richardhod

Verbatim response to that tweet, from a highly qualified friend with relevant scientific expertise:

"idk that critique is pretty dismissive of the actual experimental evidence that they generated in the study… consuming an erythritol sweetened beverage raises the erythritol level in the blood to a level that clearly causes an effect on platelets, and coincides with the difference between the measured levels in the observational study… seems pretty convincing to me".

What are the limitations to heroku that people are going to Fly for? Maybe there's a standard article that would be useful to read about it?

I was confused by your first reply at first. I think that's because you are answering a different question from a number of other people. You're asking about the conditions under which and AI might fool people into thinking it was a human, whereas I think others are considering the conditions under which a human might consistently emotionally attach to an AI, even if the human doesn't really think it's real.

I did get the order wrong. Must have been the Preston bypass that was first, which I double checked

The first sections of motorway... Definitions definitions

My father was present, as a boy, at the opening of the Lancaster bypass, which was the first section of motorway constructed in in Britain, followed by the Preston bypass. He was always very careful to remind people how the M6 was the first motorway in Britain and not the M1.

Another issue here is there is a problem with laws which are broad enough that it requires selective enforcements for them to be available. Because then you can get a problematic force which then uses them selectively for political purposes. This indeed is how the Soviet state worked and indeed current Russian state works to quash dissent.

Being 'normal' != is constitutional

Indeed. Farmed ones aren't given the same diet. I suspect it's not as healthy for them either, but it does well enough for industrial purposes. Think of cattle feed v grass, etc

I like the ending: 'If you don’t like Hancock’s story about the super-intelligent advanced civilisation being wiped off the face of the planet, here’s another that might explain how Netflix gave the greenlight to Ancient Apocalypse: the platform’s senior manager of unscripted originals happens to be Hancock’s son. Honestly, what are the chances?'

While I don't disagree with the general tenor of the article and the apparent real issues with these oils,

Regarding the section quoted below: this doesn't appear to have any statistical coherence, since naive calculations like this tend to make many implicit unwarranted assumptions.

'If heavy smoking (a pack of cigarettes per day) increases risk of death by 80% and increasing vegetable oil consumption (by 12% of calories) increases risk of death by 62%, we can use some back-of-the-napkin math to infer that every 5% increase in daily calories from vegetable oil is as dangerous as smoking 7 cigarettes per day.

Another way of looking at it: each additional teaspoon of vegetable oil you consume could increase your risk of death as much as smoking 2 cigarettes. '

It's not the same at all. At first glance it might feel the same, but the structural questions are different. The fact that the Facebook does have privacy problems does not show that Atlantic salmon does actually have parasites problems in the same way that pacific salmon does.

Argument by analogy is not good argumentation.

Where you went wrong is getting it backwards. The vaccines are orders of magnitude less likely to cause cardiac problems than the virus itself, for fairly obvious medical reasons.

And that cardiac events in both cases are a minor part of the greater effects and dangers of catching the disease.

So it's about not focusing on the wrong things