morei

My work around for this is to take a walk without headphones. For me, it means that the only thing I can do is think (can't read, can't watch video, etc). And by the time I've finished a 30 mins walk, I've usually got a bunch of things that I'm enthused about working on.

For those that don't know: Concentrated sugar solutions are indeed hydroscopic.

So it will indeed absorb water out of the atmosphere. It's still doubtful that it will do it fast enough to create any significant pressure differential though :)

No, because glass isn't a crystal. This is part of why it was such a hard problem: Because the existing sub-molecule imaging techniques couldn't image it.

What's the difference between N and N+ ? As far as I can tell, N+ just means a higher doping concentration? But it seems very fuzzy as to what 'higher' means?

This is much harder than you're assuming.

With the increasing prevalence of TLS traffic (https et al), any cache almost certainly needs to be an explicitly configured proxy which is a user configuration issue with all it's associated support and compatibility problems.

Add to that the power consumption on a satellite is often a limiting factor, the security issues, the added latency, the ever decreasing fraction of traffic that is cache-able, etc etc ... it's very much not clear that this is a good idea.

Not quite: Truetime has guarantees about ordering, rather than accuracy.

If you ask for the time and get A, and then ask for the time again and get B, then Truetime guarantees that A is less than B.

Obviously, in a distributed setting this is much easier to do if you have accurately sync'ed clocks, but that accuracy goes to reducing the uncertainty in the time (and hence making truetime faster) rather than providing accuracy.

You're right. My bad. It's much less common than I'd thought. (Intel had it on a number of chips that included the Iron Pro Graphics across Haswell, Broadwell, Skylake etc)

L3 is almost never SRAM, it's usually eDRAM and clocked significantly lower than L1 or L2.

(SRAM is prohibitively expensive to do at scale due to die area required).

Edit: Nope, I'm wrong. It's pretty much only Power that has this.

There's a couple of key things here:

1. The sheer volume of fraud attempts. Economics often dictate that it needs to be cheap and fast to reject a fraud attempt.

2. Information leakage. It's normal to see people complain that '<insert service of choice> banned them and refused to say way'. There's a very good reason for that: They're trying to slow the rate at which fraudsters learn to exploit them. So they deliberately don't detail exactly what the issue was. Yes, it's super frustrating if you get innocently caught up it, but it's not arbitrary.

TL;DR: Like everything else in life, there are real and genuine trade-offs here.

It's never that simple. You're implicitly assuming that a fraudster wants the account long term, which is rarely true.

And identity is a VERY complex area, and nothing like as simple as "plenty of ways to verify identities". Particularly noting that fraud is often carried out by leveraging many partial opportunities: I use the (false/stolen) identity from over there to carry out of the fraud over here.

It's rare, but possible.

Known of v high level example that is just incredibly disciplined about how they spend their time (e.g. "I did this, but was slowed significantly by poor code over that. Estimate about 30% time loss. I'm going to spend N weeks more in this area, so it's worth me spending 3.5 days fixing that code, but no more. I think I can fix it in about 2 days, so I will fix it. If I can't see the end of the fix by the time I'm 4 hours in, then I'm probably wrong about the expected time, so will abandon the fix").

The result is ridiculous productivity. They don't work long hours (strictly 9-5), but just very, very focused on making certain that there will be meaningful results from those hours.

So they exist, but absolutely it's not going to be common!

Google has an office in Frankfurt, and SWEs there get paid far more than 30k - 100k/yr. The comp delta between google SF SWE and google Frankfurt SWE is less than 30% (at same level obv).

I totally agree that the non-FAANG companies have terrible dev pay. But most of the FAANG companies have offices well beyond just SF/NYC, so the opportunities are definitely available.

This is a situation where automation has the advantage. With detailed position information and detailed maps, the fact that the lines on the street can't be seen is irrelevant. The car didn't need them anyway.

(noting that Waymo requires full detail maps to be able to drive an area, including all signage).

I had a quick read. The paper states in the conclusion "The more we increase the number of iterations k, the more proportion decreases, but the number of integers starting the different trajectories keeps increasing and tends towards infinity as k tends towards infinity."

This statement is not proven, and is not supported by the previous theorems. Thus the following statement "We therefore conclude that the 3x+ 1 conjecture is false" remains an unproven assertion.

Essentially the paper argues that the density F(k) of trajectories at least as long as 'k' gets ever smaller as k increases, without ever reaching zero, and thus there exists at least one value that has an infinity long trajectory.

But that's as false assertion. Just because F(k) is non-zero for every finite value of k does NOT imply that the lim as k tends to infinity of F(k) is non-zero, and consequently does not prove the existence of an infinity long trajectory.

TL;DR: paper applies intuition on finite sequences to infinite sequences and gets it wrong.

The F50 sailing class used in SailGP can remain foiling during jibes in winds as light as 8 knots. That is, the boat always sees an apparent headwind during the entire downwind run.