--- Log opened Thu Jan 25 00:00:39 2024
01:26 -!- Llamamoe [~Llamamoe@188.146.97.217] has joined #hplusroadmap
01:35 -!- Ashstar [~Ashstar@mobile-166-170-41-40.mycingular.net] has quit [Remote host closed the connection]
02:52 < hprmbridge> kanzure> RNA-guided "bridge recombination" https://twitter.com/pdhsu/status/1750316518133694734 cc @msnewgooty
02:53 < jrayhawk> that looks important
04:53 -!- Llamamoe [~Llamamoe@188.146.97.217] has quit [Quit: Leaving.]
05:13 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
05:29 < hprmbridge> kanzure> can we get a new reprap project lead, this one seems to be defective https://twitter.com/adrianbowyer/status/1750508898346250509
05:53 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 240 seconds]
05:54 < muurkha> kanzure: what did he say?  https://nitter.datura.network/adrianbowyer/status/1750508898346250509 etc. just gives me error messages
05:54 < hprmbridge> kanzure> he doesn't seem to believe in self-replication?
05:55 < muurkha> what did he say?
05:56 < hprmbridge> kanzure> I think he was making a reference to human's lack of human parthenogenesis or he doesn't believe in the existence of human cloning. Either way....
05:57 < hprmbridge> kanzure> Neither should have any bearing on whether self-replication is achievable for a mechanical project.
06:03 < muurkha> what did he say?
06:04 < hprmbridge> kanzure> "No. No more than it has for human beings..."
06:05 < muurkha> in response to what?  that fragment isn't sufficient to make sense of his utterance
06:19 < jrayhawk> https://nitter.rawbit.ninja/adrianbowyer/status/1750508898346250509 works for me
06:19 < muurkha> hooray!
06:19 < muurkha> thanks jrayhawk 
06:20 < muurkha> oh, he's just talking about "vitamins"
06:20 < muurkha> I've had this discussion with him before
06:20 < muurkha> in https://reprap.org/wiki/Wealth_Without_Money he said, "This list is an attempt to make a compromise between immediately-achievable technology and the desirable aim of shortening or eliminating it altogether."
06:21 < muurkha> at some point in the following 15 years or so he came to the conclusion that eliminating the vitamins list was uneconomical
06:23 < jrayhawk> FWIW I have 198.145.64.68 set up to redirect Host: twitter.com requests to https://twiiit.com/ , which further redirects towards ostensibly working nitter instances
06:23 < jrayhawk> you are welcome to /etc/hosts that
06:24 < muurkha> that's a great idea, thanks
06:24 < muurkha> presumably only for unencrypted HTTP requests?
06:25 < jrayhawk> i just use an invalid cert for the https ones
06:26  * muurkha Accept the risk and Continue
06:29 < jrayhawk> i should probably set up video urls to go to fxtwitter.com , now that i think about it
06:56 < hprmbridge> geneh> no, humans aren't completely self replicating. Humans still need vitamins. That's probably what he means.
07:05 < hprmbridge> kanzure> I think that definition was already commonly accepted per KSMR
07:05 < hprmbridge> kanzure> oops I mean KSRM
07:23 < hprmbridge> geneh> MIT's CBA has explicit goal of making a self replicating machine, so there are people out there working on it
07:24 < hprmbridge> kanzure> are you our original genehacker or are you a new one?
07:26 < hprmbridge> geneh> the original.
07:26 < hprmbridge> kanzure> welcome back!
08:36 -!- stipa_ is now known as stipa
08:58 < hprmbridge> astatineknight> Based on this definition nothing in the universe is self-replicating
09:00 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has quit [Ping timeout: 256 seconds]
09:05 < hprmbridge> astatineknight> Perhaps just semantics, but a truer case for humans not self-replicating is sexual reproduction
09:07 < hprmbridge> astatineknight> which funny enough should be renamed to „sexual generation“
09:10 < hprmbridge> astatineknight> as such then self-replication is antithetical to evolution minus random errors of replication
10:18 < hprmbridge> geneh> I think bowyer's wrong about needing to print conductors as good as metal. Actuators, which are probably worth replicating, can probably get away with the conductors we can print now. Electrostatic actuators which use high voltage do not need very conductive material. Anything with a conductivity less than gigaohms is basically a conductor. Lower conductivity just decreases actuation frequency.
10:19 < hprmbridge> kanzure> I was asking on twitter for a progress report and apparently nobody has answered. Surely there has been progress on reprap self-replication in the last 20 years. I mean it seems like a genuinely interesting topic for a review article at least.
10:22 < hprmbridge> kanzure> someone ought to make a backup of the reprap wiki and forums before the bitrot reaches them
10:22 < docl> lot of people doing electroplating of conductors these days
10:26 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
10:26 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
10:57 -!- cthlolo [~lorogue@77.33.24.3.dhcp.fibianet.dk] has joined #hplusroadmap
11:01 -!- Ashstar [~Ashstar@mobile-166-170-41-40.mycingular.net] has joined #hplusroadmap
11:48 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
12:00 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Remote host closed the connection]
12:01 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
12:11 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
12:17 -!- cthlolo [~lorogue@77.33.24.3.dhcp.fibianet.dk] has quit [Quit: Leaving]
12:25 -!- Anachron [~Malvolio@idlerpg/player/Malvolio] has quit [Quit: 675D 2024-01-25 20:25:37:800]
12:51 < fenn> i dumped the reprap wiki history, but getting all the images is going to be a lot of storage space, and i don't really feel like scraping forums tbh https://fennetic.net/mirrors/reprap/
12:51 < kanzure> maybe this is a good task for $30 on fiverr or something
12:53 < fenn> try to get warcs so you can dump them into the internet archive
12:54 < fenn> actually, let me check to see if it's there already
12:56 < fenn> https://archive.org/download/wiki-repraporg_mediawiki is 14GB with images, apparently. i don't see the forums
13:01 < hprmbridge> nmz787> yeah but we can go and walk to the carrot patch that we planted to get the vitamins by our own volition
13:01 < hprmbridge> nmz787> were you the one in OR?
13:02 < hprmbridge> nmz787> or was that thomasegi
13:02 < fenn> thomasegi is in germany or maybe austria
13:02 < hprmbridge> nmz787> https://arstechnica.com/gadgets/2024/01/convicted-murderer-filesystem-creator-writes-of-regrets-to-linux-list/
13:32 < hprmbridge> Katylase> Cas9🥰
13:36 < Ashstar> "<hprmbridge> kanzure> Well he doesn't want yeast to show us up. Or he wants to minimize the genome and remove unnecessary elements?" We really do not know all that is involved in the epigenome, it is like using a Blunderbus to shoot a target
13:38 < Ashstar> I remember my collegues taking bets on the total number of enes in the human genome, most thought it would be up around 120K plus
13:38 < Ashstar> genes
13:38 < Ashstar> insteat of the 22K it is
13:38 < Ashstar> instead
13:39 < Ashstar> shit, plantss have more genes!
13:40 < Ashstar> the sheer complexity of th epigenome is truly astounding and makes us who we are
14:01 -!- Malvolio [~Malvolio@idlerpg/player/Malvolio] has joined #hplusroadmap
14:13 < hprmbridge> kanzure> orthogonal DNA polymerase (in vivo) https://www.science.org/doi/10.1126/science.adk1281
14:14 < hprmbridge> kanzure> can someone add the programmable recombinase from yesterday to the wiki (gene-editing.mdwn)
14:33 < hprmbridge> kanzure> optogenetics review https://www.cambridge.org/core/journals/quarterly-reviews-of-biophysics/article/optogenetic-control-of-neural-activity-the-biophysics-of-microbial-rhodopsins-in-neuroscience/6F9E422992D9418D9418B5F5B5DC5178
14:37 < hprmbridge> kanzure> "strongly suspect certain megastructures will prove impossible to build, because their control systems would have to be superluminal. If you're dealing with a relativistic mass stream, failure propagates far faster than is physically possible to respond to" https://twitter.com/Cererean/status/1750488039489507615
14:57 < L29Ah> "ok"
15:57 < hprmbridge> gourneau> > Today, 454 Bio is releasing its DNA sequencing technology as open source. On our website, you’ll find everything needed to get started: complete plans to build our sequencing instrument, detailed lab procedures, a store to order our custom chemistry, and data analysis software.
15:57 < hprmbridge> gourneau> >
15:57 < hprmbridge> gourneau> https://454.bio/
16:06 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 255 seconds]
16:09 < L29Ah> couldn't find the BOM
16:12 < Ashstar> I wish I could get this bioreactor vessel clean
16:13 < Ashstar> concentrate NaOH, detergent
16:13 < Ashstar> doesn't cut it
16:14 < Ashstar> might break out the acids
16:18 < Ashstar> it's double jacketed glass, it might be on the inside
16:18 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
16:20 < Ashstar> shit it is
16:20 < Ashstar> https://www.msesupplies.com/products/ika-ha-gv-dw-10-glass-vessel-double-wall-bioreactors?currency=USD&variant=39869857136698&utm_medium=cpc&utm_source=google&utm_campaign=Google%20Shopping&stkn=00e3db816675&campaignid=15502649027&adgroupid=&keyword=&device=c&gad_source=1&gclid=Cj0KCQiAqsitBhDlARIsAGMR1RgfSx1zwWiMkHPHGcaSmlXGru5rsx-bU_t9IhnEq2IbxoijmXu4Uf8aAml1EALw_wcB
16:23 < jrayhawk> nmz787: thanks for posting the reiser thing. that is some good closure
16:27 < Ashstar> sorry
16:27 < Ashstar> I guess nobody does fermentation work here
16:27 < jrayhawk> it is nice to see some personal progress has been made in the 15 years he's had to reflect upon the degree to which black & white thinking and splitting has sabotaged his life
16:30 < jrayhawk> i think there's still a reiser4 filesystem sitting on one of my old desktop drives
16:33 < jrayhawk> https://en.wikipedia.org/wiki/Splitting_(psychology)
17:11 < hprmbridge> kanzure> brainwashing tip #3407 https://twitter.com/normal_brandon1/status/1750676024285704329
17:24 < muurkha> geneh: electrostatic actuators can indeed work even with fairly high resistivities, but at macroscopic scales those are pretty weak
17:25 < muurkha> electroplating constructors is a perfectly viable way to do electromagnetic actuators
17:27 < muurkha> nmz787: walking to the carrot patch was never Bowyer's goal with RepRap.  " The machine I propose will self-copy, but not self-assemble. In nature all four possibilities exist: things that neither self-copy nor self-assemble, like rocks; things that self-copy but don't self-assemble, like viruses; things that self-assemble but don't self-copy, like proteins; and finally things that both self-copy 
17:27 < muurkha> and self-assemble, like you and me. And you and I are quite dexterous at assembling machines that we want (even if we do swear at flat-pack furniture), so the second alternative (self-copying without self-assembly) is economically and practically the most interesting option. This web-page, therefore, is about making a useful virus that is as big as a fridge... "
17:31 < muurkha> that was evidently a productive direction for research but I don't think it can deliver the titular promise of "wealth without money" because you still need money for the vitamins
17:33 < muurkha> so ultimately you only get a constant-factor improvement on traditional approaches to manufacturing, though it seems that plausibly that constant factor is somewhere around an order of magnitude
17:46 < muurkha> fenn just informed me of https://archive.org/details/thingiverse, the Thingiverse collection on the Archive
17:49 < muurkha> I guess *piezoelectric* electrostatic actuators aren't weak at macroscopic scales
17:50 < muurkha> but 3-D printing PZT is pretty far outside the scope of RepRap
17:53 < fenn> there are still people grinding away at vitamin count
17:53 < fenn> the thing is, it takes a week and $20 to print out a bunch of plastic beams that snap together, and then you have error accumulation with successive generations
17:54 < fenn> vs just buying some steel rods for $20
17:55 < fenn> stepper motors and controllers is where the bulk of the expense is
17:55 < fenn> probably there is room in the world for a pneumatic stepper motor or vacuum stepper motor
17:56 < fenn> one of the big complaints about 3d printers is the noise, which this would exacerbate
17:56 < muurkha> that's an interesting idea; then you could control the flow of the compressed air with arbitrarily weak actuators
17:57 < fenn> pilot diaphragm valves are printable
17:57 < muurkha> exactly
17:57 < muurkha> fluidic controllers ought to be possible operating up to several kilohertz with existing RepRap technology
17:58 < fenn> baby steps
17:58 < fenn> you mean like an automatic transmission computer?
17:58 < muurkha> don't those contain moving parts?  fluidic gates don't contain moving parts
17:59 < fenn> i think the key is flexure mechanisms
17:59 < fenn> are flexures moving parts? the ontology breaks down
17:59 < fenn> melted goo just isn't going to have precision, so design mechanisms that don't need precision
17:59 < muurkha> flexures are moving parts, yes
18:00 < muurkha> you may be able to do higher-speed computation with flexures than with fluidic logic, but nobody has done flexure computation yet
18:01 < hprmbridge> kanzure> @geneh looked into fluidic controllers a whole ago
18:01 < muurkha> melted goo can have a lot of precision
18:01 < hprmbridge> kanzure> while ago, rather.
18:01 < muurkha> oh?  did they do a writeup?
18:01 < hprmbridge> kanzure> in fact it might be in the logs actually
18:02 < muurkha> https://youtu.be/01iiZEOwPqM is a Russian video about fluidic logic from the 60s with English subtitles
18:02 < fenn> i was thinking more for converting weak actuator outputs into strong actuator outputs, like an inchworm motor or a rotary piston engine (but with diaphragms instead of pistons)
18:03 < muurkha> you were thinking *what* for converting weak actuator outputs into strong actuator outputs?  pilot diaphragm valves?
18:03 < fenn> stepper motors have all kinds of crazy materials and optimizations to increase the power density, but that's actually not very important for our use case
18:03 < fenn> flexures, to make e.g. a rotary piston engine
18:04 < fenn> instead of pistons with hinges and crank arms
18:04 < muurkha> oh, flexures, sure
18:04 < muurkha> "rotary" is probably a misnomer applied to flexures, though hexaflexagons are rotary flexures
18:05 < fenn> https://i.pinimg.com/originals/cf/2e/f2/cf2ef2f2fe3f409aecfd3c2b5db6acf0.gif
18:05 < fenn> only the central bearing needs to rotate freely, the rest could be living hinges
18:06 < fenn> now imagine it's a vacuum powered stepper motor, and you run it by connecting the 3d printed venturi to your hose
18:06 < fenn> or directly powered by water, or whatever
18:06 < muurkha> yes, and you could avoid even the central bearing with a different design, say three sinusoidal oscillatory outputs with phases 120° apart, like 3-phase AC electricity
18:07 -!- Ashstar [~Ashstar@mobile-166-170-41-40.mycingular.net] has quit [Read error: Connection reset by peer]
18:07 < muurkha> that's not amplification, though
18:08 < muurkha> pilot diaphragm valves amplify a weak control signal to control a strong power "signal"
18:08 < muurkha> flexures can do that too
18:08 < fenn> the general idea is to not depend on transistors, which we can't make without dealing with "the englisch" which limits our freedom
18:08 < fenn> everything else is made from potatoes, theoretically
18:09 < muurkha> PLA isn't a suitable material for flexures, though
18:09 < fenn> why not?
18:09 < muurkha> well, that's not true.  it's not a suitable material for long-lived flexures
18:09 -!- Ashstar [~Ashstar@mobile-166-170-41-40.mycingular.net] has joined #hplusroadmap
18:09 < muurkha> it creeps like a motherfucker
18:10 < fenn> like a creeper
18:10 < fenn> and then it explodes, presumably. destroying your base and chests full of items
18:10 < muurkha> heh, well, machines do tend to do that
18:11 < muurkha> it's crystalline enough at room temperature that I don't think fatigue is a problem the way it is with glass (which would otherwise be an excellent flexure material)
18:11 < muurkha> PLA is an eminently suitable material for fluidic gates
18:12 < muurkha> http://web.archive.org/web/20030114133303/http://www.air-logic.com/Summary%20Cat/sum14and15.html may have been the last vendor of fluidic components; their NOR gate says, "Fluidic Transverse Impact Modulator principle has no moving parts, half-millisecond response, high recovery and fan out."
18:13 < fenn> is this not a military EMP/jamming resistant tech?
18:14 < fenn> has it "gone black" lol
18:14 < muurkha> oh, actually, they still sell it: https://air-logic.com/item/legacy/nor-logic-modulators/f3705251
18:14 < fenn> how automatic transmissions work is top secret classified info
18:14 < fenn> i dare you to find anyone who understands them
18:15 < fenn> this catalog page looks like it's from 1965
18:15 < muurkha> their datasheet actually goes into a useful amount of detail: https://air-logic.com/Asset/F370-Nor-Logic.pdf
18:16 < muurkha> it was used liberally in military stuff in the 60s
18:16 < muurkha> because it's not just EMP and EMI resistant but also vibration resistant, which vacuum tubes were not
18:17 < muurkha> not just in Russia, in the UK and USA too
18:17 < muurkha> the problem here is that you spend 200 milliwatts and you get a 1kHz NOR gate
18:18 < muurkha> well, 1.5kHz
18:18 < fenn> conceptually it seems very similar to a vacuum tube
18:18 < muurkha> how?
18:19 < fenn> you have this stream of free flowing matter and it's pushed to one side or another to bias the outputs using the energy from the stream
18:20 < fenn> also the cartridge form factor
18:21 < muurkha> that isn't how triodes work.  maybe you mean cathode ray tubes?  those are technically vacuum tubes
18:23 < muurkha> anyway I was going to say that the other vibration-resistant solid-state transistorless logic family from the 01960s is ferrite/diode systems, which were especially deeply developed in the USSR because their transistor manufacturing was way behind
18:25 < muurkha> square-loop ferrite logic can operate above a megahertz, and like fluidics, contains no moving or unsupported thin parts to wear or have vibration problems
18:26 < fenn> is this like a saturable reactor amplifier? 
18:26 < muurkha> it's kind of like magamps, yeah, but it uses different ferrites that are intrinsically bistable
18:27 < muurkha> which helps with timing restoration
18:27 < muurkha> I've been looking for this book for years but still haven't found it: https://books.google.com.ar/books/about/Square_loop_Ferrite_Circuitry.html?id=Ir0fMQAACAAJ&redir_esc=y
18:29 < muurkha> if my intuition is right, moving electrons around instead of entire atoms gives you about a factor of 30 improvement in speed at a given power
18:30 < fenn> see anything near you? https://search.worldcat.org/title/1704479
18:30 < muurkha> it suggests I go to South Africa or Florida
18:30 < fenn> now i feel spoiled
18:30 < muurkha> no known copies in South America
18:31 < muurkha> we're all spoiled now that we have libgen
18:31 < fenn> maybe i will ride my bike to the library
18:32 < muurkha> I don't, mind you, have the technical facilities available to sinter precise ferrite formulations (though those do at least exist in my metropolitan area)
18:32 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Remote host closed the connection]
18:32 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
18:33 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Remote host closed the connection]
18:33 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
18:37 < fenn> (but not tonight)
18:55 < fenn> ah right the "square loop" is the hysteresis curve
18:56 < fenn> https://www.researchgate.net/profile/Hadi-Heidari-3/publication/337805787/figure/fig1/AS:917487780626434@1595757541841/Three-main-types-of-the-shapes-of-hysteresis-loops-in-different-materials-Source.png
19:32 < hprmbridge> kanzure> behind the curve https://gattacagenomics.com/
19:37 < fenn> future partnership with soylent?
19:38 < fenn> brought to you by the torment nexus
19:41 < fenn> tl;dr it's just preimplantation genetic testing
19:55 < fenn> i can't even keep up with this shit https://old.reddit.com/r/LocalLLaMA/comments/19fgpvy/llm_enlightenment/
20:04 -!- TMM_ [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
20:04 -!- TMM_ [hp@amanda.tmm.cx] has joined #hplusroadmap
20:06 < hprmbridge> geneh> for best results fluidics really needs smooth surface finishes and small orifice sizes. You could probably print fluidics, but I doubt they would be very good. I think stereolithography may be good enough to print audio frequency ones.
20:06 < hprmbridge> geneh> here's some crazy fluidics for the SDI: https://apps.dtic.mil/sti/citations/ADA204334
20:08 < hprmbridge> geneh> idea was to make a fluidic circuit which could demodulate an AM laser signal into something strong enough to drive rocket thrusters to guide a projectile shot out of a railgun into incoming missiles
20:12 < fenn> would drawn glass tubing work?
20:13 < hprmbridge> geneh> it can be done, but that seems like it would require a lot of handwork
20:20 -!- Lando-SpacePimp is now known as Zardoz
20:20 -!- Zardoz is now known as Lando-SpacePimp
20:22 < muurkha> you can do fluidics without smooth surface finishes, they're just a lot less efficient
20:22 < muurkha> not AM, FM
20:23 < muurkha> without smooth surfaces it's hard to get laminar flow
20:34 < superkuh> Nothing new but fun, https://madsciblog.tradoc.army.mil/475-neuronudge-the-science-behind-brain-manipulation/
20:39 -!- mxz [~mxz@user/mxz] has quit [Ping timeout: 264 seconds]
21:24 -!- geneh2 [~cam@pool-173-66-190-123.washdc.fios.verizon.net] has joined #hplusroadmap
22:45 -!- mxz [~mxz@user/mxz] has joined #hplusroadmap
--- Log closed Fri Jan 26 00:00:40 2024