The defining feature of the "save democracy by making Zuck banhammer Trump" content mill — aside from being exhausting and uninspired — is that it is at once would-be authoritarian and feckless. Little written about the subject captures that farcical paradox.
"Machine learning will fix our informational ecosystem and save democracy"
– very serious aspiring Excel overlords
"At once would-be authoritarian and feckless" is a description I would once have mainly associated with HOAs, but now applies to too many public actors to count.
"A member of San Francisco’s board of supervisors called to remove Facebook CEO Mark Zuckerberg’s name from the city’s general hospital because the social media platform doesn't do enough to combat hate speech." freebeacon.com/democrats/san-…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
I remember hearing Nate Silver interviewed by a big-name reporter about the model showing Clinton with 65% odds. The reporter says "Okay, 65%, put that in context—how often does someone with that big a lead win?" There's an awkward pause. Silver says: "Well, 65% of the time."
Imperfect analogy, but this account of democracy feels like arguing that a marriage certificate not only constitutes a marriage, but is all that constitutes a marriage.
If you go to certain Melanesian islands and find long strips of pavement and wave your hands around in just the right way, will huge quantities of food drop from the sky? Well, sort of, yes! Under the right conditions. (Namely: If it's 1942 and you are a soldier for Tojo.)
In virtually every case where engineers view their life mission as "saving humanity," it's a huge win for society if they can be successfully diverted from it.
The scientists/engineers who've done the most for the survival of the species were driven primarily by ordinary careerism and a tinkerer's obsessiveness. The ones who set out specifically to "save humanity" wind up either trying to destroy it or founding a rationalist subreddit.
This article is based on a remarkable falsehood: "at nine weeks the nascent embryo is not discernible to the naked eye."
At 9 weeks of pregnancy, a human embryo is 3/4 to 1 inch long — med books compare it to a penny or a peanut. My naked eye can see that and yours can too.
Suggesting what @suzania argues here. It's not a small detail but the entire sustained claim of the piece — everyone's been duped that you can see embryos at that stage. Whether the error was deliberate or not, the article should be retracted.
Here are images of embryos up to week 8, with scale. At week 9 they are another 50% or so longer. By week 4-5 the embryo is easily visible to the naked eye.
Planting one quick flag, for the record: Not just with hindsight but knowably at the time, the masking flip-flop was an unforced error. Public health should have recommended it 4-6 weeks earlier, and before that there was no reason to belittle the public for considering the idea.
That posture of condescension set the tone for much to follow, and it was throughline when, in March, public health otherwise did a 180, not just on masks but broadly, from downplaying the risk and worrying that precautions were irrational to pushing hard on restrictions.
My Times piece on Dr. Fauci has dropped right as we're on the finish line with sending the Fall 2022 issue of @tnajournal to the printer, but I'll have more to say soon.
For those coming to my feed from my Times piece, I'm placing this here as a show that it's not an exercise in Monday-morning quarterbacking. I've been writing about public health's dysfunctional relationship to science, and to the public, since 2015: thenewatlantis.com/publications/t…
That article has a lengthy investigation of the masking question in particular, in the context of Ebola and SARS.
From 2015: "the broader institutional factors that led to the failures of public health in 2014 remain unchanged. We must understand and fix these problems, for the next outbreak may be of a disease more contagious than Ebola, and even worse understood." thenewatlantis.com/publications/t…