|
Powershift posted:That explains senior vice president of sanitation and acting divisional operations manager overseeing the structural dynamics of flow, Scruffy. I would like to imagine that in this example, Scruffy is a cat
|
#
?
Sep 25, 2019 18:53
|
|
|
|
| # ? May 23, 2024 03:58 |
|
|
Iron Crowned posted:I would like to imagine that in this example, Scruffy is a cat Scruffy is the janitor. The cat is the executive vice president of auto-hygine, superintendent of clay aggregate redistrubtion, and interim head of non-human resources. He's also the snuggly-wuggliest little furball :3 (I mean Scruffy, of course)
|
|
#
?
Sep 25, 2019 18:58
|
|
|
i just want nasa to build an OS because they landed on the moon with the computing power of a ti-86 and meanwhile silicon valley's greatest hit is "a juicer with drm"
|
|
#
?
Sep 25, 2019 18:59
|
|
|
Peanut President posted:i just want nasa to build an OS because they landed on the moon with the computing power of a ti-86 and meanwhile silicon valley's greatest hit is "a juicer with drm" actually silicon valley's greatest hit was finding a way to track every single interaction people have online and off in order to sell more
|
|
#
?
Sep 25, 2019 19:15
|
|
|
Plank Walker posted:actually silicon valley's greatest hit was finding a way to track every single interaction people have online and off in order to sell more Interestingly enough, by tracking combined searches of medications, scientists can discover previously unknown drug interactions, such as a cluster of people searching "Drug 1 + Drug 2 + symptom" and possibly save lives. they ain't fuckin gonna tho
|
|
#
?
Sep 25, 2019 19:24
|
|
|
Powershift posted:Interestingly enough, by tracking combined searches of medications, scientists can discover previously unknown drug interactions, such as a cluster of people searching "Drug 1 + Drug 2 + symptom" and possibly save lives. that sounds like an excellent way to learn why medical researchers use control groups
|
|
#
?
Sep 25, 2019 19:37
|
|
|
H.P. Hovercraft posted:that sounds like an excellent way to learn why medical researchers use control groups But they can't account for anything. Their trials involve like, 1000 patients. There are 11926 drugs listed in the canadian drug bank. With sterilized access to google search histories involving them, they could be learning all kinds of different interactions they could never determine in a drug trial. They can't account for every other possible condition within patients in a drug trial. https://www.youtube.com/watch?v=avuhY7D71sQ Unfortunately, there's more money in prescribing a third drug to handle the interactions of the first two. Big data CAN actually be used for good. It ain't fuckin gonna be tho
|
|
#
?
Sep 25, 2019 19:46
|
|
|
Powershift posted:They can't account for every other possible condition within patients in a drug trial. people thinking that they have all kindsa side effects that they ascribe to medications they just started taking is exactly why they use control groups, people are dumb as poo poo and eliminating self-reporting bias is a big drat deal
|
|
#
?
Sep 25, 2019 19:54
|
|
|
I like how part of this year's Apple keynote was a bunch of blue ribbon medical studies for rich people with expensive Apple products
|
|
#
?
Sep 25, 2019 20:01
|
|
|
Turns out every single drug is paired with "am i pragnet??" Also not to out myself as a computer toucher, but NASA had much less computing power than a ti-86. The good news is that even incredibly slow computers are much, much better than us at figuring out math.
|
|
#
?
Sep 25, 2019 20:03
|
|
|
H.P. Hovercraft posted:people thinking that they have all kindsa side effects that they ascribe to medications they just started taking is exactly why they use control groups, people are dumb as poo poo and eliminating self-reporting bias is a big drat deal That's where the benefit of the larger pool of users comes in. The difference between anecdotes and statistics. It's something double blind studies couldn't possibly account for.
|
|
#
?
Sep 25, 2019 20:05
|
|
|
Powershift posted:It's something double blind studies couldn't possibly account for. why is this not ringing bells for you
|
|
#
?
Sep 25, 2019 20:06
|
|
|
H.P. Hovercraft posted:why is this not ringing bells for you Why do you think I'm the one who's missing the point?
|
|
#
?
Sep 25, 2019 20:09
|
|
|
Powershift posted:Why do you think I'm the one who's missing the point? because you posted a ted talk about diluting medical research results with google search histories lol
|
|
#
?
Sep 25, 2019 20:10
|
|
|
fyi what this reeks of is from last year when facebook et al wanted to start selling medical data to get around hipaa which is extremely capitalism.txt nevermind that it's also a buncha software people working for big ad companies thinking that they have special insight into how medical research works e: whoops two years ago H.P. Hovercraft has issued a correction as of 20:15 on Sep 25, 2019 |
|
#
?
Sep 25, 2019 20:13
|
|
|
H.P. Hovercraft posted:because you posted a ted talk about diluting medical research results with google search histories lol So you don't think tens of thousands of people, out of millions of people taking a combination of two drugs, experiencing and searching for specific side effects that aren't apparent in the usage of either drug individually is significant? You think drug companies should, in labratory settings, test every combination of commonly used drugs in patients with the most common genetic conditions, and that is the only possible way to get accurate data? This is probably relevant to you. https://www.bmj.com/content/363/bmj.k5094 e: facebook does evil poo poo because that's loving facebook. I'm saying the data currently being used to sell us boner pills could be used in more productive ways. There is no current legislation, and no current way to ensure the data would be used productively if shared. Hence "they ain't fuckin gonna tho" Powershift has issued a correction as of 20:20 on Sep 25, 2019 |
|
#
?
Sep 25, 2019 20:15
|
|
|
Powershift posted:So you don't think tens of thousands of people, out of millions of people taking a combination of two drugs, experiencing and searching for specific side effects that aren't apparent in the usage of either drug individually is significant? lmao no but i do think that google wants to sell their lovely incomplete data to academic hospitals and pharma and call it an advanced medical research tool while breaking down hipaa law
|
|
#
?
Sep 25, 2019 20:23
|
|
|
perhaps instead of relying on poo poo people type into an advertising search engine we should get to a point where "hey doc you prescribed me x and y and i feel z now" is something people can do without getting charged out the rear end for it
|
|
#
?
Sep 25, 2019 20:24
|
|
|
T-man posted:Turns out every single drug is paired with "am i pragnet??" yeah but how did they collect data on all your activities and serve Michael Collins ads while he was orbiting the moon?
|
|
#
?
Sep 25, 2019 20:27
|
|
|
medical researcher with phd in public health: "i must have these webmd search histories, they are essential to determining public policy for this new drug, i will pay top dollar!"
|
|
#
?
Sep 25, 2019 20:27
|
|
|
Zamujasa posted:perhaps instead of relying on poo poo people type into an advertising search engine we should get to a point where "hey doc you prescribed me x and y and i feel z now" is something people can do without getting charged out the rear end for it I'm sorry, where does the shareholder value increase in this model?
|
|
#
?
Sep 25, 2019 20:27
|
|
|
Zamujasa posted:perhaps instead of relying on poo poo people type into an advertising search engine we should get to a point where "hey doc you prescribed me x and y and i feel z now" is something people can do without getting charged out the rear end for it They can under very limited circumstances, but right now it is like Hovercraft said, self reported, and the data isn't compiled outside of drug trials and other studies. Google is being used as an example, because google is what most people ask most of their questions right now. There absolutely should be a publicly run data bank of that sort because drugs.com and webMD are loving dangerous in the hands of most people. A headache means brain cancer and your xanax made you grow tits. What's really frustrating Is that there is no dedicated agency for this type of thing basically anywhere. The FBI and CIA have billion dollar budgets to prevent a couple hundred deaths a year. Adverse drug reactions could be in the hundreds of thousands. Canada has the drug bank, and I've seen my medical record that has every drug i've ever been prescribed in my adult life, but those two aren't associated in any way, and there are extremely limited ways in which my medical information can be used. H.P. Hovercraft posted:medical researcher with phd in public health: "i must have these webmd search histories, they are essential to determining public policy for this new drug, i will pay top dollar!" You're kind of a thick oval office. It seems we agree on the fundamentals, companies shouldn't have ownership of private medical data because it's only used for ads and denying insurance claims. but you can't seem to understand the limitations of controlled trials. If you're game, we can test the hypothesis of that double blind parachute study. Or maybe we can go as a single person who has jumped out of a plane once and lived whether or not they had a parachute on, and just extrapolate that to everybody who ever will jump out of a plane. I sure hope they were wearing one.
|
|
#
?
Sep 25, 2019 20:44
|
|
|
T-man posted:Turns out every single drug is paired with "am i pragnet??" we must do way instain patient> who take thier druggs
|
|
#
?
Sep 25, 2019 20:46
|
|
|
Powershift posted:but you can't seem to understand the limitations of controlled trials. If you're game, we can test the hypothesis of that double blind parachute study. Or maybe we can go as a single person who has jumped out of a plane once and lived whether or not they had a parachute on, and just extrapolate that to everybody who ever will jump out of a plane. ok cool so you have no idea what you're talking about, thank you for clearing that up
|
|
#
?
Sep 25, 2019 20:47
|
|
|
H.P. Hovercraft posted:ok cool so you have no idea what you're talking about, thank you for clearing that up Correct me
|
|
#
?
Sep 25, 2019 20:58
|
|
|
I think drug trials are bad because they're being done for the cheapest amount literally legally allowed, they're mostly being done for big sellers like dickpills or new and improved painmeds, the scientific culture is incredibly toxic and has massive epistemic blindspots you could build a pot shop in, and obviously data reporting is hampered by outdated technology (fax machines motherfuckers) and an inability to share productively.
|
|
#
?
Sep 25, 2019 21:00
|
|
|
Istvun posted:yeah but how did they collect data on all your activities and serve Michael Collins ads while he was orbiting the moon? https://en.wikipedia.org/wiki/STS-51-F quote:In a heavily publicized marketing experiment, astronauts aboard STS-51-F drank carbonated beverages from specially-designed cans provided by competitors Coca-Cola and Pepsi. Post-flight, the astronauts revealed that they preferred Tang, in part because it could be mixed on-orbit with existing chilled-water supplies, whereas there was no dedicated refrigeration equipment on board to chill the cans, which also fizzed excessively in microgravity.
|
|
#
?
Sep 25, 2019 21:09
|
|
|
Powershift posted:Correct me I think the issue is that the data you want to add is completely untrustworthy. You'll get a bunch of people saying things like "when I mixed Aspirin and Oxycodone, it made me more vulnerable to the wifi poisoning my mind with astral projected nightmares" The double-blind part of the controlled studies, as well as actual medical professionals checking out reported symptoms, is vitally important in order to produce results that can be trusted at all. There are definitely limitations on studies that make determining all of the different drug interactions, yes. That doesn't mean the solution is to throw in a bunch of completely unverifiable untrustworthy data.
|
|
#
?
Sep 25, 2019 21:31
|
|
|
A dumb derail, but isn't the idea that you take in aggregate all of the search queries and use that to inform a new round of studies that investigate the specific interaction/side effects? What else would you even do with that data that scares you so much?
|
|
#
?
Sep 25, 2019 21:47
|
|
|
Powershift posted:Correct me ok, well for starters clinical trials aren't black box "throw poo poo at the wall and see what sticks" - they're way more incremental than that the whole point of them being double-blind is it allows them to determine whether any given participant's set of symptoms or observed drug efficacy or medical outcome or whatever is actually happening or not. it's extremely doubtful that some database of medical term search histories would be useful to anyone except the person selling it, because of the myriad biases inherent to such a set of data on top of that is the laughable idea that the Big Data salesman has identified some huge blind spot in how human trials of clinical research are performed, which conveniently can be solved by allowing said Big Data to abrogate patient privacy laws moreover, that these medical outcomes being somehow missed by all of medical research are significantly impactful to the health or quality of life of millions, instead of just hundred of thousands of people googling lightheadedness and nausea and sleeplessness ie "things people not on these drugs also google"
|
|
#
?
Sep 25, 2019 21:47
|
|
|
H.P. Hovercraft posted:ok, well for starters clinical trials aren't black box "throw poo poo at the wall and see what sticks" - they're way more incremental than that Yeah "but what if the computer sees a pattern we don't" is such a persistent mythology. Unstructured data is rarely informative
|
|
#
?
Sep 25, 2019 21:49
|
|
|
T-man posted:I think drug trials are bad because they're being done for the cheapest amount literally legally allowed, they're mostly being done for big sellers like dickpills dude dickpills got discovered by accident
|
|
#
?
Sep 25, 2019 21:50
|
|
|
T-man posted:I think drug trials are bad because they're being done for the cheapest amount literally legally allowed, they're mostly being done for big sellers like dickpills or new and improved painmeds, the scientific culture is incredibly toxic and has massive epistemic blindspots you could build a pot shop in, and obviously data reporting is hampered by outdated technology (fax machines motherfuckers) and an inability to share productively. p sure the big money's been in cancer treatment for quite some time now
|
|
#
?
Sep 25, 2019 21:52
|
|
|
They Epsteined a guy who was set to testify in the largest money laundering scandal in history. Police concluded no foul play immediately after finding his body. https://www.bbc.com/news/business-49824367
|
|
#
?
Sep 25, 2019 21:52
|
|
|
amazing
|
|
#
?
Sep 25, 2019 22:08
|
|
|
Day Man posted:I think the issue is that the data you want to add is completely untrustworthy. You'll get a bunch of people saying things like "when I mixed Aspirin and Oxycodone, it made me more vulnerable to the wifi poisoning my mind with astral projected nightmares" I’m not saying take the google data and start modifying people’s meds off it. I’m saying it’s a piece of the puzzle that largely seems to be missing. It’s a pointer that can’t be discovered through small studies where rare dispositions for drug interactions are statistically insignificant. Russ Altman does follow up connections he finds with further experimentation, on rats which itself is problematic, but a solid conclusion isn’t drawn from the collection of data alone. Through my worst I kept a food journal documenting what I a and how I felt before. Things really went off the rails and could easily include the various drug cocktails I was on. I would gladly give all of that to any pharmaceutical researcher it would be useful to. I don’t give a gently caress who knows that none of their drugs did anything or that if I eat a spinach Salad I shoot flaming hot blood out of my rear end for 3 days. If there was a legitimate mechanism for that information to be collected and analyzed beyond posting on Facebook that you spray painted the bathroom red and they used that post to sell you pressure rated butt plugs, more significant progress could be made. I did the test at Americangut.org so now people know what’s growing inside me and I’ll do it again if I can find a drug that works to help them help somebody else. I’ve tried to get into trials for simponi, Entyvio, FMT, and I’m waiting to hear on esketamine but have been turned down because my collective issues would skew the results, (expect for FMT which was because I couldn’t afford to live in Sydney Australia for 12 weeks. )This is the kind off scenario data from the a collective of users could improve, and it seems the only real avenue to that is through involuntary submissions to the pool of data owned by the companies that already know whether you’re circumcised or not and what you BAC is at any given time of day. H.P. Hovercraft posted:ok, well for starters clinical trials aren't black box "throw poo poo at the wall and see what sticks" - they're way more incremental than that i never said they were? quote:on top of that is the laughable idea that the Big Data salesman has identified some huge blind spot in how human trials of clinical research are performed, which conveniently can be solved by allowing said Big Data to abrogate patient privacy laws Not some big data salesman. A researcher who is actually utilizing the data to solve actual problems and talks regularly about the difficulties in obtaining it, likely because he lacks a profit motive and BVI holding comment to receive it through. http://helix.stanford.edu/ I’m glad to took a break from sitting back and smugly saying “nun uh” to beat these big bad straw men to death through. You really saved us. You’re a champ. H.P. Hovercraft posted:p sure the big money's been in cancer treatment for quite some time now Actually it’s in big rear end nips and I have proof.  You heard it here first. Stuff all your money in between take two largest nipples you can find. Powershift has issued a correction as of 22:25 on Sep 25, 2019 |
|
#
?
Sep 25, 2019 22:23
|
|
|
Tunicate posted:dude dickpills got discovered by accident naw the connection was known, viagra was just better at that then dealing with the heart and they didnt want to waste the money they had already spent
|
|
#
?
Sep 25, 2019 22:24
|
|
|
Powershift posted:
|
|
#
?
Sep 25, 2019 22:30
|
|
|
Knowing Trump he was probably slipping her big dick pills when them nips showed up Stock up on big dick pills now, in 2020 warren is going to usher in the era of women as the dominant sex and everyone is gonna want them big nips pills
|
|
#
?
Sep 25, 2019 22:37
|
|
|
|
| # ? May 23, 2024 03:58 |
|
|
Powershift posted:Not some big data salesman. think again
|
|
#
?
Sep 25, 2019 22:42
|
|