Did I? The person that programmed Slashdot to silently filter mu characters made the mistake, IMO.

I actually typed the mu characters in my comment, but slashdot ate it. Thank you for pointing the error. Next time I will use mg/mL.

PFAS are special here because they are quite hard to get rid of. The C-F bond is the most resistant that chemistry can produce, it is very hard to destroy. They are also very hard to excrete from the body: half-life of PFOA (the oldest PFAS) in your blood is several years. As a consequence, the accumulate in the environment and in living organisms, including human beings.

Firefighters in this study are mildly contaminated (lucky them!). 95th percentile for PFOA in this study is 3.30 g/L. In France, the Esteban study reported 95th percentile for PFOA at 5.257 g/L. This is the general population, not firefighters.

As always, check crustal abundance. Ok for titane, but Sc and Sn are as rare as Colbalt. Sourcing cobalt is not that easy right now.

It is not obvious that learning academic stuff is the goal of a business curriculum.

It would also be nice if every company would not run the exact same software. Monocultures makes crops vulnerable to diseases, this is the same story for software.

Too bad, now this makes me feel sympathetic with this toxic UK project.

I guess that explains the huge interpersonal variation on PFAS excretion in feces. Everyone's liver excretes a small amount of PFAS in bile, some people have gut bacteria that absorb it, other people reabsorb the PFAS from the intestine.

Not sure this will be a big change for PFAS decontamination, as the PFAS must first be excreted by liver, and it seems it is not a very efficient process.

You face a statistical problem. Even if we find convergent studies concluding your habits give you a X% increase of dying before age N, that will not tell you when you will die. You could be lucky and get in the (100-X)% side. Or you could be unluncky and get a cancer at age N-20. Or you could even die from an unrelated reason, like being hit by a truck.

One still has to trust the code to actually implement what is advertised. And to be bug-free, too.

But we do not know if current GPU mitigations would protect against that new attack. Probably they would not.

Someone cited a Phoronix article. It explains that the GPU security mitigation were against Spectre attacks. This attack class is now addressed in the kernel and there is no need to take care of it in the GPU anymore.

But rotation is measured in rad/s, c is m/s, so how does it fits?