Live data from Hacker News

Coronavirus and Credibility

paulgraham.com

841–850 of 968 posts

Re: Coronavirus and Credibility

#841
post #791

Earlier quoted context omitted.

> We are doing this because there are O(10^6) preventable deaths That doesn't appear to be true. Almost everybody who goes on a ventilator dies. Corona virus will infect as many people as any other flu or cold. So it's a valid question - why do we still have a lockdown?

Flu doesn't spread as well, because large number of people take a vaccine. The reason for lockdown is to slow things down, so we don't get to the point where doctors have to decide who to help or who not. Also a lot of people can't connect in their mind that when hospitals are full, it also affects people who didn't even get the virus.

> The reason for lockdown is to slow things down, so we don't get to the point where doctors have to decide who to help or who not.

But one must think that this can't go on indefinitely or for very long time frames. It buys time to reduce the number of "active" cases, but only to ensure that once lockdown is lifted one is able to do tracing and isolation for new cases.

Realistically, we'll have to live with this virus (and the associated risk) for quite a while. The time frames for a "cure" if it is found vary from months for drugs (if those in current trials prove to be useful) to years for a vaccine (which like the drugs may not be effective).

You can keep people holed up for a few months at best. You won't be able to do so for one or more years, unless you want to face severe consequences (and I'm not talking about the economy, I'm talking of long-term psychological effects).

Re: Coronavirus and Credibility

#842
post #534

Earlier quoted context omitted.

> the minimal predictions in this report Those minimal predictions don't justify spending trillions of dollars on CO2 mitigation. They basically say "the climate is going to change, so be ready to deal with it".

This is a false economy. Several studies have demonstrated that it is cheaper to avert climate change than to deal with the effects. Switching to low- and zero-carbon fuels, achieving greater efficiencies and so on, is expensive, but still a lot cheaper than e.g. moving coastal cities or engineering mega-projects to combat flooding, dealing with huge numbers of migrants, dealing with massive changes in agriculture, w…

> Several studies have demonstrated that it is cheaper to avert climate change than to deal with the effects.

Studies based on climate models that are overpredicting warming, plus economic and sociological models that have even less predictive power.

> moving coastal cities or engineering mega-projects to combat flooding

If your coastal city has a problem with sea level rising a couple of feet, your coastal city has already had a problem for a century or more, and you ignored it. How is that all of a sudden a climate emergency?

> your problem has not gone away

I've already said climate change is something we're always going to have to deal with, just as humans have had to deal with it during all of human history. The idea that we can magically stop the climate from changing ever again is ridiculous.

Re: Coronavirus and Credibility

#843
post #524

Earlier quoted context omitted.

> the IPCC report from 1990 Which overpredicted the warming that has occurred from 1990 to now (though its predictions overall were certainly much milder than the catastrophic ones that are being trumpeted as reasons to spend trillions of dollars on CO2 mitigation). Not to mention that there have been four more assessment reports since then (and the sixth is in the works now), and the more alarmist ones (in particula…

Yes, but it's not like the 1990 model was massively wrong or something. The (now historical) observations are riding the lower end of the predicted range from 1990, but even the low estimates still predict a 3C temperature rise by the end of the century. All models are wrong and some are useful. Even a 3C rise would be pretty bad. Just because some of the 6 or 7C predictions (high end of range) don't appear to be com…

> The (now historical) observations are riding the lower end of the predicted range from 1990

No, they're lower than the low end of the predicted range. Plus, the low end of the predicted range was based on assumed CO2 emissions that were significantly lower than what has actually occurred since then. So we have more CO2 than predicted and less warming than predicted even from the lower than actual CO2.

> even the low estimates still predict a 3C temperature rise by the end of the century

That's not 3C rise from where we are now. It's not even 3C rise from where we were in 1990. It's 3C rise from the "pre-industrial" temperature, i.e., the mid to late 1700s, i.e., in the middle of the Little Ice Age. Temperatures have already risen (largely due simply to coming out of the Little Ice Age) by about 1.6C since then, so that low estimate is about 1.4C rise from now to 2100. And, as noted, actual temperature rise is running below the low end estimate. We're looking at roughly a degree C rise in a century. The idea that humans cannot adapt to that so we have to spend trillions of dollars to stop emitting CO2 right now is laughable.

> Just because some of the 6 or 7C predictions (high end of range) don't appear to be coming true we should just ignore the rest of the models?

We should not be using the models to drive public policy decisions with multi-trillion dollar costs. They're simply not good enough for that.

Re: Coronavirus and Credibility

#844
post #392

Earlier quoted context omitted.

What does "taken seriously" mean? We don't even know that from a prediction that vague. If the prediction is "we can expect a serious epidemic once every 10 years", then we have something definite that we can plan for, or at least discuss the costs of planning for. There's no way to discuss the costs of planning for "this will happen, some time in the future" without more detail.

Seems like the only correct response to that level of prediction is increasing investment in studying the problem. Which has largely happened in a lot of areas... except for the people who think the correct response is to ignore any prediction that isn't perfectly specific?

> increasing investment in studying the problem.

That would be nice, yes.

> except for the people who think the correct response is to ignore any prediction that isn't perfectly specific?

If a prediction isn't giving you any useful, actionable information, you can't act on it.

That doesn't necessarily mean you do nothing. If our public policy decision is "well, we don't really know how often epidemics like this will occur, but just to ballpark, let's plan on one every 10 years and see what we would need to do to prepare", that's fine. Just don't say we need to do that because the experts made a prediction. It's just a best judgment call based on very incomplete information.

Re: Coronavirus and Credibility

#845

Earlier quoted context omitted.

That the media have (with some regularity) promoted the most catastrophic predictions, predictions that were subsequently falsified, is not exactly news.

So you're going to ignore actual climate experts because you don't like how the media has portrayed the issue?

No, that's not the argument.

Again, my root question is "why have there been so many prominent predictions of climate disaster that have failed to materialize?"

The sources for these predictions have largely been the media, but the sources are generally respected institutions or individuals, and the articles have claimed the backing of scientific authority. To what extent have these failed predictions stemmed from the media, and to what extent have they stemmed from the scientific authorities themselves?

A good part of the blame certainly rests with media. The media is motivated by what is advertiser friendly, and fear sells. However, I do not think the scientific establishment is without fault here.

No doubt part of the problem is the sheer complexity of the system they are studying, and perhaps premature confidence in the methodology of using computer models in the study of complex systems. But it also seems that a bias towards certain policy measures has led some scientists to overstate the confidence, severity, or urgency of predictions.

I do not mean to imply some sort of conspiracy theory here. I simply think that, psychologically, scientists tend to be a of a sort more attuned to "far problems". This tendency allows scientists to sometimes correctly anticipate threats, such as those from a global pandemic, but can also lead to a false equivalence between future uncertain costs and present certain ones.

I am not ignoring experts, nor am I claiming that they are completely wrong about global warming. I would more consider myself in the company of folks such as economist David Friedman or climatologist Judith Curry (neither of whom are in the employ of Big Oil) who both express doubts about the degree of consensus[0] and the severity and certainty of the warming and the value of drastic action in the present[1].

[0]https://daviddfriedman.blogspot.com/2014/02/a-climate-falseh... [1]https://curryja.files.wordpress.com/2019/02/curry-testimony-...

Re: Coronavirus and Credibility

#846

Earlier quoted context omitted.

Science isn’t consensus! Proof isn’t determined by “very nearly 100% of scientists agree”. Just ask Copernicus. We don’t vote on scientific proof. Yes evidence leads us to believe climate change is happening, but why and predicting it’s trajectory and effect should be met with scientific standards not scaremongering and “CLIMATE DENIER” hyperbole. If the pursuit of scientific rigour and fact has taught us anything ov…

You are right. It is decided because not only do almost 100% of climate scientists agree, but there in indisputable proof of the basic method of action (CO2, CH4, H20 greenhouse effects) and we can measure that effect in the lab and then correlated it to measured climate change that normalized for various known climate cycles. The consensus in the fields is driven by the overwhelming evidence of the multidisciplinary…

> but there in indisputable proof of the basic method of action (CO2, CH4, H20 greenhouse effects) and we can measure that effect in the lab

Yes, and those measurements tell us that without a large amount of positive feedback, the effect of greenhouse gases alone is too small to worry about.

> and then correlated it to measured climate change that normalized for various known climate cycles

Nope, this has not been done. The climate models cannot reproduce the data at all with just the measured greenhouse effect of the various greenhouse gases. They have to dial in a large amount of positive feedback, and even then the only period of time for which they reproduce the data tolerably well is the latter half of the 20th century. The model outputs don't match the early 20th century warming, they don't match the mid 20th century cooling, and they don't match the early 21st century pause.

The correct, honest scientific conclusion from all of this is that the models are simply not good enough at this point, i.e., we don't understand how the climate works well enough at this point, to support multi-trillion dollar policy decisions.

Re: Coronavirus and Credibility

#847
post #746

Earlier quoted context omitted.

You guys are still doing this..?

Being rational? Yes. This is not smallpox. This isn’t even MERS from a mortality perspective. It’s real, rapid-spreading, and terrifying in how it kills those few who succumb, but people need balanced perspective. All they are getting is impending doom reported on television 24/7 and it’s nothing but. The future is bright. The 1918 Spanish Flu was followed by the roaring 20s!

At best, that’s “straw Vulcan” rationality. The emotional reaction of a crowd is an important part of any world model.

At worst, without all the anxiety you’re opposing, too many people catch it at the same time and the death rate rises to match the hospitalisation rate. When that happens you’re not talking 75 million dead, you’re talking 375 million to 2.25 billion dead depending on which of the enormous range of hospitalisation estimates I look at.

Re: Coronavirus and Credibility

#848
post #842

Earlier quoted context omitted.

This is a false economy. Several studies have demonstrated that it is cheaper to avert climate change than to deal with the effects. Switching to low- and zero-carbon fuels, achieving greater efficiencies and so on, is expensive, but still a lot cheaper than e.g. moving coastal cities or engineering mega-projects to combat flooding, dealing with huge numbers of migrants, dealing with massive changes in agriculture, w…

> Several studies have demonstrated that it is cheaper to avert climate change than to deal with the effects. Studies based on climate models that are overpredicting warming, plus economic and sociological models that have even less predictive power. > moving coastal cities or engineering mega-projects to combat flooding If your coastal city has a problem with sea level rising a couple of feet, your coastal city has…

> Studies based on climate models that are overpredicting warming

Models in general haven't overpredicted warming, as several people in this discussion have tried to educate you.

> If your coastal city has a problem with sea level rising a couple of feet, your coastal city has already had a problem for a century or more, and you ignored it.

I'm not at all clear what your point is here. It looks like you're arguing that because something is changing over the course of several decades it's not as much as a problem. Surely that can't be it though, because that makes no logical sense. From an economic sense it's just as expensive.

> The idea that we can magically stop the climate from changing ever again is ridiculous.

A strawman argument. That is not what I wrote.

The climate is changing faster by far than it has in all human history, and probably in the history of our planet.

Dealing with the effects is also much more difficult now compared to most of human history. When you have 7 or 8 billion people living on the planet, largely in huge cities, it's far more challenging.

It's just a lot easier and cheaper to avoid breaking something in the first place than it is to try to fix it afterwards.

Re: Coronavirus and Credibility

#849

Earlier quoted context omitted.

I don't think the sixth edition of the assessment report is complete. It's "AR6 Climate Change 2021" so I assume it will be released next year. But if it's available somewhere, let me know. "The world is going to end in 12 years..." is a recent famous quote by a prominent Democrat. I'm surprised you haven't heard it. Personally, I bought into the climate alarmism about 30 years ago, and my skepticism is based on seei…

>The world is going to end in 12 years..." " is a recent famous quote by a prominent Democrat. This isn't true, and you should probably cut whatever source told it to you out of your life. The world is going to end if we don't take action within 12 years , which is the gist of what the IPCC report says.

My quote is literally what she said. (She later claimed it was "dry humor plus sarcasm")[1]. Perhaps she meant something more like what you said.

It doesn't matter.

"The world is going to end if we don't take action within 12 years" is also false, and not at all what the IPCC report says. When O'Rourke said something similar, the AP fact checked the claim:[2]

> THE FACTS: There is no scientific consensus, much less unanimity, that the planet only has 12 years to fix the problem.

> “This has been a persistent source of confusion,” agreed Kristie L. Ebi, director of the Center for Health and the Global Environment at the University of Washington in Seattle. “The report never said we only have 12 years left.”

And no, I'm not going to cut the AP and PBS out of my life just because some alarmists are delusional.

1: https://apnews.com/1c663bcb294c1752573f45fc5bf5e0f9/

2: https://www.pbs.org/newshour/politics/ap-fact-check-orourke-...

Re: Coronavirus and Credibility

#850
post #847
post #746

Earlier quoted context omitted.

Being rational? Yes. This is not smallpox. This isn’t even MERS from a mortality perspective. It’s real, rapid-spreading, and terrifying in how it kills those few who succumb, but people need balanced perspective. All they are getting is impending doom reported on television 24/7 and it’s nothing but. The future is bright. The 1918 Spanish Flu was followed by the roaring 20s!

At best, that’s “straw Vulcan” rationality. The emotional reaction of a crowd is an important part of any world model. At worst, without all the anxiety you’re opposing, too many people catch it at the same time and the death rate rises to match the hospitalisation rate. When that happens you’re not talking 75 million dead, you’re talking 375 million to 2.25 billion dead depending on which of the enormous range of ho…

5% to 30% of the world dying?

Yeah, that's bollocks.

At least 80% of people don't even notice they have it (so the BBC has told me today).

Only 5% of cases require ICU, let alone 5% of infections (and if Imperial modelling is right that number is far lower). ICU saves at most half of the people who go in based on CDC numbers.

I have not seen any modelling by any source of even the slightest repute suggesting a 2 billion dead number is even possible. This alarmism makes people roll their eyes at all things COVID-19 and violate lockdowns, stick to the facts, they're bad enough.

Post reply on HN