Let's just prosecute each as evidence appears.
"But everybody does it" isn't a great excuse.

So, someone installs a non-google app, and records a call where it isn't allowed. Then, somehow the blame gets shifted not to the person, not to the app writer, but to google. And exposes google to X-million dollar fine? Compare that to the usual exposure of google to fines - privacy laws, self driving car running someone over, etc, etc. Doesn't make any sense.

Or playing against the same players who actually did the bidding...
and
"Each champion played [...] against a pair of opponents. These opponents were the best robot champions in the world to date – robots that have won many robot competitions but that are universally acknowledged to be nowhere near as good as expert human players."

Our universe is an anti-universe running backwards in time! That's why I get out of bed every day in the evening and slowly forget what happened to me over the day, until I finally get back in bed in the morning.

Maybe they can also invert the map? I’ll provide the summary, and the AI will fill in the rest of the doc? That would be really helpful!

Because these workers only handle packages sent to prime members, and without prime members Amazon would be in the red.

I'm running WSL, and when I need to I launch an Xserver. How will this be different?

Oh, cool reference. Thanks!
I was more thinking of random jumping around, so first square takes 1 step. Then chance to hit non visited on next step is 1999/2000, and once that happens, chance to visit another non-visited 1998/2000 and so on, giving 2000*sum_{i from 1 to 2000) (i/2000), or about 8 times the whole house.

But isn't the result of the paper that to cover a torus of size n x n would take 4 n^2 log(n)^2, which if n^2=2000 gives log(2000)^2/pi or about 18?

Strange that my random jumps gives about log(size of house), and the random walk gives log(size of house)^2 / pi.

Exercise: If a house is 2000ft large, and a roomba is 1ft large, how many times does the house need to be cleaned with a random walk till every sq ft has been cleaned?

Also standing on the ground. You and planet Earth are in free fall.
Wheee !!!

I'm not sure about BLAS - I had to get it working for the compile, but I'm not sure how much it is actually used. I'm running pomp in R, where the main model consists of thousands of lines of C++ code. It is mostly floating point, and I run exactly the same code in both cases. And, as I said, mac mini is only faster with a fan, which probably should count as cheating but is good enough for me.
And, the i9 was great for the months I was using it till the mac mini came out.

Sorry, I wasn't clear. I don't run it single-thread. I run thousands of threads. But when I debug and test I run a single thread. So for actual analysis I use the supercomputer (a tiny fraction of it...), but for quick debugging the mac mini. I haven't figured out how to split that single task, which takes 1-2 hours if run in full, among cores.

I need a job to run as fast as possible. R + C++ (covid analysis). Normally I run on a supercomputer, but for checking and debugging, i9 9900k is faster (33%) single thread. That's what I've been using till now. Now I got a mac mini m1, and it is again 25% faster. For this job, single thread. And, I have a fan directed at the mini. .
It was a pain getting everything compiled for m1, but worth it!

For some reason we are assuming aliens are rational?
Why did people go to the most remote parts of the Amazon? To teach the natives about the one true god. There must be religious fanatics among the aliens, too somewhere.

I'm a zoom fan. It works better for me than any other platform I've worked with over this pandemic. Zoom calls with many participants work better in terms of video quality than skype calls one-on-one. Screen sharing works well, control sharing works well. For me it was indispensable over the last months.