pm_me_your_quan

Yeah, I'm surprised they released the demo about multi-temporal crop prediction. Their accuracy is, frankly, pretty terrible. It's basically what I managed the first time I tried to run a classifier against the CDL dataset across years.

I'd be really curious if you're willing to expand more on how it has helped with those workflows. Do you copy/paste chunks in and ask it to explain them? Have it try to refactor them and then clean up?

eh, I think it's arguable here that making a good experience here would really help cement people swapping over. Then again unclear if twitter is going to remove the super low limit, so people may abandon it anyway.

That's clearly not enough. Tech executives are doing too many drugs and it's leading to this kind of behavior, it's a cultural problem. They need fewer broken families and more time spent in church. And these rotten ones they need to be under constant watch by strict, controlling wardens.

e.g., the system deployed in Germany: https://www.youtube.com/watch?v=_3P_S7pL7Yg&t=193s

This exact problem happened with an optic in my lab in graduate school. For two years the senior grad student and postdoc blamed each other over the entire apparatus becoming misaligned every couple of days. (It was a really toxic environment.) Eventually, they both left, I was the only one there, and it still became misaligned. In one day I tracked it down to a prism from Thorlabs whose glue had gone bad positioned at the very beginning of the laser line- it was sliding in its mount.

I wish I had pushed more strongly about it. We spent probably a full person-day of work every week on that.

Ceres Imaging (Aerial and satellite imagery analytics for farming), Convoy (Trucking load auctions), etc. There are plenty of companies doing very real work that need this kind of heavy numeric lifting.

Except that's not what the study says. Quoting a comment below, "The linked study (and the Merten's study it's built ontop of) classifies defaults as "structural" interventions. In the linked meta-analysis, after stripping out estimates of publication basis, structural interventions have the most "robust" evidence left (95% CI of 0.00-0.43)"

I took it when I was an undergrad (that was ten years ago, but I hope the spirit remains the same) as my first CS class. It was very tough given my preparation, but the thing I appreciated was that it gave a very general introduction to many high-level concepts in fairly good detail. I felt I had a solid starting point in many things- programming paradigms, understanding data abstractions, levels of abstraction in program design, etc that I don't always see replicated in other introductory material.

This is a terrible idea. Paper ballots work, are secure, and the processes are well understood. The value an attacker could garner via control over elections is enormous, so there's a big incentive to do so.

>Counterpoint: why not just just take a fulltime job and quit / move on if it's not a good fit?

The company doesn't want to be on the hook for unemployment or wrongful termination I think.

> I agree in principle but in practicality, there are some difficulties such as needing healthcare coverage for my family. (Yes, the US system of employer-subsidized healthcare is utterly broken, but that's what I'm dealing with)

I actually did this for my last job, because I'm terrible at interviewing (I both hate practicing leetcode, and have fairly bad anxiety). I had the same concern as you, and the way I got around it was to ask for them to pay me on contract an amount equivalent to what my salary would be AND benefits. I guess if you're employed then that doesn't work so well, but as an unemployed (at the time) person it worked out great.

Yea, the core questions are around superposition states, at least from a rough reading. Centrally, if the _same_ clock is "at two altitudes" (by superposition), which time dilation will it experience?

It is. We haven't had the necessary experimental equipment to be able to measure it until recently. You need an extremely precise clock that you can put into a quantum superposition- that's difficult to make.

But dense urban cores tend to subsidize all of the infrastructure that makes suburban life possible. If the money weren't funneled in the wrong directions, there's no way it would make sense to be out of the urban core.

academics are largely the only people who will be able to understand the work, but sure.

The fact that it's not out in the open is somewhat complicated. You're perhaps right that it would lead to better outcomes, but it's also important that researchers feel free to speak openly.

> If it was about accountability you would have to prove the flight was really needed in the first place,

Friend, at a certain point the overhead to administrate these kinds of checks is more costly than just letting people buy tickets to go to conferences. And at this point it isn't corruption in the university, it's in the form of handouts to large corporations.