COVID-19/Iluvalar/politics

I wrote my script in August 2020 simulating my best understanding of science. It works great, it predicted the right amplitude of the second wave 6 months in advance. If you doubt of it, feel free to go visit the page and compare my predictions with reality all you want. However, I can't help but to witness a large discrepancy between my best understanding of science and what is commonly believed. And I collected reasons why the common beliefs diverge for mine. Now in a comical irreconcilable way. This is a psychological analysis of what happened around us. Some of my statement might shock a bit. I must confess, it's kind of the goal of this page.

General hatred for Chinese's dictatorship

edit

An appreciable amount of organizations are not fans of the Chinese communist party (CCP). They find convenient to blame China for the pandemics. They are motivated by political reasons to do so, not science. This cause our first bias toward covid's perception. In order to blame china, ones need the pandemic to be dangerous. Fairly dangerous as a matter of fact. I'm not saying that all China's enemies met in a meeting room to plan a smearing campaign. Nor that they even aware of the phenomenon, but it cause a lot of people secretly wishing that covid is a bit more dangerous.

What China did right

edit

The first few cases of covid-19 in the hospital at Wuhan were dying at ~15%. They were healthy 40yo men and they had symptoms akin to diarrhea, which have been documented, but not common for Covid-19. They should have a CFR of ~0.1% not 15%. Something else happened to those men, the doctors didn't know what to do with it. They detected Covid-19 for the first time, but it was not the main cause of those deaths as far as I can tell. There have been a resistance in the hierarchy to reveal that there was a deadly virus with 15% CFR from the small 11 men samples of the first weeks. They DID covered up the 15% deadly hypothesis and it was the right thing to do.

The WHO

edit

The World Health Organization failed pretty hard in the early stages of the pandemic. They changed the definition of pandemics after the H1N1 early warning fiasco. Now they were to only declare a pandemics once it was detected on each continents. The insane flaw with this idea is that countries around the world were counting on the WHO for a warning, in order to MAKE those tests in the first place. We needed to mass produce the tests as soon as the science determined that Wuhan was full of cases. For some inexplicable reason they had this incredible source : "We estimated that 75 815 individuals (95% CrI 37 304–130 330) individuals had been infected in Greater Wuhan as of Jan 25, 2020." [1]. But waited until 11 March 2020 to declare the pandemic. What they were thinking exactly ? That the 75k infected people that was building up for months were quietly all sitting in Wuhan city ? Why waiting 40 extra days before telling countries to start testing ? I can't...

"Better safe then sorry" creep

edit

It's better to prepare of 20% more deaths and be wrong, then do nothing and face the consequence. Right ? But who should add that 20% margin ? The civilian who go take a test even if he don't feel sick ? The test makers who intentionally make the test with more false positive then false negative ? The lab technician who spin the sample at 10k RPM instead of 8k just to be sure ? Should the scientist who estimate the total amount of infection lean of the "safe" side ? Or is it the scientist who later will calculate the IFR from the infection estimation ? The advisor who is hired by the different government to read and comprehend those study would do nightmare at nights if he don't err on the safe side. Surely no one would blame the government itself for preparing for the worst ! And surely no one will listen to all those recommendations, better ask for 2m instead of 1 and have them all put a mask instead of trying to explain to them that they spit too much on each other. And of course those recommendation should be followed precisely by local authority even if they sound a bit extreme.

I count 9 steps in my demonstration here. It means 500% measures instead of 100% if everyone in the loop do it. Too many individual steps by too many individual people cause a bloat. We needed one centralized entity that would process all the information and gives ALL the results openly so they can be reexamined. (looking at you again WHO, you failed us so bad...)

I also want to say that it's only an illusion, you probably noted and smile dear reader as you read this chapter. A direct death from covid look certainly more menacing then forcing a quarantine on 20 000 people. But is it really ? The exercise here was to balance BOTH side of the equation for BOTH to be as bad than the other. There was no direction to lean toward as the governments. If anything, they should have been slow to implement anything. but it's not what happened.

Never Trumper politics

edit

In late March, Deborah Birx and Trump came with an estimation of 100k-240k death in the first wave. In reality, the first wave did 172,978 deaths by September 1st. The political attacks came from every angle possible... It was insane to see 100k deaths as a "victory", it was all trump fault. It was incompetent to ignore the worst scenario at 2+ millions death and more measure was necessary. it was bad and racist to close the airport before the advice of the WHO. The airports should have been closed much earlier. Testing people at airports is a breach of their freedom. More tests are needed. Bla bla bla... In retrospect, his estimation was just right. Estimation 100k-240k, Reality: 173k. In retrospect, retrospection is what medias lack. They played a massive role into confusing the public, and they are not going to admit it.

I'm a Quebec separatist Canadian who is leaning toward moderate liberals. I have no personal interest in American politics.

Big tech and AI safety

edit

On March 16, Microsoft, Facebook, Google, LinkedIn, Reddit, Twitter and YouTube. [2] Launched AIs in order to "throttle" misinformation. I do not know which architecture was in use in each individual company. But Gpt-2 was open source and freely available and gpt-3 was about to be announced by open-ai. It was the first time in history where an AI had this kind of interactions with humans. It raise AI safety concerns.

Specifically that day, google stop giving appropriate search results if they were saying that covid-19 is false or is a hoax. I truly understand the intention behind. But indeed that 15% death rate horror virus which was covered by China government didn't exist. It was a good rhetorical argument which was perfectly adequate to counter that overly dangerous version of covid-19 which never existed.

The intention behind this choice was to get people to accept the quarantine, masks and distanciation in order to prevent the second waves. The intention was good, but retrospection is also great.

The New-York crisis

edit

Between march 15 and April 25 or so, 15% of new-yorkers got covid at once. It forced the WHO to recognize the pandemic, but it was WAY TOO LATE to start mass producing the tests. The first problem was the media confusing IFR with CFR. See my main page COVID-19/Iluvalar for more about the "initial misconception". By mid-march, the media started explaining it themselves, but they never mentionned the obvious retrospect meaning of it : Covid, couldnt just appear on march 15, it had to grow to about 1% of all the population before that point for the data to make any sens. It created a wave of panic in the city which killed a lot of people. Lots of nurses were terrorized by a 5% CFR and were wearing nearly hazmat suit just to go turn a patient in his bed. If they were going at all. We have archive of news from that era [3]. Nursed scrambling for PPE when 5% of new-yorkers outside the hospital was infectuous anyway... There is no doubt that the quality of the service to the patient suffered a lot from the stacked amount of misinformation already in place.

I heard that Andrew Cuomo also took bad decisions as well. Making things even worst wikipedia:New_York_COVID-19_nursing_home_scandal.

Finally, the age pyramid in new-york is different, I'm not from the region, but I'm guessing it's a nice place for retirement. Tons of services in a short distance.

These 3 factors drastically inflated the death toll in New-York. I don't know why so many people took it as the base value including Wikipedia when obviously it's the worst case scenario. Specially nearly 2 years after, it still stick as a huge outlier.

Big tech bis

edit

While New-York was reaching his peak and it became clear that Covid was going to be worst then flu. Big Tech moved it bar from throttling "covid is a hoax" to throttling "Covid is a cold". Most people don't even know yet that those AI exists. We have, in essence AI moderators who reads everything on the internet and throttle free speech. I don't think this bar was reasonable, it prevented REAL discussion about comparing covid to other breathing illness. The word "covid" became equivalent to a swear word in youtube for exemple.

Impact on media

edit

It's been documented already, not only youtuber avoided the subject all together. No one knows where the bar for censorship is, those AIs are a black box. It's a new type of intelligence that are not quite human. All we know for sure is that it read every single zombie apocalypse books and novel on google book and use it to make decisions. The AIs are NOT 100% accurate, it typically turns around 82%-90%.

Any news outlets, being paper or TV who would attempted to talk about Covid in a moderated way was running a serious risk of being "throttled" which for any media that attempt to be competitive with ad revenue basically amount to suicide. No one looked interested in those content for the same reason. No one had any friends or family on twitter and facebook who questioned the narrative.

This is how human beings works. If no one questioned it, it must be true. Specially if you are not an expert. Our collective intelligence is not human anymore, we ARE cyborgs in that regard, And that cybernetic part of us is not programmed to be relevant or seeking truth anymore, the cybernetic part of our intelligence is now coded to enforce the fact that Covid is dangerous.

And it is dangerous ! I personnaly predicted on April 7 2019 13% more deaths that it actually did in reality.