I was going to say "borrowed time", but looks like the majority of the answer may be "Universities"? The first stats I found showed black people making up 4.1% of Google tech employees vs 7% of Computer field employment. That's barely more than the ratio of underrepresentation that white people have among Google tech employees. (which might also be a factor? "you picked too many whites" can become a lawsuit even without allegations of racial animus, but I'd expect "you picked too many Asians" to raise eyebrows in any crowd less racist than a Harvard admissions committee)
Edit: I initially misread that 7% as being "CS degrees", rather than employment in the field as a whole. It sounds like the gap among new graduates has narrowed, if "In computer science fields, Black students earned 9% of bachelor’s degrees, 13% of master’s degrees and 7% of all research doctorates over the 2017-2018 school year." Comparing Google's cumulative hiring stats over decades to new graduate stats a few years old is a bit apples-to-oranges, but if I were one of Google's legal compliance people I'd now definitely be looking for some apples-to-apples and oranges-to-oranges numbers before I felt safe.
roystgnr
6d ago
Oh, there's no general consensus; to non-nerds the original perpetual internet flame war may have been Kirk-vs-Picard, but to nerds it was vi-vs-emacs.
I'm a happy vim user, but I would recommend it if and only if you expect to spend a significant portion of your life editing text; it's great to use but time consuming to learn.
I have a lot of coworkers, including the ones who wouldn't touch Windows with a 10 foot pole, who are big fans of Visual Studio for C/C++ development, but I don't know how well it works for Python.
roystgnr
6d ago
I'm not much of a Python guy in particular (though I think it's fantastic that the same language is useful for both teaching kids and writing cutting-edge software; when I was a kid we had various forms of BASIC, which were used for and useful for neither).
But my most useful tips are language-agnostic:
Write and comment and document (three separate things!) all your code so thoroughly that even a complete stranger doesn't need to ask you questions to understand it all. This doesn't sound so important for a "sole developer" role, but at some point you'll have to extend some of your own code that you haven't looked at in years and you'll be the complete stranger who can't ask your past self questions.
Cover your code with tests. Set something up to automatically run tests before allowing any new merge (I'm assuming you're using a version control system; if not then let's call that tip #0). You will write bugs, but it won't matter so much as long as you're the first person who's hit by them, because then you have a chance to make sure you're the only person who's hit by them.
roystgnr
2d ago
If the rough estimates Wikipedia collected are reasonable then you're a surprisingly good guesser.
roystgnr
5d ago
There is simply no way that most people would prefer years of incarceration to caning or similar physical punishments.
I think more important than whether we've properly calibrated the amount of punishment is whether we've optimized the effects of punishment.
Why do we punish people? For incapacitation, for denunciation, for retribution, for a deterrent, for reparation, for rehabilitation, and for expiation. The further you go down that list in that order, the worse prison looks.
Incapacitation is probably what prison is best at, better than any punishment short of the death penalty (and a lot more flexible than that...). Every year you keep an offender away from potential victims is a year likely to have fewer actual victims. I suspect no amount of caning or stockades or whatever else we might bring back would be enough to completely eliminate the need for prison as a "backup" for repeat offenders.
You'd think calibrating any sort of punishment would make it reasonably effective for denunciation and retribution, right? We have The System tell the offender that they did a horrible thing, it prescribes a certain level of suffering for the offender, and this gives us a shared ethical code and some feeling it's being enforced (at least if the police and the prosecutors are doing their jobs, but that's a requirement with any form of punishment). Thinking about incarceration from this perspective, already it's possible to see cracks in the system. Is it even possible to calibrate the suffering we prescribe to different offenders? If you're accosted by some thug and have to fight back before the police arrive, do you think his prison sentence would deliver as much suffering as yours would, if a jury doesn't think your self-defense was justified or proportional and convicts you in addition to or instead of him? If you're very upper class you may have the social/financial/cultural capital to recover (respect, Martha Stewart!), or if you're very lower class you may be okay with a little free room and board, but if you're middle class your career may never recover. Other forms of punishment have similar flaws here, though, so it's hard to fault incarceration specifically.
As a deterrent, incarceration is probably specifically much less effective than the same level of suffering would be if delivered as corporal punishment. The sort of high-time-preference offender who thinks crime is a good idea in the first place is not going to be nearly as deterred by suffering which is scheduled years into the future, and because the suffering from prison is so more gradual than the suffering from corporal punishment there's no way to avoid letting it stretch long into the future for serious crimes.
For reparation (aside from "the victim feels better to see the offender suffer"; I'm counting that with retribution), incarceration is basically useless (it doesn't transfer any value to the victims) or worse (it conflicts with possibilities like wage garnishment that could transfer some value to victims).
For rehabilitation, in theory prison could be helpful, but in practice it seems to be worse than useless. Criminals are not being isolated from the bad influences that led them to crime, they're being put into a community full of them. Depending on what connections a prisoner makes, they may end up more disposed to a life of crime when they leave than they were when they came in.
And for expiation, incarceration is probably grossly counterproductive! In theory The System has told everyone that "they've paid their debt to society" upon release; in practice any significant sentence length makes it difficult to maintain relationships with non-fellow-criminals (and nearly impossible to continue providing friends/family/dependents any support) and difficult to find a (legal) job when the punishment is over. Arguably these are the most punishing aspects of prison, but they're also precisely the aspects of an offender's life we want to encourage, not punish!
Once there were enough non-nerds there, it wasn't the early days of the Internet anymore.
More options
Context Copy link