--- Log opened Tue Dec 02 00:00:54 2025
00:18 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has quit [Ping timeout: 246 seconds]
00:58 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
01:11 < fenn> Z-Image-Turbo works well in my limited testing (see the iili link above)
01:11 < fenn> apple's video model "starflow" is a joke
01:12 < fenn> trinity seems alright, about the same as gpt-oss-20b (but perhaps without the internal psychopathic monologue)
01:12 < fenn> trinity-mini
01:47 -!- Gooberpatrol66 [~Gooberpat@user/gooberpatrol66] has joined #hplusroadmap
02:07 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
02:08 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
02:21 < kanzure> this seems not terrible for organizing local project docs in sprawling files? https://www.mkdocs.org/user-guide/writing-your-docs/
02:31 < kanzure> supercollider sucks how does anyone live with this
02:34 < kanzure> these look nice tho https://github.com/sonic-pi-net/sonic-pi/tree/dev/etc/synthdefs/designs/supercollider
02:34 < kanzure> whoops i meant these https://github.com/SCLOrkHub/SCLOrkSynths/tree/master/SynthDefs
03:08 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
03:08 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
03:35 < hprmbridge> kanzure> https://quillette.com/2025/12/02/embryo-selection-a-guide-for-the-curious-herasight-intelligence/
03:40 < pasky> fenn: the default models MuaddibLLM has set up should rarely succumb to dumb loops, but it would make sense for these in case they'd be more prone (btw there is already an automatic fallback to unsafe model in case of explicit rejection, though it's currently not a very used codepath since anthropic models are the main ones good at signaling rejection explicitly, and they aren't default for anything 
03:40 < pasky> rn)
03:44 < hprmbridge> kanzure> "Snake venom protection by a cocktail of varespladib and broadly neutralizing human antibodies" https://www.cell.com/cell/abstract/S0092-8674(25)00402-7
03:44 < hprmbridge> kanzure> "We iteratively recovered two broadly neutralizing antivenom antibodies from the memory B cells of a hyperimmune human donor with extensive snake venom exposure. The antibodies recognized conserved neutralizing epitopes on prevalent long and short snake neurotoxins, with crystal structures revealing antibody mimicry of the interfaces between these neurotoxins and their host target, the nicotinic
03:44 < hprmbridge> kanzure> acetylcholine receptor. We combined and tested these antibodies and the phospholipase inhibitor varespladib. A 3-component cocktail rescued animals from whole-venom challenge of all species in a 19-member WHO Category 1 and Category 2 elapid diversity set, with complete protection against most snakes observed."
03:48 < hprmbridge> kanzure> "Synthetic development of a broadly neutralizing antibody against snake venom long-chain α-neurotoxins" https://pmc.ncbi.nlm.nih.gov/articles/PMC7617732/
04:01 < hprmbridge> kanzure> "De novo designed proteins neutralize lethal snake venom toxin" https://www.nature.com/articles/s41586-024-08393-x (2025)
04:09 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
04:09 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
04:16 < hprmbridge> kanzure> "Somatic mutations impose an entropic upper bound on human lifespan" https://www.biorxiv.org/content/10.1101/2025.11.23.689982v1.abstract
04:27 < hprmbridge> kanzure> "Our analysis reveals fundamental asymmetry across organs: post-mitotic cells such as neurons and cardiomyocytes act as critical longevity bottlenecks, with somatic mutations reducing median lifespan from a theoretical non-aging baseline of 430 years to 169 years. In contrast, proliferating tissues like liver maintain functionality for thousands of years through cellular replacement, effectively
04:27 < hprmbridge> kanzure> neutralizing mutation-driven decline. Multi-organ integration predicts median lifespans of 134-170 years —approximately twice current human longevity. This substantial yet incomplete reduction indicates that somatic mutations significantly drive aging but cannot alone account for observed mortality, implying comparable contributions from other hallmarks."
04:33 -!- catalase [~catalase@user/catalase] has quit [Quit: Ping timeout (120 seconds)]
04:33 -!- catalase [~catalase@user/catalase] has joined #hplusroadmap
05:29 < L29Ah> kanzure: ppl often just build on top of it, like tidalcycles
05:29 < hprmbridge> kanzure> yea but the nonrealtime stuff barely works.
05:30 < hprmbridge> kanzure> i wouldn't mind typing out sclang, tidalcycles looks okay too
05:31 < L29Ah> pasky: maybe detect rejections with a 3B model and paste them separately for the user's feedback?
07:07 < kanzure> https://securitylabs.datadoghq.com/articles/shai-hulud-2.0-npm-worm/
07:17 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=bec2f061 Bryan Bishop: add lots more pro-reading mutations >> http://diyhpl.us/diyhpluswiki/genetic-modifications/
07:17 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=eecf90ee Bryan Bishop: Merge remote-tracking branch 'origin/master' into master >> http://diyhpl.us/diyhpluswiki/
07:47 < kanzure> https://diyhpl.us/~bryan/irc/chatgpt/synaptogenesis.mp3
08:27 -!- adlai [~adlai@user/adlai] has joined #hplusroadmap
08:31 < kanzure> adlai!
08:31 < adlai> kanzure!
08:31 < kanzure> so you're still alive
08:32 < adlai> yeah; ... not much to report that would be of significant interest to the channel
08:32 < adlai> didn't get re-drafted in the latest war, so I guess that's good news.
09:38 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 246 seconds]
10:03 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
10:03 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
10:26 < kanzure> https://mistral.ai/news/mistral-3
10:40 < kanzure> https://github.com/stuhlmueller/strudel-llm
10:42 < kanzure> run with https://codeberg.org/uzu/strudel
12:06 -!- stipa [~stipa@user/stipa] has joined #hplusroadmap
12:14 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
12:14 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
13:31 < RangerMauve> Are y'all takin K2 with your vitamin D3?
13:31 < RangerMauve> One of my paramedic friends recommended the adjustment
13:33 < jrayhawk> I eat inordinate amounts of menatetrenone, yes
13:35 < hprmbridge> Eli> I do when I take it. But I also walk around in the sun to get vitamin d.
13:36 < hprmbridge> Eli> Yeah I made a video on this. I sent the video to the researchers and also to Tim Friede and they like it: https://m.youtube.com/watch?v=0pyGHgtqt8A
13:37 < hprmbridge> Eli> This is one of those things where they will save a lot of lives. Biotech is becoming magical.
13:39 < L29Ah> RangerMauve: no, i take d3 twice a week and k2 every other day
13:40 < RangerMauve> L29Ah Oh interesting. I was advised to take both daily. Is that bunk?
13:40 < hprmbridge> Eli> I rendered black mamba venom on the acetylcholine recptors. This causes you to die of flaccid paralysis. You are fully conscious and know you are dying. https://cdn.discordapp.com/attachments/1064664282450628710/1445530062341279865/venom_.jpeg?ex=6930ae47&is=692f5cc7&hm=9e3e771cab74f644310d5b1bc29b7cbcbdf32dc5242c2ffd7e7f9f7e156b7fcb&
13:40 < jrayhawk> in healthy humans, you have an enormous capacitor for fat-soluable nutrients in the form of adipose tissue
13:40 < L29Ah> idk i saw no good reason to
13:41 < RangerMauve> I wish I got my D from the sun but the dayjob + Canadian winter are working hard against that option
13:41 < L29Ah> i wish i never got D from the sun but i'm born white
13:42 < RangerMauve> Is in vivo melanin increase therapy on the wiki? :P
13:42 < kanzure> pretty sure?
13:42 < kanzure> ctrl+f melanin https://diyhpl.us/wiki/genetic-modifications/
13:43 < RangerMauve> Is in vivo gene editing actually viable already? I thought that even with crispr your default genome ends up outcompeting the cells eventually.
13:43 < kanzure> huh i guess not... explicitly. feel free to add it.
13:44 < kanzure> RangerMauve: ah! well our method is not as stupid as single-cell CRISPR microinjection of an embryo.
13:44 < kanzure> our approach is based on modifying the genome before inserting it into an egg or embryo
13:44 < kanzure> allowing for normal stem cell expansion in vitro
13:44 < kanzure> and then multiple cycles of genetic engineering and sequencing for quality control
13:44 < RangerMauve> I was thinking existing living humans rather than freshly made ones
13:45 < RangerMauve> That's good to know though
13:45 < kanzure> adult somatic gene therapy is very difficult in comparison, you can get good edits on liver and uhh maybe your retina
13:45 < RangerMauve> Slowly warming my partners up to the tought of doing some custom editing of our offspring rather than the usual methods
13:45 < kanzure> maybe some dermis modification... you can get some localized gene therapy via electroporation if you want... uhh if you're very risk friendly then you could do viral gene therapy but you cannot get 100% coverage.
13:46 < L29Ah> viral coverage sucks
13:46 < jrayhawk> RangerMauve: your paramedic friend is sortof correct in that menatetrenone is needed to carboxylate osteocalcin to aid in putting calcium where it needs to be (with vitamin d) and keep it out of where is shouldn't be (with retinoids). menatetrenone is additionally is an essential intracellular carrier in the mitochondrial electron transport chain. retinoids and vitamin d also need to be balanced 
13:46 < jrayhawk> because their receptor heterodimerize in immune function.
13:46 < L29Ah> on the other hand it is a good thing as you don't get killed by the immune response
13:46 < kanzure> yea herasight can only do +9 IQ with 20 embryos or something. and if you and your partner are retards, then there's not much that can be done with selection, hence genetic engineering.
13:46 < jrayhawk> er, their receptors
13:47 < kanzure> i think this would be a good edit to add to the wiki.
13:48 < RangerMauve> I'm pretty happy on the IQ front. I'm thinking more along the lines of the sleep modification or any immune system related changes.
13:48 < kanzure> MuaddibLLM: propose an entry for the germline genetic modifications page based on the above conversation and melanin skin pigmentation so that we can go outside again.
13:49 < MuaddibLLM> Searching for constitutive eumelanin-boosting gene targets (MC1R, SLC24A5, ASIP) suitable for germline editing...
13:49 < kanzure> no it's on the page already
13:49 < RangerMauve> jrayhawk Yeah re: menatetrenone, they mentioned it specifically to combat vascular calcification
13:49 < MuaddibLLM> For germline editing targets to boost constitutive eumelanin (sun protection): 1) **MC1R** (Melanocortin 1 Receptor) super-active variants (e.g., "African" consensus alleles or synthetic high-activity mutants) to drive eumelanin over pheomelanin; 2) **ASIP** (Agouti Signaling Protein) knockout/disruption, as ASIP naturally inhibits MC1R—breaking this brake constitutively activates darkening (seen in "dominant black" 
13:49 < MuaddibLLM> horses/mice); 3) **SLC24A5** (ala111thr variant) is the classic "golden" gene, fixing the European low-activity variant back to ancestral high-function significantly boosts... full response: https://pasky.or.cz/muaddib/3bf4aaca10b6465196cafdc5496de743.txt
13:50 < kanzure> vantablack humans
13:53 < jrayhawk> RangerMauve: https://www.westonaprice.org/health-topics/abcs-of-nutrition/beyond-good-and-evil/#:~:text=a%20balance has a good summary, plus https://lirias.kuleuven.be/server/api/core/bitstreams/399ed5b2-4fa8-4945-b178-4e40eb4d04b7/content for the electron transport stuff
13:54 < hprmbridge> Eli> To some degree. They have some approved therapies, like for sickle cell (but it's not fun). They also are doing it for metabolic disorders that would normally kill a kid. They can put the payload into a lipid nanoparticle and it preferentially accumulates in the liver. They did it on a kid and saved his life. Now, they are expanding this to others.
13:55 < jrayhawk> notably mk4 has greater liposoluability than mk7, so it's bioavailable to lipoprotein-lipase-based cellular uptake in ways mk7 can't be
13:56 < jrayhawk> which is also the reason it compliments coq10 in electron transport
13:56 < RangerMauve> I wish there was a multi vitamin thing at Costco that just had all this stuff balanced out already. 😅
13:58 < jrayhawk> very few people would be willing to enough for that to happen
13:59 < jrayhawk> er, willing to pay enough
13:59 <+gnusha> https://secure.diyhpl.us/cgit/diyhpluswiki/commit/?id=8f73d35b Bryan Bishop: more melanin to protect against solar radiation damage >> http://diyhpl.us/diyhpluswiki/genetic-modifications/
13:59 < L29Ah> RangerMauve: there are crazy amounts of different balance offers everywhere
14:00 < RangerMauve> Maybe someday we can implant new organs with bacteria in them to produce the right balance in realtime or some other scifi scenario
14:00 < RangerMauve> Like this but for everything. https://news.mit.edu/2023/implantable-device-enable-injection-free-control-diabetes-0918
14:00 < L29Ah> or chop our current organs away and outsource even more chemistry
14:00 < kanzure> wow what if there are microbes that colonize the gut already?
14:01 < L29Ah> gut stuff is often poorly available
14:01 < RangerMauve> Oh has there been research on conciously modifying the gut biome? That sounds more fun.
14:01 < L29Ah> for example there was a study that demonstrated wild gut microbes producing enough B12 to feed a non-supplementing vegan
14:01 < jrayhawk> what if we had portable self-maintaining fermentive bioreactors that convert grass to fat-soluable vitamins in a convenient colloidal form?
14:01 < L29Ah> but one had to eat their own shit extract
14:02 < jrayhawk> such a thing is unimaginable
14:02 < kanzure> RangerMauve: what if we do fecal matter transplants from mathematicians to non-mathematicians
14:02 < jrayhawk> https://letter.palladiummag.com/p/the-bison-sphere-manifesto
14:02 < kanzure> i call it SETIp - the search for exceptional terrestrial intestinal poo
14:02 < RangerMauve> Unironically I believe the fecal matter transplant stuff is slept on too much.
14:03 < RangerMauve> jrayhawk My rabbits seemed to have a blast eating their shit all day
14:03 < jrayhawk> i will stick with pastured eggs and grass-fed butter, thanks
14:04 < kanzure> various other gut microbiome proposals are listed here https://diyhpl.us/wiki/nootropics
14:04 < kanzure> in particular i think that targeting fetal brain development via maternal gut content could work
14:04 -!- adlai [~adlai@user/adlai] has quit [Ping timeout: 256 seconds]
14:06  * L29Ah still didn't perform that poo treatment on himself
14:07 < L29Ah> MuaddibLLM: give us a litany against procrastination
14:07 < MuaddibLLM> I must not delay; procrastination is the slow mind-killer that steals lives in five-minute chunks - I will begin the smallest possible step now, and where action has passed, only progress will remain.
14:07 < RangerMauve> Also surprising real time monitoring of vitamins isn;t a thing. I'm planning to integrate a continuous blood glucose monitor sometime in the next year.
14:07 < jrayhawk> you lost your fermentive hindgut for a reason https://en.wikipedia.org/wiki/Appendix_(anatomy)
14:08 < kanzure> RangerMauve: if you attend the group event IRL then you will meet someone doing continuous peptide monitoring device for human
14:08 < L29Ah> sounds expensive
14:08 < RangerMauve> Dang! I would love to if it was outside the US. 😅
14:08 < L29Ah> with a good antibody chip it can work tho
14:08 < kanzure> uh oh are you on a sanctions list
14:09 < RangerMauve> Not yet! There's just been random Canadians harassed at borders already so I'm minimizing my risks.
14:10 < RangerMauve> https://www.stelo.com/
14:10 < RangerMauve> This is the glucose one I'm going to try first BTW
14:11 < RangerMauve> Wish there was some sort of open source alternative for their apps, but maybe I'll be the one to reverse their protocols and add it to gadgetbridge or whatever
14:12 < RangerMauve> Off to do meatspace adventures. Hope your evening or day is fruitful!
14:18 -!- adlai [~adlai@user/adlai] has joined #hplusroadmap
14:30 < hprmbridge> kanzure>  https://cdn.discordapp.com/attachments/1064664282450628710/1445542759497990204/image0.png?ex=6930ba1a&is=692f689a&hm=85757995737aa1483e10455f76971adcc0cb8148060d344289dfc1b92c57c1a9&
14:56 -!- adlai [~adlai@user/adlai] has quit [Quit: WeeChat 4.7.1]
15:15 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
15:15 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
16:00 < jrayhawk> how was that imaging/modelling done?
16:03 -!- stipa_ [~stipa@user/stipa] has joined #hplusroadmap
16:04 < hprmbridge> kanzure> nooo idea.
16:05 < hprmbridge> kanzure> just scrolled by in a feed. no source.
16:05 -!- stipa [~stipa@user/stipa] has quit [Ping timeout: 246 seconds]
16:05 -!- stipa_ is now known as stipa
16:13 < hprmbridge> kanzure> "A French teenager has spent the past few years systematically reviewing historical biostasis case reports and other documents. His independent research echoes the detailed complaints raised by Mike Darwin, Aschwin de Wolf, Max More, Emil Kendziorra, T. Theodorus Ibrahim, and others. Standby, stabilization, and transport continues to lag far behind where it could be."
16:13 < hprmbridge> kanzure> https://www.reddit.com/r/CryonetReddit/s/4JbMkWoaBo
16:32 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has quit [Ping timeout: 252 seconds]
16:34 -!- rafspiny [~raffaele@a120210.upc-a.chello.nl] has quit [Ping timeout: 260 seconds]
16:36 -!- rafspiny [~raffaele@a120210.upc-a.chello.nl] has joined #hplusroadmap
17:36 < hprmbridge> kanzure> "Impaired DNA double-strand break repair contributes to the age-associated rise of genomic instability in humans" https://pmc.ncbi.nlm.nih.gov/articles/PMC5071568/ "an age-related decrease of NHEJ efficiency"
17:37 < hprmbridge> kanzure> "FOXO3 activity also declines with age, which leads to decreased dsDNA repair and autophagy. "
17:38 < hprmbridge> kanzure> "Changes in DNA repair during aging" https://pmc.ncbi.nlm.nih.gov/articles/PMC2190694/
19:18 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
19:18 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
19:19 < fenn> "if you and your partner are retards, then there's not much that can be done with selection" <- not correct, please come up with a better way of stating what you really mean and practice using it
19:20 < fenn> if you and your partner completely lack high-IQ alleles, sure, it would be difficult or impossible to achieve a high-IQ offspring with selection alone
19:21 < fenn> if you're both carriers of dominant deleterious mutations that can simply be avoided, it would be shameful not to do embryo selection
19:21 < fenn> i'm not even sure if "high-IQ alleles" is an accurate way of phrasing this
19:35 < fenn> the 3d mitochondria could be done with plenoptic imaging and/or center of gaussian superresolution like PALM/STORM with narrow depth of field
19:36 < fenn> i think i've seen similar stuff on neuron slice SEM reconstructions
19:44 < fenn> i eat a lot of cheese now
19:46 < fenn> it turns out dutch cheese has a lot of K2 and is probably the best source in a diet. think gouda, jersey, emmenthaler, camembert
19:46 < fenn> not sure if similarly named cheeses in the west are adequate, since it's microbially created during the aging process, and they might use different strains
19:47 < fenn> enough different european cheeses have K2 that i think it was artificially selected for somehow
19:47 < fenn> excess K2 beyond what was present as K1 in the milk
19:48  * fenn pokes RangerMauve to read the cheese blither
19:49 < fenn> https://fennetic.net/irc/vitamin_k2_in_cheese_radar_linear_zoomed.svg and https://fennetic.net/irc/vitamin_k2_in_cheese_radar_european_cheeses.svg from https://pmc.ncbi.nlm.nih.gov/articles/PMC5946231/
19:50 < fenn> i think "cheddar" from USA is not the same at all, nor "munster" 
19:51 < fenn> er. excess K2 beyond what was present as K2 in the milk*
19:51 < fenn> it could be K1 derived for all i know
20:02 < L29Ah> MuaddibLLM: how much is K2 in gouda?
20:02 < MuaddibLLM> Roughly on the order of 50–80 µg of K2 per 100 g for properly aged Dutch gouda (PMC5946231 puts 13‑week 50% fat gouda at ~65 µg/100 g total menaquinones), with younger/US-style stuff often a lot lower.
20:03  * L29Ah sticks to his pills
20:03 < fenn> i think there's too much alterna-health focus on "microplastics" (whatever that is) and not enough on "fragrance" (whatever that is)
20:04 < fenn> fragrances are designed to be bioactive, they're literally selected to bind to pheromone receptors and alter mood
20:05 < fenn> polymers are mostly designed to be as inert as possible
20:05 < L29Ah> stinky shit (and colorful shit) is frowned upon among the eco-conscious, but not loudly enough
20:05 < fenn> you can't avoid it
20:05 < L29Ah> it is virtually impossible here to buy detergent without these
20:06 < L29Ah> MuaddibLLM: how problematic is titanium dioxide used as pills coloring?
20:06 < fenn> i see biodiesel soap in your future...
20:06 < MuaddibLLM> Orally it’s pretty low on my worry list: TiO₂ is basically inert bulk-wise, the EFSA/EU ban was about chronic food use with nano-sized particles and possible genotoxicity, but the tiny amounts in tablet coatings look like a small, not-zero-but-not-huge risk compared to stuff like fragrances/air pollution, so avoid it if it’s easy, don’t panic if it’s not.
20:09 < fenn> bryan johnson claims to have reduced his blood and semen microplastics count by ~ an order of magnitude by not doing dumb stuff (microwaving in plastic)
20:09 < fenn> and sauna, which shows the immune system can be trained to remove it
20:09 < fenn> perhaps there's a *-mab for that
20:10 < L29Ah> is he into plasma dumping?
20:10 < fenn> yes but not hardcore enough
20:10 < fenn> i haven't read anything from him first-hand actually
20:11 < fenn> my info is out of date, in early 2025 he did total plasma exchange
20:12 < L29Ah> does total plasma exchange mean he was totally out of antibodies for quite some time?
20:13 < fenn> "the most up-to-date information can be found on his X account." no wonder why i can't find anything
20:18 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
20:19 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
20:35 < fenn> he's eating flax :(
20:39 < fenn> not very much of it, good
20:40 < fenn> lots of ALA nuts in general though
20:41 < fenn> his olive oil brand touts peroxide value so you'd think he would understand why it's a bad idea
20:42 < fenn> there is some vegan propaganda probably
20:44 < fenn> wonder why black lentils is his main staple
20:49 < fenn> i found a nutrient profile for black lentils, identical to the USDA nutrient profile for lentils, mature, cooked, boiled
20:50 < fenn> it could be bogus data
23:20 -!- _flood [~flooded@45.87.214.26] has quit [Remote host closed the connection]
23:20 -!- _flood [~flooded@45.87.214.26] has joined #hplusroadmap
23:53 -!- darsie [~darsie@84-113-82-174.cable.dynamic.surfer.at] has joined #hplusroadmap
--- Log closed Wed Dec 03 00:00:55 2025