four minutes ago
20:17 < muurkha> when I didn't understand that
20:19 < fenn> it's all pretty abstract until you've seen some simulations of tethers catching and throwing things
20:20 < fenn> https://www.youtube.com/watch?v=mPx1Nq80jm8&t=3m
20:28 < fenn> also fwiw people are working on atmospheric ion thrusters https://www.esa.int/Enabling_Support/Space_Engineering_Technology/World-first_firing_of_air-breathing_electric_thruster
20:29 < fenn> it might be necessary if *cough russia cough* some rogue nation-state makes a mess out of the starlink constellation and kicks off kessler syndrome for real
20:38 < fenn> all the replacement satellites would have to be below 250km or so to ensure their orbit is clear of debris
20:39 < fenn> the ideal altitude varies with solar activity which puffs up the atmosphere
22:11 -!- codaraxis [~codaraxis@user/codaraxis] has quit [Ping timeout: 260 seconds]
--- Log closed Thu Feb 02 00:00:32 2023