Live data from Hacker News

YouTube removes interview with professor of medicine on Covid stats and policy

unherd.com

571–580 of 596 posts

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#571

Earlier quoted context omitted.

Nationally, the clearance rate for arson is typically under 23%, meaning the vast majority of arsonists get away with it. In general, clearance rates are pretty bad across most crimes. The homicide clearance rate is typically below 62%, clearance rates for non-fatal shootings is typically below 55%, clearance rates for reported rape is typically below 40%, clearance rates for motor vehicle theft is typically below 20…

> If you want to solve crime problems, you have to explore preventative solutions, namely give people better paths and eliminate paths that empirically lead to only bad destinations. Education is the preventative solution. Censorship is just security through obscurity.

> Education is the preventative solution.

It's a possible preventative solution, but when designing anything, you should never forget that you are not the user. You have to design for the user.

How does the ignorant person who is seeking education know which source to believe? Educated or uneducated, when people don't have the time or prior knowledge to evaluate claims on their own, they look to the sources they already trust, especially when they're scared. In the US, tens of millions of people deeply trust Trump, and many people are afraid of dying from COVID-19. Trump has been regularly praising hydroxychloroquine as a miracle cure ever since an attorney/blockchain enthusiast named Gregory Rigano misrepresenting himself as a Stanford Med School Advisor on Tucker Carlson's show [0] (see too_much_detail below) and misrepresented a bad study to assert hydroxychloroquine had a 100% cure rate. Trump has been telling people to take this drug constantly, claiming he's taking it personally. However, in a study by the VA of hydroxychloroquine's effect on 368 COVID-19 patients, the hydroxychloroquine-only group was associated with increased overall mortality (Rates of death in the HC, HC+AZ, and no HC groups were 27.8%, 22.1%, 11.4%, respectively) [3]. Considering

Many people will die because Trump and the GOP are miseducating people. Why shouldn't YouTube improve the quality of the content on their platform by removing lethal misinformation? YouTube and FaceBook were able to greatly limit the spread of the ChristChurch mass murder (by rejecting video's with matching hashes, looping in human reviewers, and other means) such that I never saw the video, so it's not a technological issue.

too_much_detail: {where he falsely presented himself as a "Stanford Univ Med School Adviser" [1] and claimed a "well-controlled [false], peer-reviewed [false] study carried out by the most eminent infectious disease specialist in the world Didier Raoult MD Ph.D. ... enrolled 40 patients, again a well controlled study [still false, no randomization, no blind, treatment group was from one hospital in Marseille (mean age 51.2), while the control group was from hospitals in southern France (mean age 37.3) [2]] peer-reviewed [still false] study, showed a 100% cure rate [false, 100% of the people who were swabbed every day, which excluded 1 patient that died and 3 that went into the ICU [2]]. The study was released this morning on my Twitter account."}

[0] https://www.youtube.com/watch?v=C4zTt8oLD44

[1] https://www.nytimes.com/2020/05/12/magazine/didier-raoult-hy...

[2] https://www.theguardian.com/world/2020/apr/06/hydroxychloroq...

[3] https://www.medrxiv.org/content/10.1101/2020.04.16.20065920v...

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#572

Earlier quoted context omitted.

Their automated system has a non-zero false-positive rate. It will from time to time accidentally take down legitimate videos, which they restore fairly quickly once you request a manual review, such as in this case. Yet people still jump to conclusion and assume malintent, posting headlines which imply Youtube is intentionally removing these videos. Clickbait.

> which they restore fairly quickly once you request a manual review Seems like an overly optimistic view, based on stories you hear that small publishers don't stand a chance.

Yet people ignore all the cases where it was indeed recovered very quickly, as is the case here.

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#573

Earlier quoted context omitted.

Their automated system has a non-zero false-positive rate. It will from time to time accidentally take down legitimate videos, which they restore fairly quickly once you request a manual review, such as in this case. Yet people still jump to conclusion and assume malintent, posting headlines which imply Youtube is intentionally removing these videos. Clickbait.

I've heard enough of these machine learning buzzword bingos every week when things like this happen. The thing is, I don't care. I just want things like this to never happen again.

Then we should shut down videos on the internet entirely, because 500 hours of video per minute is not something anyone can manually moderate. We would all love to live in an utopia where everything works perfectly and there's a solution to every problem, but that's not realistic.

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#574

> professor of medicine According to this guy's Wikipedia page, he has claimed many times to be a professor of oncology at Imperial College: but he is not, and Imperial College has sought legal options to stop him from making this claim. The page is full of other fairly astonishing stuff, not the least of which is apparent fraud connected with the Lockerbie Bombing. https://en.wikipedia.org/wiki/Karol_Sikora I get th…

this type of victim playing by knowing fraudsters is becoming the modern day 'hack' of the concept of free speech and really needs a counter.

The problem is that there is no counterargument to an argument made in bad faith.

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#575
post #510

Earlier quoted context omitted.

But this favors one recommendation algorithm (time based) over others. What makes ordering by time non-editorial?

Laws don't have to be philosophically pure. A law can define or describe what counts as an editorial algorithm and what does not, and it could exclude simple sorting while including the complicated thing Youtube does to keep you watching. What might be harder is to distinguish sophisticated search algorithms from recommendation algorithms. There's some overlap, but I'm inclined to think that the two are mostly differ…

Indeed, but (at least in the US), you can't really do that without running afoul of the first amendment, presenting content is just a form of speech. Government restrictions on speech are the thing that everyone's afraid of, right?

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#576

Earlier quoted context omitted.

Their automated system has a non-zero false-positive rate. It will from time to time accidentally take down legitimate videos, which they restore fairly quickly once you request a manual review, such as in this case. Yet people still jump to conclusion and assume malintent, posting headlines which imply Youtube is intentionally removing these videos. Clickbait.

No, that isn't what's happening. You're probably not aware of this if you only read particular media outlets but YouTube keep doing this and they keep defending it. It's absolutely intentional and the headlines aren't misleading people, you are. Read the article: "They rejected our appeal to have the video reinstated." - appeals are at least looked at by humans and the humans are affirming they made the right call. M…

> They rejected our appeal to have the video reinstated

Yet the video is up and working just fine...

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#577
post #554

Earlier quoted context omitted.

Advertisers will always have power, and platforms will always decide to publish different types of content. If an advertiser today wants to advertise with pornhub, that is a different editorial decision than choosing to advertise on YouTube I shouldn't have to spell this out because it seems pretty obvious. One can imagine a world where YouTube complies with your regulation by providing YouTube, YouTube for kids, and…

> Advertisers will always have power Not if laws lessen it. Which is my whole point. > If an advertiser today wants to advertise with pornhub, that is a different editorial decision than choosing to advertise on YouTube. I shouldn't have to spell this out because it seems pretty obvious. It's only obvious within a certain puritan culture. In Europe nobody would bat an eye if an advertiser for a, say, fast food store,…

Legislation can't possibly prevent YouTube from sharding into a bunch of disparate publishing entities

In fact, the "break up big tech" groups might laud such a thing

There is no way of framing your regulation so that it does something useful. Instead it will only ever be a burden on smaller distributers and hurt their chances of advertising (because in order to find advertisers, those advertisers will either have to not care about the content or be convinced that the content is pg -- an incredibly onerous technical burden)

This idea that the regulation wouldn't have consequences you don't like because you don't like them is a fantasy

None of your ideas make any sense here

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#578

Earlier quoted context omitted.

People are worried about economies not for some abstract reasons. Bad economies literally kill and have a potential to kill much more people than Covid-19. From what I gather, the scale of current economic damage is seriously compared to the Great Depression in mainstream media[1][2]. Well, the Great Depression brought Nazi into power in Germany[3], WW2 followed and took lives of 70-85 million people[4] to say nothin…

I don't think anyone said worry about the economy was an abstract idea. It obviously has a very real effect on people and their health, it has been well studied. I was simply pointing out that what people will consider more significant in five years time (the effect of the pandemic itself or the subsequent harm to the economy due to the response) depends on who you ask. Furthermore, since we don't know the full exten…

"I was simply pointing out that what people will consider more significant in five years time"

Yes, I think I understand what you're saying.

My point is the harm to the economy directly translates to lives lost too. Bad economy leads to surge in violent crime, opiods addictions, gang violence, lost access to health care...

"... the Great Depression was one of many factors that led to the rise of the Nazis in Germany. The tensions that led to WW2 were brewing for before WW1 even started so don't try to claim that an economic depression caused all those deaths. There are far too many factors to make that claim."

It wasn't the only cause, sure. But it was a huge one, afaik. History isn't a hard science, of course, and I'm not a historian (not even an amateur historian) but this quote from Wikipedia doesn't strike me as particularly controversial:

"The unemployment rate reached nearly 30% in 1932, bolstering support for the Nazi (NSDAP) and Communist (KPD) parties, causing the collapse of the politically centrist Social Democratic Party... Large-scale military spending played a major role in the recovery."[1]

[1] https://en.wikipedia.org/wiki/Great_Depression#Germany

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#579

Earlier quoted context omitted.

My bad, here's a better link. It's hard to find a "mainstream" source because most of the old media is ignoring a lot of newsworthy items these days. https://www.breitbart.com/tech/2020/02/13/rand-paul-blasts-y... It's widely known that Google/Youtube took down any videos where the name of the supposed whistleblower were being mentioned, at least according to a lot of regular content creators who had videos deleted o…

> Google/Youtube took down any videos where the name of the supposed whistleblower were being mentioned Presumably because they didn't want to be exposed to the legal ramifications of broadcasting a crime?

Broadcasting a crime is illegal?

Re: YouTube removes interview with professor of medicine on Covid stats and policy

#580

Earlier quoted context omitted.

Nationally, the clearance rate for arson is typically under 23%, meaning the vast majority of arsonists get away with it. In general, clearance rates are pretty bad across most crimes. The homicide clearance rate is typically below 62%, clearance rates for non-fatal shootings is typically below 55%, clearance rates for reported rape is typically below 40%, clearance rates for motor vehicle theft is typically below 20…

The last bit sounds like the plot of Minority Report. I get that being a homicide researcher you're rather working on the assumption that a non-zero homicide rate is a problem that can be solved with a sufficiently clever policy. What if it's just a natural fact, like the laws of gravity? What possible solution could there be to homicide as a category of crimes beyond literally being able to see into the future or pe…

> What possible solution could there be to homicide as a category of crimes beyond literally being able to see into the future or people's brains, a la Minority Report or Black Mirror? Isn't there a risk the mentality of a "war on homicide" (you talk of surrender) is actually more dangerous than the problem itself?

First, I'm going to plant the goalposts. A homicide rate of 0 would be nice, but it's not realistic for any large population. Even in the very well educated, upper-middle class Detroit suburb I grew up in, there was a homicide every 10 years or so (the most recent one was actually done by the dad of one of my friends/a soccer coach of mine; he had a great reputation/was a philanthropist/was president of the local Rotary club, but apparently he was also a pillar of the local S&M community and his wife found out so he had her killed). The goal is to reduce the rate of homicides. And 5 minutes of thought will reveal that that's absolutely possible.

Like many US cities, Chicago is sharply segregated [0]. That link also includes maps of counts of narcotics arrests, homicides, and shootings, and if you cross-reference the narcotics, homicide, shooting, and black residents as a percent of census tract population maps, you'll see they're highly correlated. The vast majority of the violence is in the black neighborhoods. It's an uncomfortable fact, but stay with me. As a liberal, seeing this was initially like a punch to the gut, but it makes sense when you ask the question "why would anyone choose to live in these neighborhoods", and very few choose to live there, rather, most people can't leave. In 2016 (and for the US), the median net wealth of black families was about $17k, while the median net wealth of white families is about $171k, 10x the median net wealth of black families. White people have the funds to climb to better areas with better schools with better economic opportunities and they can afford to buy real estate in good or improving areas which provides those white people with a durable store of value (that has typically appreciated at a compounding rate over the past century), but this increases rents and this pushes black people to areas with lower rents, worse schools, brain-damaging pollution (eg lead paint, lead contaminated water [2][3], air pollution from industry), and much higher rates of violence.

Now to get to the question: what's driving the violence? And it's largely economic. A massive portion of the shootings and homicides are committed by gang members (really neighborhood cliques now, with little or no hierarchy and on the order of 10 members on average) who are in cliques that sell drugs, and from an analysis of cleared homicides and shootings, the shooter and victim are very frequently in geographically adjacent cliques that both sell drugs. In the 1920's, alcohol became a black market good and there was a lot of violence between alcohol traffickers that evaporated when alcohol returned to being a regular market good.

So, do I have to be an omniscient precog from Minority Report to know that [legalizing drugs, eliminating lead exposure, and improving the economic opportunities for black youths by improving the quality of education in all neighborhoods to the level white people enjoy] would drastically reduce the homicide rate? No. Those are obvious steps.

> The source wasn't YouTube. It was just her social circle.

That's kind of like saying "guns don't kill people, blood loss kills people". How do the ideas initially get into the minds of people in her social circle? Per Facebook, a lot of misinformation is financially motivated [4] or geopolitical (from a certain malicious surveillance state that wants time to catch up on 5G development) [5] and spread widely through public groups. If the misinformation was stopped from spreading publicly, it would reach the last leg of the trip to the private group/WhatsApp group phase far less frequently and lead to far less destruction.

> The article also lumped people protesting against lockdown (normal, expected)

No, it's not normal, and the protests have been very small and sparsely attended, but they've been massively amplified by the media [6]. As of May 7th, 68% of Americans (10957 people surveyed) were more worried about restrictions being lifted too early vs 31% that were worried about the lifting too late. And saying you're more worried about the economic harms than the virus is way more normal than actually going into public and gathering in a group with other people who are all intentionally rejecting safety recommendations.

> For these conspiracy theories to be successfully debunked will require the debunkers to stop arguing from authority. People don't trust the authorities.

Yeah, it's not possible to change a conspiracy theorist's mind with facts, because any inconvenient fact can just be interpreted as another part of the conspiracy. To those susceptible to conspiracy theories, there is no known cure to the infection. The only way to avoid outbreaks is to stop transmission.

[0] https://imgur.com/a/K9IJKHr

[1] https://www.brookings.edu/blog/up-front/2020/02/27/examining...

[2] https://www.cambridge.org/core/services/aop-cambridge-core/c...

[3] https://en.wikipedia.org/wiki/Lead%E2%80%93crime_hypothesis

[4] https://www.facebook.com/facebookmedia/blog/working-to-stop-...

[5] https://www.nytimes.com/2019/05/12/science/5g-phone-safety-h...

[6] https://www.vox.com/2020/5/10/21252583/coronavirus-lockdown-...

[7] https://www.pewresearch.org/fact-tank/2020/05/07/americans-r...

Post reply on HN