--- Log opened Tue Dec 03 00:00:07 2024
00:16 < fenn> is there something like "Certified Organic by Quality Assurance International" but for replacing the function of the FDA?
00:16 < fenn> an independent third party safety and effectiveness testing organization that stamps its trademark on whatever
00:17 < fenn> and then you fly to barbados to have genetically engineered racoon poop inserted
00:20 < fenn> once there are millions of proven glowing green healthy raccoon children the FDA could be required by politicians to use data from already-conducted offshore trials
00:22 < fenn> also important is structuring the incentives so that med-QAI doesn't just become the successor to FDA with all the same problems, or a racket like Underwriters Laboratories
00:23 < fenn> there's a chicken-and-egg problem and a free rider problem
00:24 < fenn> nobody knows about med-QAI until it's a big thing, so having the stamp in the early days doesn't help a medical research business grow their customer base
00:25 < fenn> if instead the customer (patient) pays for the safety review data, then inevitably it will leak and they can't sell that data for very much
00:26 < fenn> mumble mumble dominant assurance contract
00:48 -!- gl00ten [~gl00ten@bl5-237-194.dsl.telepac.pt] has quit [Ping timeout: 276 seconds]
00:59 -!- Croran [~Croran@user/Croran] has quit [Ping timeout: 252 seconds]
01:11 -!- Croran [~Croran@user/Croran] has joined #hplusroadmap
02:02 < L29Ah> 09:16:59]<fenn> an independent third party safety and effectiveness testing organization that stamps its trademark on whatever
02:02 < L29Ah> FDA doesn't test anything, it only reads your published papers, IIRC
02:03 < L29Ah> maybe inspects you don't have rats running around in your GMP facility
02:31 < fenn> FDA publishes requirements for clinical trials and refuses to certify anything unless they meet those standards, which are too expensive in practice. also they will send guys with guns to kick your door down and seize equipment
02:33 < fenn> Reason was saying there should be some kind of middle ground between third world medical tourism anarchy, and big pharma controlled federal overreach
02:33 < fenn> i'm not aware of any such middle ground, but then i haven't really looked
02:33 < fenn> this is the textbook libertarian solution to this sort of thing
02:44 < L29Ah> yes, and this thing is not appealing for both big pharma (they figured out how to satisfy FDA at tolerable costs and don't want you disrupting their market) and congresscritters (you don't have a lot of money to outbid the big pharma lobbying)
02:46 < L29Ah> so your middle ground boils down to DIYing all the regulated stuff and streamlining equipment to DIY (still there are controlled drugs where you're punished for even trying)
02:46 -!- etc-vi6 [~etc-vi@user/meow/girlchunks] has joined #hplusroadmap
02:47 -!- etc-vi [~etc-vi@user/meow/girlchunks] has quit [Ping timeout: 246 seconds]
02:47 -!- etc-vi6 is now known as etc-vi
03:01 < fenn> i assumed you would just buy all the equipment/materials in $unregulated_country and the customer/patient has to go there, as in the medical tourism scenario. the difference is that it has a stamp of "someone looked at this and decided it was safe or effective enough to put a stamp on it"
03:09 < L29Ah> pretty sure we already have the FDA analogs in every "unregulated" country you can imagine, so wrt stamps you can pick your poison
03:10 < L29Ah> still leaves out things like barges in international waters tho
03:14 < L29Ah> https://www.prrowess.org no news :<
03:16 < L29Ah> https://weber.house.gov/news/documentsingle.aspx?DocumentID=1783 ok apparently it works, but wtf they have no website!?
04:13 < fenn> ^ that's banning barges in "federal waters" to skirt state laws
04:14 < fenn> also it's not a law
04:16 -!- Croran [~Croran@user/Croran] has quit [Ping timeout: 260 seconds]
04:25 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
05:30 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has joined #hplusroadmap
06:57 -!- Guest39 [~Guest39@ipb218e480.dynamic.kabel-deutschland.de] has joined #hplusroadmap
06:58 -!- Guest39 [~Guest39@ipb218e480.dynamic.kabel-deutschland.de] has quit [Client Quit]
09:06 < kanzure> nice article on E11 Bio https://www.bloomberg.com/news/articles/2024-12-03/ex-google-ceo-wants-to-learn-about-brains-by-infecting-them
09:29 < hprmbridge> Eli> Economists have been writing about the FDA since its inception. One idea is just to allow automatic approval in a country if the drug has been approved by any of the OECD countries regulatory authorities. Then, instead of a monopoly, the regulatory agencies of different countries have to compete with each other to find the optimal level of safety vs. efficiency of determining safety.
09:29 < hprmbridge> Eli>
09:29 < hprmbridge> Eli> Also, there should probably be a continuum of approval. For example, I can get multiple types of certifications for a supplement I sell (GMP, cGMP, gluten free, organic, kosher, sugar-free, etc ...). Why can't I do the same thing with medicine? Tell me the p-value, number of people treated, number of product recalls, phases of clinical trials completed (can be more than 3), etc ... And if I lie
09:29 < hprmbridge> Eli> about my certifications, the market can sue me. Currently, this info is hidden from the consumer. Most companies voluntarily want to get the most number of certifications as it increases the potential customer market size.
10:13 -!- flooded [flooded@gateway/vpn/protonvpn/flood/x-43489060] has joined #hplusroadmap
10:19 -!- flooded is now known as _flood
10:33 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
10:34 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
13:07 -!- darsie [~darsie@84-112-12-36.cable.dynamic.surfer.at] has quit [Remote host closed the connection]
13:08 -!- darsie [~darsie@84-112-12-36.cable.dynamic.surfer.at] has joined #hplusroadmap
14:31 < fenn> there's no incentive for the FDA bureaucrats to approve anything, only disincentives
14:32 < fenn> same with any government agency, although it could be different in theory
14:39 < fenn> consumers are too stupid and coddled to be able to think critically about safety and effectiveness, which is by design
14:39 < fenn> you're supposed to go through life believing the authorities and doing what you're told, under the implicit social contract that the state will take care of you when things go wrong
14:39 < fenn> unfortunately that's not how any of this works
14:40 < fenn> so a business plan that relies on consumers wanting information needed to be able to think critically is bound to fail from the start; the necessary population of critical thinkers isn't there
14:40 < fenn> look at examine.com, total flop
14:42 < fenn> write to RFK with your "nutrition facts label for drugs" idea, i dunno, maybe it could go somewhere
14:43 < fenn> i agree that a binary yes/no decision is not ideal
14:53 -!- cslr [~cslr@85-76-12-1-nat.elisa-mobile.fi] has joined #hplusroadmap
14:54 < cslr> spam: I have developed free audio machine learning software to alter brainwaves (you need to listen sounds in the background for 1-3 hours before possible effect (reinforcement learning requires lots of data and iterations)). Requires EEG and *fast* cpu. https://www.blackclinic.net
15:01 < nsh> "Why not to use this software" section should be mandatory imho
15:01 < nsh> (also nice background)
15:03 < nsh> when i used to kidnap people from the nursing home [or whitehouse] and drag them to the rave it was often hard to avoid causing upset. but nowadays you can cite all kinda of science for why they just needed some strong bass to jostle out their neurogunks
15:04 < nsh> should note this down as a business venture if i ever need to increase numberwang
15:04 < nsh> can't go wrong selling mildly dubious scienctically proven way to outstay your mortal welcome to americans 
15:05 < nsh> license to print money while enjoying good music
15:06 < fenn> i officially disapprove of your attitude, mister
15:06  * nsh smiles
15:06 < fenn> all are free to continue their mortal stay, in the republic of fennlandia
15:07 < fenn> er, did i say republic? might want to change that later
15:07 < RangerMauve> c
15:07 < RangerMauve> oops
15:07 < RangerMauve> cslr: Do you dogfood your thing? See anything noticable?
15:08 < nsh> dogfood?
15:09 < nsh> cslr, i think i'd need to understand the theory (and/or see the source code) before using this
15:09 < RangerMauve> "Eat your own dogfood". Sorry, corpo brained
15:09 < nsh> ah right
15:10 < TMA> fenn: remember that dictatorship is a republican institution :)
15:10 < RangerMauve> I'm not clear on why getting the brain waves to a specific target is beneficial. Could see it useful for frying somone's brain tho.
15:10 < nsh> i guess it's using the EEG and some ML to modulate somehow the aural tone malarkeys to be optimal wrt to the biofeedback
15:11 < nsh> but that still leaves open what it's targeting towards modulo that
15:11 < nsh> (as in you can use the general method to make a better torture device too)
15:12 < nsh> oh there's visual simuli also?
15:12 < RangerMauve> Reminds me of Daemon by Daniel Suarez. Good book with EEG based torture
15:12 < RangerMauve> fMRI but same diff
15:14 < nsh> i was thinking of that book the other day
15:15 < RangerMauve> Did you read Change Agent? I found it inspirational.
15:15 < fenn> it could make you mistake your wife for a hat
15:15 < RangerMauve> Oooo also a good book
15:16 < L29Ah> cslr: i'm allergic to binaries so no thx
15:26 < nsh> RangerMauve, oh no, i don't think so. thanks!
15:29 < RangerMauve> nsh: Yeah he did a great job with some of the vibes from Daemon but with more modern tech around bio engineering.
15:30 < RangerMauve> https://libgen.li/edition.php?id=5382312
15:30  * nsh nods
15:31 < fenn> i read some criticism that suarez's schtick was being "two steps into the future" but this backfires when reality catches up
15:32 < RangerMauve> Oh? IMO the stuff in Daemon is still pretty scifi compared to reality. Like we have kill bots deployed more but the AR stuff and brain hacking IRL is still pretty stone age in comparison. The AR stuff is a bit closer these days tho
15:48 < hprmbridge> kanzure> tokenized uranium https://x.com/uranium_io/status/1863952782107005004
15:52 < fenn> a little late to start naming startups *.io
15:57 < hprmbridge> Eli> TBH, I think one of the best public health policies would be to collect data on patients over time. Allow people to volunteer their data for the benefit of society. If only 1% of the population did this, it’s over 3 million people’s data over their lifespan which would make it the largest epidemiology study of all time for very little cost. It would help illuminate negative drug-drug interactions,
15:57 < hprmbridge> Eli> which are currently not really well tested for by the FDA. The data from human health outcomes in this study would save an enormous number of lives.
15:57 < hprmbridge> Eli>
15:57 < hprmbridge> Eli> It will never be done due to onerous legal reasons, but the arguments against it are weak.
15:57 < hprmbridge> Eli>
15:57 < hprmbridge> Eli> I’ve sort of wondered if you could do this anonymously on the blockchain. Use a QR code for your private key at the doctors office. Your doctor would immediately have your electronic health record across all doctors over your adult life. If you receive free health care then there is a strong argument that you should at least provide anonymized data for societal benefit. Socialism requires
15:57 < hprmbridge> Eli> contribution to society and reducing the free rider problem. People supporting socialized health care tend to claim to want it in order for society to save lives. This would save a lot of lives.
15:57 < hprmbridge> Eli>
15:57 < hprmbridge> Eli> I spoke with a guy who was working on electronic medical record data mining on blockchain but he said you can’t even volunteer your data for the benefit of society which is crazy. There’s a lot to be said on this topic but I feel like I’m being long winded already.
16:00 < RangerMauve> Yeah I think a good first step is patient-centric data collection where the patient owns all their data and the health system merge gets given temporary access to it. I have a bunch of designs for how to do it on a technical level, but the issue is that patient data is owned by EMR systems that are focused on centralization and having data live in hospitals or clinics which adds cost everywhere and makes it harder to share the data
16:02 < RangerMauve> auditable zero knowledge computation is a thing and it could be technically possible for users to allow health practitioners and analysts to run analysis on subsets of their data without needing to fully reveal it. No need for blockchains or central EMRs. The difficulty is in the social layer to get there.
16:22 < nsh> working out how to create a cryptographic ecosystem to facilitate a minimal friction transition into a situation of personal data sovereignty by default without breaking continuity of operations for the systems currently relying on its cooption
16:22 < nsh> would go a very long way towards enabling a path towards a better state of affairs
16:24 < nsh> especially when you have the value offering (to counteract the pushback against resovereignisation) to similarly reducing friction for cooperative ventures on compositional predicate basis
16:24 < nsh> i know some people who are thinking about this (in cruder terms) wrt to using chain-based interoperability of digital assets and some licensing considerations around that
16:25 < nsh> (in the context of gaming)
16:26 < nsh> there must be somewhere to meet in the middle between what we sort of pretend to achieve through traditional IP law based licensing and what is rigourously achievable demonstrably with potentially much easier to evolve granular emergent dynamics through cryptographic engineering
16:27 < nsh> it's the bridges between the mires of here-and-now-as-we-find-it and the loft heights of cloud cuckoo land that are the hardest to build
16:27 < nsh> *lofty
16:27 < RangerMauve> IMO all the chain based stuff I saw around health data is about either provenance or speculative investment. Provenance and attestation is valuable but usually it's still corpo / data broker first instead of person first. For making data available I've been messing with some p2p file transfer stuff using my web browser agreogre. Basically you can have data in the users control and they can choose to make temporary encrypted content archives...
16:27 < RangerMauve> ... which a doctor or whatever can scan.
16:28 < nsh> stuff that exists is often banal and mediocre. existence seems to unfairly favour those characteristics
16:28 < RangerMauve> I think bridges are hard because just connecting two EMRs between two healthcare practitioners is thousands of dollars in licensing fees. I really doubt they'd open it to users
16:28 < RangerMauve> Truuu
16:29 < RangerMauve> I'm hoping this one client I'm helping do integrity/provenance stuff for might pivot to some p2p data transfer if we land some funding for it
16:29 < RangerMauve> Probs not the end goal you described but at least a step towards it
16:29 < RangerMauve> I kinda hate working on healthcare tech but it ends up bringing me back somehow :P
16:30  * nsh smiles
16:31 < nsh> the post was a lot about attestation and provenance
16:31 < nsh> and evolved into the global supply chain
16:31 < nsh> so there are worse things to begin with
16:32 < RangerMauve> Hopefully I can pivot to something more scifi once I get this EEG on my head and train some neural nets on it.
17:22 -!- darsie [~darsie@84-112-12-36.cable.dynamic.surfer.at] has quit [Ping timeout: 252 seconds]
17:33 < hprmbridge> Eli> Ultimately the state would need to drive something like my proposal. You have to probably change laws and then incentivize doctors to onboard this system. The government can do both. Medicare is expected to start tapping out in the 2030s and there will be intense pressure to better invest what remains as healthcare rationing occurs.
17:33 < hprmbridge> Eli>
17:33 < hprmbridge> Eli> What is the status of proprietary data formats for EMRs? Did they not pass a law stating data had to be exportable in json/xml or something open source?
17:47 < L29Ah> healthcare rationing always occurs
18:09 -!- flyback [~flyback@2601:540:c701:900:8577:8ec5:8bdd:b7bb] has quit [Ping timeout: 248 seconds]
18:22 -!- flyback [~flyback@2601:540:c701:900:e584:eaf4:a278:88f9] has joined #hplusroadmap
19:02 -!- TMM [hp@amanda.tmm.cx] has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
19:02 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
20:13 -!- TMM [hp@amanda.tmm.cx] has quit [Ping timeout: 248 seconds]
20:22 -!- TMM [hp@amanda.tmm.cx] has joined #hplusroadmap
20:31 -!- L29Ah [~L29Ah@wikipedia/L29Ah] has left #hplusroadmap []
20:47 -!- mxz__ [~mxz@user/mxz] has joined #hplusroadmap
20:47 -!- mxz [~mxz@user/mxz] has quit [Ping timeout: 265 seconds]
20:48 -!- mxz__ is now known as mxz
20:48 -!- mxz_ [~mxz@user/mxz] has quit [Ping timeout: 252 seconds]
20:58 -!- gl00ten [~gl00ten@bl5-237-194.dsl.telepac.pt] has joined #hplusroadmap
21:15 -!- gl00ten [~gl00ten@bl5-237-194.dsl.telepac.pt] has quit [Remote host closed the connection]
21:16 -!- gl00ten [~gl00ten@bl5-237-194.dsl.telepac.pt] has joined #hplusroadmap
21:32 -!- gl00ten [~gl00ten@bl5-237-194.dsl.telepac.pt] has quit [Remote host closed the connection]
21:32 -!- gl00ten [~gl00ten@bl5-237-194.dsl.telepac.pt] has joined #hplusroadmap
21:50 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has quit [Ping timeout: 260 seconds]
21:59 -!- justanotheruser [~justanoth@gateway/tor-sasl/justanotheruser] has joined #hplusroadmap
22:55 -!- mxz_ [~mxz@user/mxz] has joined #hplusroadmap
23:12 -!- Hooloovoo [~Hooloovoo@hax0rbana.org] has quit [Ping timeout: 252 seconds]
23:15 -!- Hooloovoo [~Hooloovoo@hax0rbana.org] has joined #hplusroadmap
--- Log closed Wed Dec 04 00:00:08 2024