Hacker News Reader

Comment by Shank | Hacker News Reader

This is worth reading purely for the writing alone. Scott Alexander is a treasure:

> As best I can tell, this is some kind of Egyptian trial. It might or might not be an RCT; it says stuff like “Patients were self-allocated to the treatment groups; the first 3 days of the week for the intervention arm while the other 3 days for symptomatic treatment”. Were they self-allocated in the sense that they got to choose? Doesn’t that mean it’s not random? Aren’t there seven days in a week? These are among the many questions that Elalfy et al do not answer for us.

View on HN

> This is worth reading purely for the writing alone. Scott Alexander is a treasure

This post is a breath of fresh air for the blog and feels refreshingly well-researched. Between that and the unquestionable popularity of COVID-related discussion, it’s no wonder this article is popular.

Scott is undeniably a good blog writer with an entertaining writing style that resonates with his reader base. However, I’ve become disillusioned with a lot of his other recent writings like the “Great Families” post in which he presented a lot of conjecture that pandered to his core readership without any research or evidence to support it. Specifically, his position that genetics are the primary explanation for “Great Families” while ignoring obvious heritability of things like wealth and social status. Even the old Slate Star Codex and new Astral Codex Ten communities started calling him out on it after that article.

Scott’s writings are undeniably entertaining and usually quite fun to read, but I’ve long since learned to read them more critically because his writings have a way of projecting a sort of humble confidence while either hedging any statements so much that little is actually communicated, or sometimes inserting conjecture or feelings with a lot of linguistic window dressing such that it’s easy to mistake it for fact.

I hope this article is a sign that the writings are shifting more toward the old-school, heavily-researched style that made the blog so popular.

> ignoring obvious heritability of things like wealth and social status

He addresses that in the section section - besides the intro, he leads with it:

"the next generation becomes semi-famous fashion designers and TV personalities and journalists, which seem like typical jobs for people who are well-connected and good at performing class, but don't need to be amazingly bright. Sometimes they become politicians, another job which benefits from lots of name recognition.

But I've tried to avoid mentioning these careers, and focus on actually impressive achievements that are hard to fake. And also, none of these families except the Tagores were fantastically rich; there are thousands or millions of families richer than they are who don’t have any of their accomplishments. For example, Cornelius Vanderbilt's many descendants are famous only for being very rich and doing rich people things very well (one of them won a yachting prize; another was an art collector; a third was Anderson Cooper)."

True, but that's an example of the endless hedging I mentioned: His writing style is very good at throwing in little counter examples here and there to give a "both sides" type impression, but then he pokes holes in the counter-arguments in the build-up to the conclusion he actually wants to present.

If you read a few paragraphs past your quote, Scott comes right out and declares the "genetics" is the explanation that he thinks is correct:

> The other obvious answer is “genetics!” I think this one is right

He then applies some softball counter-arguments to the genetics argument to hedge and give the "both sides" treatment. However, the article is clearly structured to give a tour of possible arguments while subtly giving the most credence and the least hedging to the genetics argument. Combined with the not-so-subtle declaration from Scott that "I think this one is right" in the middle of the article, he expertly walks the reader into the conclusion he wants while making them feel like they came to that conclusion on their own after an objective tour of the research.

Are you accusing him of anything else than having an opinion, and arguing for it?

I think the fact is that he presents himself as speaking from a highly informed, relatively unbiased presentation of a subject he finds interesting. In fact, he is arguing for a position. Reading an article meant to be informative vs reading an article meant to convince you of a position are two very different things.

The ideal is that an author is unbiased before seeing the evidence, then forms an opinion based on the strength of the evidence. An article that chronicles such a journey, presenting an opinion but also the strongest counter-arguments to that opinion, is about the best one can hope for in scientific writing.

I mean, it's hard to be informed and also not have a position. I think the best you can do is read an article about the information that convinced the author (and the information that made them less confident).

Most of the writing is fantastic, but I'm very disappointed by the take that "everyone was wrong", by the implication that doctors didn't know they should treat for worms before giving immunosuppressants. Guess what, doctors in the developing world already knew that, it was always the standard practice on Brazilian hospitals. In fact, ivermectin was in very high demand in hospitals before it was suspected to be a Covid19 treatment. That whole paragraph reads like some really bad exceptionalism (it it doesn't exist here it doesn't exist).

So why hasn’t that come up before in these studies?

Because it was not related to Covid. If someone with heart problems is admitted with Covid they will receive their heart medications in the hospital, that doesn't mean that now heart medication is a Covid medication, so doesn't get mentioned in studies. The same goes for dewormer, it was in normal use against worms. Only after people started believing that it could be a treatment for Covid, that it started being studied for that (with different protocols). And then we found out that it is just a dewormer like we thought in the beginning.

Except in this case the dewormer is the central focus of the study. The relationship between worms in the population and its influence on covid outcomes makes that highly relevant to the publications. If it's known to be present in the general public, then not controlling for that in your study is just shoddy science.

More likely: a case of wishful thinking. Need miracle cure, found miracle cure, look no further.

Wishful thinking IS shoddy science

but it is essential to good engineering

Ivermectin being cheap and widely available is really important here. Something like that- instead of a novel vaccine made with novel tech in very specific first world factories- is very attractive to cash strapped governments and public health systems.

Of course. But that doesn't mean it exists.

Is it standard practice to test for worms with every COVID patient that comes in?

Just an fyi, worms/parasites generally suppress the immune system.

The everyone-was-wrong part was about scientists missing the worm connection when looking at the aggregate data, not about practicing doctors and worm treatments.

I actually come from one of those so-called Great Families. Three generations back, there were 3 siblings who all had national or smaller success in creative, financial, and mathematical fields, including one who won a nobel prize. Many of their descendants also had successes, although of a more modest variety.

However, in my generation, there were lots of money problems. And they did what money problems tend to do -- delay, deny, disappoint, cause health problems, and make everything harder.

The "brains," such as they were, were still there. If we take the usual markers* of intelligence there were Ivy League admissions, Fulbright scholarships, and so on. But for me at least, not having affordable healthcare at age 20 impacted my life much more in a negative way than getting a 5 on the Calculus AP impacted it in a positive way.

I'm much less interested in the concept of inherited genius than I am in universal health insurance and free college.

*Which are themselves problematic, and often merely markers of class, but it's the best I can do here for a HN comment

This is super interesting, because it actually makes the case for 'old money' taking better care of the rest of the world, not for the benefit of the rest of the world but for their downstream.

Sometimes it skips a generation.

Oh, no wonder this was so entertaining to read, didn't even catch that this was astral codex. I laughed out loud at Cadegiani et al:

> A crazy person decided to put his patients on every weird medication he could think of, and 585 subjects ended up on a combination of ivermectin, hydroxychloroquine, azithromycin, and nitazoxanide, with dutasteride and spironolactone "optionally offered" and vitamin D, vitamin C, zinc, apixaban, rivaraxoban, enoxaparin, and glucocorticoids "added according to clinical judgment". There was no control group, but the author helpfully designated some random patients in his area as a sort-of-control, and then synthetically generated a second control group based on “a precise estimative based on a thorough and structured review of articles indexed in PubMed and MEDLINE and statements by official government agencies and specific medical societies”.

Spoilers ahead, but I loved this pun:

> The fraud-hunters have examined this paper closely and are unable to find any signs of fraud. I think this paper is legitimate and that its findings need to be seriously considered. ... And there’s always the chance it was a fluke, right? Can something have a p-value less than 0.001 and still be a fluke?

...

> And it was! It was a fluke! A literal, physical, fluke!

Edit: If you like puns and this writing, you'll likely enjoy some of Scott's other works, like "The Study of Anglophysics" [1] and (especially if you're at least passingly familiar with Judaism and Kabbalah) Unsong [2].

[1]: https://slatestarcodex.com/2014/04/03/the-study-of-anglophys...

[2]: https://unsongbook.com/

Sort of undermined the entire payoff there. Maybe this is a rare time to use all caps: *** SPOILER ***

Also in the SPOILER reveal, you could just quote ”And it was!” to show you enjoyed the setup and delayed gratification pun w/o tipping the whole game.

‘Twas a gem though.

Small correction: you can very much enjoy and love Unsong without understanding of Judaism and Kabbalah.

In the pilot of House, M.D., a patient is treated with Albendazole for a parasitic worm infestation. Not a fluke, but closely related :)

Albendazole is also considered for treating covid.

edit : hm, I think I remember Dr. House making the fluke joke ...

The patients got Sunday off of course

If the study was in Egypt, that would have been Friday.

based on logic and statistics, that would be true

based on the rest of study design I'm less confident

Reply on news.ycombinator.com