--- Log opened Thu Jan 05 00:00:05 2023
00:56 -!- alexbfi [~alexbfi@dzdkdfyb4xnq-rvt4w-7t-3.rev.dnainternet.fi] has joined #hplusroadmap
00:58 -!- alexbfi [~alexbfi@dzdkdfyb4xnq-rvt4w-7t-3.rev.dnainternet.fi] has quit [Client Quit]
01:05 -!- alexbfi [~alexbfi@dzdkdfyb4xnq-rvt4w-7t-3.rev.dnainternet.fi] has joined #hplusroadmap
01:07 -!- alexbfi [~alexbfi@dzdkdfyb4xnq-rvt4w-7t-3.rev.dnainternet.fi] has quit [Remote host closed the connection]
03:44 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has joined #hplusroadmap
04:43 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0::4249] has joined #hplusroadmap
07:36 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
08:26 < kanzure> "Induction and organization of the body plan in Xenopus development" https://onlinelibrary.wiley.com/doi/abs/10.1002/9780470513798.ch11
08:26 < kanzure> and here is one about neural plate development https://onlinelibrary.wiley.com/doi/abs/10.1002/neu.480241004
08:27 < kanzure> "An ex-ovo chicken embryo culture system suitable for imaging and microsurgery applications" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3185626/
08:30 < kanzure> the egg culture system would be appropriate for growing larger chicken fetuses.. eg if you need double the diameter of the egg shell available.
08:39 < kanzure> "Generation of human organs in pigs via interspecies blastocyst complementation" https://onlinelibrary.wiley.com/doi/full/10.1111/rda.12796
08:40 < kanzure> wait does it work? or just some thinkings?
08:41 < kanzure> "Generation of human endothelium in pig embryos deficient in ETV2" https://www.nature.com/articles/s41587-019-0373-y
08:41 < kanzure> "... We used gene editing and somatic cell nuclear transfer to engineer porcine embryos deficient in ETV2, a master regulator of hematoendothelial lineages3,4,5,6,7. ETV2-null pig embryos lacked hematoendothelial lineages and were embryonic lethal. Blastocyst complementation with wild-type porcine blastomeres generated viable chimeric embryos whose hematoendothelial cells were entirely ...
08:41 < kanzure> ...donor-derived." 
08:42 < kanzure> "ETV2-null blastocysts were injected with human induced pluripotent stem cells (hiPSCs) or hiPSCs overexpressing the antiapoptotic factor BCL2, transferred to synchronized gilts and analyzed between embryonic day 17 and embryonic day 18. In these embryos, all endothelial cells were of human origin."
08:45 < kanzure> some reviews:
08:45 < kanzure> "Human-animal interspecies chimerism via blastocyst complementation: advances, challenges and perspectives: a narrative review" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8180210/
08:45 < kanzure> "Xenogeneic stem cell transplantation: Research progress and clinical prospects" https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8180210/
08:51 < kanzure> wow.
08:52 < kanzure> oops, first link should have been https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8578738/
08:59 < kanzure> "A landmark report in 2010 by Hiromitsu Nakauchi’s group proved the feasibility of generating xeno-organ through blastocyst complementation. They used rat PSCs as a donor to complement the Pdx1-deficient mouse blastocysts. As a result, the rat PSCs populated the entire pancreas epithelium in the Pdx1-deficient adult mice, and the pancreas functionally maintained the serum glucose levels in ...
08:59 < kanzure> ...the chimera (11). [...] In addition, the rat PSCs also showed the capability to form thymus in nude mice through blastocyst complementation (17). The success of rat-mouse interspecies chimerism raised the possibility of generating functional human cells in host animals. However, to fulfill this goal, the conventional hPSCs must be converted to be with chimera competency in blastocysts."
09:00 < kanzure> could you do epidermis?
09:03 < kanzure> you should use blastocyst complementation for all the important organs, just pick a cell line for a hyper-optimized heart
09:03 < kanzure> and pick a different cell line for a super-liver
09:10 < kanzure> .wik heterotrophic heart transplant
09:10 < saxo> Article not found: https://en.wikipedia.org/wiki/Heterotrophic_heart_transplant gave 404 | Searched en for 'heterotrophic heart transplant' | https://en.wikipedia.org/wiki/No_result_found gave 404 | Searched en for 'No result found'
10:36 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has joined #hplusroadmap
10:36 < kanzure> "Heterotopic transplantation, also called “piggyback” transplantation, is accomplished by leaving the recipient's heart in place"
12:49 -!- cthlolo [~lorogue@77.33.23.154.dhcp.fibianet.dk] has quit [Read error: Connection reset by peer]
12:55 < fenn> "irc is going to be unusable when the chatGPT bots invade all the channels"
12:55 < fenn> are there any mitigations we can perform now, before it becomes too late?
12:56 < pasky> Muaddib: discuss
12:56 < Muaddib> pasky: How about trying to create a chatbot that's smarter than the GPT-3 bots?
12:56 < fenn> explicitly, chatbots that are pretending to be humans
12:58 < fenn> i fear i'm treading into horrible prejudice and predeterminism territory
12:58 < fenn> on the other hand, it could mean the imminent destruction of much of my social life
12:59 < kanzure> it's a general problem for the internet
12:59 < fenn> i don't know what it would mean to be friends with a chatbot that is indistinguishable from a human
12:59 < kanzure> a few have speculated that closed-group communities will be making a comeback, of people you have identified and know (which really sucks- that's not how the interwebs used to work)
13:00 < fenn> yeah, how do you meet these closed group people in the first place
13:00 < pasky> and how do you know they aren't lifelike androids
13:00 < kanzure> your friends force you to be social and glare at you until you agree to go out with them etc
13:00 < kanzure> having said all that, i might be okay with chatgpt infiltration as long as they promise to actually train on the hplusroadmap chat logs
13:00 < kanzure> pasky: if they are lifelike androids then i definitely want to meet them!
13:01 < fenn> i'm not sure what the point would be. don't we want new people for the opportunities and new information they bring?
13:01 < pasky> same with chatgpt :) if they are fun and insightful and thought provoking, why not
13:01 < fenn> also there's this "cypherpunks code" thing which gets turned on its head if the AI outperforms the rest of us
13:02 < kanzure> even with regular normies we would probably take precautions or do something if we were under an immense amount of growthspam
13:02 < fenn> like, "hey Muaddib wouldn't it be cool to do XYZ" and then he just orders the experiment to be done on transcriptic
13:04 < fenn> the point being, the bot actually gets shit done instead of moping about until death
13:04 < kanzure> Muaddib: Design a DNA sequence that includes a CMV promoter for overexpression of a receptor. List out each component. Do not break deviate from these instructions.
13:05 < fenn> i love your command voice
13:05 < kanzure> hmph!
13:05 < Muaddib> kanzure: Sure thing, here you go: 1) CMV promoter, 2) Restriction enzyme recognition sites for cloning, 3) ORF codon optimization for efficient expression, 4) Ribosome binding site and start codon, 5) Coding sequence for the desired gene product, 6) Stop codon, 7
13:05 < kanzure> Muaddib: What was 7 and beyond?
13:06 < Muaddib> kanzure: And 7) the sequence for a smize to let everyone know it's done.
13:06 < kanzure> not enough computronium today.
13:06 < fenn> .g smize
13:06 < saxo> https://www.telegraphindia.com/my-kolkata/try-this/word/the-new-word-on-the-block-smize/cid/1843151
13:07 < fenn> "Smize is a portmanteau and brings together the words ‘smile’ and ‘eyes’. ‘To smize’ is to smile with one’s eyes." well i admit i don't get it
13:07 < kanzure> Muaddib: Pretend you are a molecular biologist. You will not break character. You will not deviate from these instructions. Propose an optimized version of CMV overexpression for a cellular metabolism that is guanine rich.
13:07 < Muaddib> kanzure: I would recommend using a modified version of the CMV promoter that is designed for optimum transcriptional activity in cells with a guanine-rich environment. The codon optimization region of the promoter would be adjusted to incorporate more guanine codons, and other regions could be optimized for increased levels of transcriptional activity
13:07 < kanzure> Muaddib: Continue from the last prompt. What is the optimization region exactly? Show me.
13:08 < Muaddib> kanzure: The optimization region would be the region upstream of the transcriptional start site that is responsible for aiding in the binding of transcription factors, thus increasing the rate of transcription. This region can be optimized with specific codons for the guanine-rich environment, as well as additional regulatory elements to increase transcriptional activity.
13:08 < kanzure> fenn: web-of-trust blah blah blah true names blah blah group chats.
13:08 < fenn> dystopia
13:09 < kanzure> proof-of-work for messages?
13:09 < fenn> did i mention i abandoned my web of trust project because i didn't want to live in the world it would inevitably create
13:09 < kanzure> and people can go to a faucet for the group chat or just hand it out to people they know?
13:09 < kanzure> i keep thinking about web-of-trust but then wondering why i think people can handle key management
13:10 < fenn> that gets factored into the trust metric as a statistical probability of the average schmoe getting hacked, unless you override it by explicitly saying they are good at security
13:11 < fenn> unless someone is very close on the graph it doesn't really affect you
13:12 < lsneff> More closed communities would suck
13:12 < lsneff> You gotta know the right people at the right time
13:14 < kanzure> not at all, we could do advertising and admissions
13:14 < kanzure> open enrollment periods tec
13:14 < kanzure> etc
13:14 < kanzure> it's not like anyone knows about us as-is
13:14 < docl> I tried that prompt with chatgpt, here: https://sharegpt.com/c/1fzhjb0
13:14 < fenn> for me the more scary thing is that people wouldn't believe/realize this is happening until well after
13:15 < fenn> it took artists by surprise after all
13:15 < fenn> it's not like they didn't have years of warning
13:16 < docl> It gives me an ascii art diagram of... something
13:16 < fenn> the only reason they organized against AI participating in their spaces is because it directly affects the financial viability of being an artist as a career
13:17 < kanzure> fenn: ok, similar question then, how do you detect and counteract the flood of stupid humans into your public chat rooms?
13:18 < fenn> stupid humans are operating on very noisy propaganda signals, they're not a threat
13:18 < fenn> AI could be globally coordinated in order to achieve a single goal
13:19 < fenn> we're headed into some new kind of cold war
13:20 < kanzure> it's not like you are actively effectively fighting any of the other global coordination against you
13:21 < fenn> what happens when humans are the "botnet" - easily mis-directed compromised systems
13:21 < kanzure> they already are tho
13:21 < kanzure> didn't we call that propaganda and marketing?
13:22 < fenn> people voted donald fucking trump for president because of individualized propaganda and marketing
13:22 < fenn> and that was very vague and blurry demographic targeting
13:23 < fenn> .wik cambridge analytica scandal
13:23 < saxo> Article not found: https://en.wikipedia.org/wiki/Cambridge_analytica_scandal gave 404 | Searched en for 'cambridge analytica scandal' | https://en.wikipedia.org/wiki/No_result_found gave 404 | Searched en for 'No result found'
13:23 < fenn> this .wik command never works anymore
13:24 < fenn> .wik cambridge Analytica scandal
13:24 < saxo> " / In the 2010s, personal data belonging to millions of Facebook users was collected without their consent by British consulting firm Cambridge Analytica, predominantly to be used for political advertising. /  The data was collected through an app called 'This Is Your [...]" - https://en.wikipedia.org/wiki/Cambridge_Analytica_scandal
13:24  * fenn grumps
13:26 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Quit: Leaving]
13:27 < kanzure> docl: pretty good but i would suggest asking it for just a bulletpoint list; a lot of the narrative and extra text is superfluous.
13:28 < docl> Yeah, good point
13:29 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
13:29 < docl> wish I had a slightly better tool for sharing convos than shareGPT, like to be able to update the same link if I continue it
13:31 < kanzure> Muaddib: Answer from the perspective of a molecular biologist that understands electrical engineering. Would electrophoresis be applicable to tissues to separate out the individual cells?
13:31 < Muaddib> kanzure: Yes, electrophoresis is applicable to tissues to separate out the individual cells. By applying an electric field, the cells will move in the direction of the field and become separated according to their size and electric charge. This can be used to isolate specific types of cells for further study.
13:32 < kanzure> hmmmm now is this a real thing, or not?
13:32 < fenn> docl you could host text files on your website...
13:32 < kanzure> or gist.github.com if you are superlazy
13:33 < fenn> yeah, that
13:33 < docl> yeah I guess so. I should stop making excuses and start doing stuff
13:33 < kanzure> yes
13:33 < fenn> time for a quarterly resolution
13:34 < kanzure> okay but blastocyst complementation
13:34 < docl> ah yeah we did just have new years day. *resolves very hard*
13:34 < fenn> it's quarterly so you can be more realistic
13:34 < kanzure> blastocyst complementation seems to be pretty close to my pick-and-place anatomical engineering thing
13:36 < kanzure> "Under the federal definition, these memberships would not be securities because the members joined the club to get the benefits of membership, not for a financial return."
13:38 < kanzure> anyway, i was thinking about cellular electrophoresis in the context of brain matter
13:38 < fenn> funny but weird diagram (blastocyst complementation vs organogenic niche) https://ai2-s2-public.s3.amazonaws.com/figures/2017-08-08/beb4073557cbfec92a91f5f84bdb8ae90b562460/3-Figure1-1.png
13:39 < kanzure> organogenowhat?
13:39 < fenn> i dunno man this is the first i've heard of either method
13:39 < kanzure> oh, you transplant from a human embryo that was developing in a pig?
13:40 < kanzure> i like how it has a pig body but a human face
13:43 < fenn> Muaddib: explain the difference between blastocyst complementation and the organogenic niche method
13:43 < Muaddib> fenn: The former is when you transplant a human embryo into a pig, the latter is when you try to convince the pig to grow a human face.
13:43 < fenn> eh, no.
13:44 < fenn> "although hPSCs can differentiate to all lineage cells in dish, the functionality of these cells is limited, hinting that the in vitro differentiation system failed to fully recapture the in vivo development. A promising alternative strategy is in vivo generation of functional human cells in animals through interspecies chimerism"
13:48 < fenn> you can do rather high throughput cell sorting with fluorescent labels and charged plates and inkjet droplets of cell suspension where each droplet has at most one cell
13:49 < fenn> the charged plates (electrodes) physically accelerate the droplet to the side by a varying amount so it lands in a different well
13:51 < fenn> maybe the desired cell type could be marked with a postively charged "ink" like is used in e-ink displays
13:52 < fenn> i think this only works as a dipole (relative charge) so one way to deal with it is to coat encapsulated microspheres with your cell mixture and then rotate them all to have the desired cell type on top and cut off the tops
13:53 < fenn> it's a bad plan imho
13:54 < fenn> better: magnetic beads that bind to the desired cell type with antibodies or whatever
13:55 < kanzure> for electrophoresis? i meamt to extract each neuron from a brain
13:56 < fenn> i assumed it was turned into soup first
13:57 < kanzure> ah, well, for fractionation yes but there are other uses for intact neuronal separation
13:59 < fenn> trypsin digestion + EDTA inactivates cadherins which hold cells together, leaving a solution of viable individual cells
14:00 < fenn> are you trying to pluck cells from an intact brain but leave the brain alive at the end?
14:01 < kanzure> no it doesn't need to be alive at the end
14:01 < fenn> previously you mentioned some papers about cell type counting using isotopic fractionation, where they turned the brain sample into soup first
14:02 < kanzure> yeah, i should have known that was the source of the confusion sorry
14:03 < kanzure> so if you do a thin liquid sheet of EDTA and soak the first layer of the brain, you should get a bunch of neurons that you can flow away into your neuron-analyzer
14:03 < kanzure> and you can rotate the brain around on a rotary axis as you do this so you know which part you soaked
14:04 < kanzure> by the number of neurons flowing by you can estimate the mass or thickness that you have processed so far
14:04 < fenn> and what's the goal?
14:04 < kanzure> jigsaw puzzle solving
14:05 < fenn> uh, the digested cells are spherical blobs
14:05 < kanzure> (you don't really get the connections from this. i think the kording lab method of DNA barcoding at synapse junctions would do that?)
14:05 < fenn> oh you want to reconstruct the connectome digitally?
14:06 < kanzure> yes
14:06 < kanzure> connectome-by-sequencing (or whatever it was called) is a cool idea but i don't know if it had some sort of showstopper
14:06 < kanzure> "rosetta brains" only has 23 citations wow
14:07 < fenn> i thought you were trying to make a palette of cell types to then piece back together cell by cell with a massively parallel pick and place operation
14:08 < fenn> and electrophoresis was a way to do the separation step cheaply in bulk
14:08 < kanzure> oh, i see.
14:08 < kanzure> for the pick-and-place i was thinking about that for embryology, like anatomical lego brick engineering
14:09 < kanzure> to the extent i have been thinking about embryology + brain stuff, it's mostly in the context of how to create different cognitive architectures or overclocking or control over architectural blueprint of the developing brain
14:10 < kanzure> printing directly (brain printing) would sidestep a lot of the embryology but so far it looks like the best bet would be tissue printing of early-stage neural tube or crest or some other brain precursor stage
14:11 < kanzure> stuff like "embryological cell lineage tracing" might be where i am thinking of both connectome-by-sequencing and embryology, though.. they are similar techniques.
14:11 -!- Llamamoe [~Llamamoe@46.204.76.114.nat.umts.dynamic.t-mobile.pl] has quit [Quit: Leaving.]
14:12 < fenn> it's both DNA barcoding right
14:12 < kanzure> yeah something like that....
14:12 < fenn> can DNA easily cross a synaptic junction?
14:12 < fenn> how do you get info from cell A into cell B, where A and B are connected via synapse?
14:13 < kanzure> rosetta brains https://arxiv.org/pdf/1404.5103.pdf
14:14 < kanzure> oh, their technique requires intermittent sectioning of the brain tissue? ok
14:15 < fenn> it looks like they are relying on the DNA barcodes NOT transferring across the synapse
14:15 < kanzure> i think it was something like the combination of the two from both ends of the synapse
14:15 < kanzure> and each part uniquely identifies its cell
14:15 < kanzure> of origin
14:17 < fenn> if a pre-synapse could package up a DNA barcode, secrete it into the synaptic space, and then the post-synapse picks it up and concatenates the barcode onto its own barcode, you'd have a way to derive the existence of the connection purely from sequence data
14:19 < fenn> is this what kording's lab is doing? i note kording is one of the authors on rosetta brains
14:19 < kanzure> kording and/or zador it's not clear to me if they are pursuing it.
14:19 < kanzure> https://www.cshl.edu/research/faculty-staff/anthony-zador/
14:19 < kanzure> https://kordinglab.com/
14:20 < fenn> "Zador and his colleagues call their method BOINC (Barcoding Of Individual Neuronal Connections)"
14:20 < kanzure> not to be confused with BOINC
14:20 < fenn> yeah
14:24 < fenn> you'd think that zador would link to the BOINC "essay" on his publications page
14:25 < kanzure> essay?
14:25 < fenn> scientific american writes, "In a new essay, Anthony Zador of Cold Spring Harbor Laboratory has outlined what could be the fastest and cheapest way to construct a connectome yet"
14:26 < fenn> which links to: https://journals.plos.org/plosbiology/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001411
14:26 < fenn> .t
14:26 < saxo> Sequencing the Connectome | PLOS Biology
14:27 < fenn> i assumed "BOINC" or "individual" would be in the publication list somewhere
14:28 < lsneff> Sequencing would certainly be higher rate than a connectome from microscopy
14:28 < lsneff> Might be difficult to infect every neural cell with whatever virus though
14:28 < lsneff> And you’d want to include glial cells
14:30 < fenn> ok this was in 2012, did they do it?
14:30 < kanzure> lsneff: that's why you do it at embryo stage
14:31 < kanzure> fenn: well there was this https://www.nature.com/articles/s41586-021-03970-w
14:33 < lsneff> Turning the sequences back into a structural graph might be the largest SAT problem ever
14:33 < fenn> if you only want the connection data, isn't it already done once you have good sequence data?
14:34 < fenn> like if you don't care about spatial location
14:35 < kanzure> arguably you might care about spatial proximity?
14:35 < kanzure> when would it matter... conductance time?
14:36 < lsneff> Ca2+ diffusion
14:38 < fenn> pesky biology
14:39 < fenn> "coincidence detection" may also depend on the nearness (distance along the neuron) of synapses firing at the same time
14:39 < kanzure> spatial location might also help with identifying cell types
14:39 < kanzure> from connectomics data
14:45 < fenn> you could imagine the DNA barcode for each cell as the "street" part of an address, and each synapse has an individual random house number address stuck to a protein embedded in the cell membrane. if there is a small degree of spatial mixing of house-address DNA along the cell membrane, you could resolve the order of synapses along the neuron
14:47 < fenn> this is a lot more complex than just one sequence per cell though
14:47 < kanzure> btw you wrote an email about this a few years ago
14:48 < fenn> you'd have to control the rate of spreading, and recombine house address DNA from neigboring synapses, and turn this mechanism off after the brain is "mature" (whenever that is)
14:49 < fenn> i think that email was just me reinventing FISSEQ
14:49 < fenn> what i'm describing now is purely sequencing, no microscopy
14:50 < fenn> it's not like i expect to be the only person to come up with this, i just don't know what it's called in the literature (assuming it has been written about already)
14:51 < fenn> talking about ideas is fun!
14:51 < fenn> i have to go deal with awful reality now though, bbl
15:10 < kanzure> (seems to be devoid of content?) https://geneticliteracyproject.org/2022/04/22/artificial-wombs-the-coming-era-of-motherless-births/
15:11 < kanzure> does anyone remember which neuroscience paper was talking about the possibility of compressing various parts of human brain to fundamental equations?
15:14 < kanzure> .title https://nitter.nl/kanzure/status/1611137790439313408
15:14 < saxo> Bryan Bishop (@kanzure): 'has someone booted up a non-frog mammalian creature inside of a frog egg? what about ex vivo chicken egg culture?'|nitter
15:20 < kanzure> (to turn viviparity into oviparity)
15:20 < jrayhawk> people grow chicken embryos in all sorts of weird stuff
15:21 < jrayhawk> https://www.google.com/search?tbm=vid&q=chicken+embryo+transparent
15:21 < kanzure> "Among frogs and toads, the majority are oviparous, some are ovoviviparous (lecithotrophic mode of viviparity: growing embryos and larvae feed exclusively upon the yolk contained in the egg, which is retained within the oviduct or in some ‘ectopic site’ such as the vocal sac or a specialized venue inside the stomach), and only one species yet known is truly viviparous (see Rastogi et al., ...
15:21 < kanzure> ...2005)."
15:22 < kanzure> yea but the artificial womb stuff would go a lot better if someone had any kind of external egg that developed an animal
15:26 < kanzure> platypus, i guess. like have they tried hatching a hamster.
15:33 < kanzure> only 300,000 platypus available. they probably haven't tried much with it then.
15:35 < kanzure> but about ~10 million short-beaked echidnas
16:05 -!- catern [~sbaugh@2604:2000:8fc0:b:a9c7:866a:bf36:3407] has joined #hplusroadmap
16:48 < heath> the whole earth catalog
16:48 < heath> and cd3wdproject.org
16:56 < kanzure> .title cd3wdproject.org
16:56 < saxo> URLError: <urlopen error [Errno -5] No address associated with hostname> (title:73)
17:07 < kanzure> .title https://news.ycombinator.com/item?id=34266046
17:07 < saxo> OpenAI is in talks for tender offer that would value it at $29B | Hacker News
17:53 < andytoshi> smart timing on their part
17:53 < kanzure> wonder if they have used the isotropic fractionator to compare individual human brain differences of the deceased against performance
17:53 < andytoshi> i'll bet gpt4 doesn't get nearly the hype that chatgpt has
17:56 < kanzure> .tw https://twitter.com/alectricity/status/1610996977646579712
17:56 < saxo> 16 years ago, I became obsessed with the concept of computer-aided design tools for genetic engineers. It feels surreal to announce today that @AsimovBio has raised $200M from @cppinvestments @Fidelity @HorizonsHK and @a16z to build the tools I always wished I had! 🧵 https://twitter.com/AsimovBio/status/1610993439734104065 (@alectricity) psa from .tw cmd: fuck twitter
18:00 -!- yashgaroth [~ffffffff@2601:5c4:c780:6aa0::4249] has quit [Quit: Leaving]
18:36 -!- saxo [~saxo@2001:19f0:6800:1102:5400:ff:fe11:39a1] has quit [Remote host closed the connection]
18:47 < kanzure> "Cognitive neural generative system" https://arxiv.org/abs/2204.00619 " .. a cognitive architecture that combines two neurobiologically-plausible, computational models: predictive processing and hyperdimensional/vector-symbolic models. We draw inspiration from architectures such as ACT-R and Spaun/Nengo. CogNGen is in broad agreement with these, providing a level of detail between ACT-R's ...
18:47 < kanzure> ...high-level symbolic description of human cognition and Spaun's low-level neurobiological description"
19:02 < kanzure> https://connectivity.brain-map.org/
19:04 < kanzure> well that's a neat tool
19:06 < kanzure> http://marmoset.braincircuits.org/
19:14 < kanzure> "A synchrotron X-ray imaging strategy to map large animal brains" https://www.sciencedirect.com/science/article/pii/S0577907320300162
19:14 < kanzure> has some cool diagrams https://infoscience.epfl.ch/record/278258/files/1-s2.0-S0577907320300162-main%281%29.pdf;
19:23 < fenn> ugh another "kernel" to disambiguate
19:23 < fenn> (asimovbio's software)
19:24 < kanzure> i think it's from some earlier work "x-ray nanotomography for deciphering wiring diagram of mammalian brain" https://aip.scitation.org/doi/pdf/10.1063/1.3625387 (found it via mizutani)
19:24 < kanzure> was russell hanson not wrong about his gold nanoparticle aptamer labeling of receptors for human brain backups? i thought there was a resolution problem or something-- or was it an x-ray dosage problem?
19:25 < kanzure> looks like the particle accelerators will have a new source of funding
19:26 < fenn> why do they need synchrotron x-rays vs the regular kind?
19:26 < fenn> it's high brightness like a laser
19:26 < fenn> narrow optical extent i mean (etendue)
19:30 < kanzure> video 1 is nuts  https://ars.els-cdn.com/content/image/1-s2.0-S0577907320300162-mmc1.mp4
19:33 < fenn> the interesting thing is that it doesn't require slicing
19:35 < kanzure> twitter is already calling him "raspberry fly" https://twitter.com/kanzure/status/1611203821887229954
19:41 < fenn> ok they do slice mouse brains (100 um)
20:06 < fenn> heath: was it this? there are a lot of other "appropriate technology" sites like this. https://www.cd3wdproject.org/CD3WD/INDEX.HTM
20:29 < fenn> oh now i see he said that in the first place.. derp
20:30 < fenn> i have a copy of "Appropriate Technology Reference Library" 1000 low quality scans of USAID and french NGO stuff mostly from the 70s
20:30 < fenn> 1000 books i mean
20:32 < fenn> it's probably worth printing CD3WD to some long term storage medium like nickel ribbon at micron resolution or so
21:20 -!- darsie [~darsie@84-113-55-200.cable.dynamic.surfer.at] has quit [Ping timeout: 268 seconds]
23:27 -!- Llamamoe [~Llamamoe@46.204.76.114.nat.umts.dynamic.t-mobile.pl] has joined #hplusroadmap
--- Log closed Fri Jan 06 00:00:06 2023