Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Can we can an update to your blog post lamenting the death of science, now that science is only reason we will have a vaccine for a virus that has killed a quarter-million people in the US? Could also expound on your interpretation of the LIGO data while you're at it.


[flagged]


I did, and it’s pretty clear your gripes about social sciences carry over into taking cheap shots at a bunch of other fields.

2: it’s the literal cause of their death. Don’t know what you’re getting at with trying to compare that to how much science is dying. Sounds like incomparable quantities to me - be the good scientist you want to see in this world!


> it’s the literal cause of their death

Are you sure about that? Last I heard (from Dr. Birx on TV), many states count anyone with a positive test as a “COVID death”, regardless of what the coroner’s report says.

If we counted deaths for any cold the same way, it would be a shocking (perhaps not as big, but still shocking) number.

> be the good scientist you want to see in this world!

Unfortunately there is no place for a good scientist in most fields, you either play the political game or you get frozen out. I would rather fight to tear the whole system down to its foundations, so more people will have the opportunity to be good scientists in the future.


Does no one - by this logic - ever die of AIDS because it’s something else (like pneumonia) that does them in at the end?


That’s an interesting question. Officially the answer is “no”, but what if someone with AIDS contracts COVID-19 and dies? Does that person count toward both statistics? How many statistics can one death count towards?

Now try “diabetes and hypertension”, or “lung cancer”, or “temporary immunodeficiency” and you can start to see how dishonest it is to represent the death toll of this disease as a single number, based on an interpretation that generates the largest possible number, and an interpretation not used with any other respiratory disease.


Is it the bullet or the blood loss that kills me?


Again, an interesting question. You will find out that a concept like “death toll” does not fit neatly into linear or trivially-nonlinear equations.


You are going to be shocked when you learn how flu deaths are counted. Hint: positive test is not even necessary.

And fascinatingly, you don't worry at all about people who died without test missing from statistics.


Flu deaths are estimated, not counted. Often victims are not tested for it at all.

> And fascinatingly, you don't worry at all about people who died without test missing from statistics.

Actually “presumed COVID deaths” are included in some states’ numbers, but yes there is some undercounting too.


I don't understand. You consider the alleged mistakes public health officials who provided guidance about pandemic response to signal the end of basic science, but the successes of vaccine production are exempt from that analysis because it is applied?

How are those public health recommendations less applied than vaccine creation (especially mRNA vaccines, which are only possible today due to a ton of basic research in the last 2 decades)?


To the extent that epidemiology is an applied science, it does not on its own “herald the end of science.”

But even in that field, ludicrously simple and narrow-minded theories that are “tested” by running computer games, and then memory-holed as soon as they are no longer politically useful, are a big part of the problem.


I agree that this is problematic. Can you point to instances where this was found to be true? Or, better yet, evidence this is widespread?

tbh I'm not clear what you mean by a lot of the terms like "memory holed" - do you mean like the work of Bayes or Boole? I love learning about stuff that we learned, then forgot (scurvy, comes to mind).


"memory holed" is a reference to 1984. 'guscost appears to believe that epidemiologists eradicate evidence of their theories and simulation results in a manner comparable to the Ministry of Truth's purging of evidence and rewriting of history.


Instead of active malice to rewrite the truth, can't this just be explained by "people forget over time as new information overwrites old"? Maybe one model for our attention spans is a FIFO queue, organized per topic.

That model can't always work, however - we also know that people's first impressions will continue to be strongly held for quite a bit of time. EG, once a news story is published, retractions lack the same impact.

What other possible explanations are there? I think it's worth considering alternatives, as it seems clear there are multiple relevant ways of looking at the problem.


> What other possible explanations are there?

Corruption. Individual journalists may not have the active intent to do harm with every editorial decision, but there is no way the pattern of things that are “amplified” or “remembered” vs “forgotten” is an accident.




Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: