I've found there's a big difference in OCR accuracy where it comes to handwriting. For printed text, I've used tesseract, but it seems to miss a lot for handwriting. In my experience, google cloud vision is far more accurate at transcribing handwriting. Haven't tried other cloud based tools, so I couldn't tell you if it's better, but I would say that overall, the cloud based ones seem to be much better at handwriting or oddly formed text, but that for basic typeset printed text, open source apps like tesseract do well.

Wow, that's bad.

I also think that structurally, universities have academic career advancement problems. It's a very difficult system to to into and out of, in a way that is toxic for everyone but unusually so for women.

It's almost perfectly crafted to disadvantage women. Compare it to law, and imagine someone who moves through the elite path quickly. Graduate college at 22, finish law school at 25, go into heavy debt but start at around $200k a year, work brutally until age 30. Ok, not a great spot for starting a family, but you have now earned well over a mil (you probably started at 200 and are now well above 300k). You don't have much time, but you do at least have money and the ability to buy help. You also (if you're a woman) still have a few years before age related issues in having children manifest.

Now, again, it's not great, but compare this to academic. You finish college at 22, you finis your PhD at 28. That's moving quickly. Next, you do a post-doc or two. Now you're 30. And now, even if you're a winner with a tenure track position (not tenured, just the kind that can lead to it), you're still in grave danger, career-wise. It's do-or-die. You are also, at this point, earning about half of what that law graduate earned at age 26. You have no ability to buy your way out of this.

Now, to me, it's nutty that anyone voluntarily signs up for this, male or female [1]. But it's unusually harsh on women, because taking time off at this junction to have kids derails your career at the worst moment. And academic is unusually bad for on-ramps later in life. There essentially are none. Honestly, in spite of age-related discrimination in tech, I'd rather try my odds finding an on-ramp in tech at age 40 than in academia.

There is a commonly cited study that showed women applying for academic STEM positions received over twice as many offers as similarly qualified men. I actually think the research was good, but people didn't dig into the data. The comparison was for women and men who had achieved a high level position in academics. Yes, I have no trouble believing that women who make it through this gauntlet (which again, I absolutely insist on pointing out is lousy for men too) are very appealing to departments that don't want to change structurally. If a well funded department can hire a group of women who made it despite the odds, they can put a halo up and appear to be one of those virtuous departments even as they participate in the structure that makes it very unlikely women will become one of that small successful group.

[1] indeed, part of the reason universities like worker-visa programs is that it creates a pool of candidates who aren't allowed to live and work in the US with economic freedom.

I like the idea of a points system like Australia has, but I'm wary of the excessive emphasis on STEM, and I'm even less enthusiastic about the idea of empowering universities to "sell" access to US residency and work rights through overpriced MS degrees, many of which are, frankly, bullshit cash cows.

Keep in mind, in Australia, plumbers and electricians get the maximum points for professional skills. So would nurses, pilots, and even lawyers (though they must be admitted to the Australian bar).

Hell will freeze over before the US lawyers who write the immigration system rules agree that foreign lawyers and patent agents should have access to the US job market the way they think STEM workers should.

Your reply is important. There is a myth that bullies are weak cowards. Now, they may actually turn into cowards when they are faced with a fight that is important but they are likely to lose. But the idea that you can poke them in the nose and they'll go running is very false. It is absolutely true that fighting back without something to back it up can turn a bad situation into a far more dangerous one.

Not arguing that people shouldn't fight back against odds, or that there's no benefit to fighting even if you'll lose, but I insist that we be realistic about the situation a lot of people getting bullied are facing.

What do you think prevents prices from rising?

I think prices are still rising. That said, the market is split. The agencies are businesses and hence pay taxes, the direct tutors usually do not declare income and operate in a black market/cash in hand fashion. Most STEM students are financially naive. To them the prices already feel high and they seem reluctant to go truly maximaly price seeking. This in turn limits the agency market as the gap would grow to wide.

While in theory there are great alternatives in online materials etc for math remediation, the truth is most of the students using these tutor services don't lack potential but motivation. The parents send them to tutoring, the student does not want to be there and so wiuld also not fi d or consult alternatives.

What are "the excesses of the far left"? What is the "far left" even?

> I'm not sitting around scratching my head bewildered as to what he's talking about.

Strange, because despite the extra paragraph that you spent dancing up to the subject, I have no idea what you're referring to. Or rather, I do, but I know why you wouldn't (try to) spell it out here.

I am still interested in knowing what you think Carville was talking about, without prompting from me. You might want to read the interview where he said this.

But I'll give an example. Again, I want to point out that I don't use the term "woke", which I consider to be unhelpful. But here's an example of what I would view as the self-defeating behavior of the far left.

Let's take Judge Duncan's talk at Stanford. Now, I probably don't need to convince you to be concerned about the 5th circuit court or Duncan's judicial record. That much, at least, we already agree about. And anyway, there are far better people than me to explain it.

Unfortunately, those better people didn't get a chance to do so. I would say that the behavior from the Stanford Law school students who chose to disrupt the speech was an example of harmful behavior from the radical left that represents a pretty serious setback for moderate liberals. Duncan won this round, and there really was no need to hand him such a victory. Stanford looked bad, DEI admins looked bad, the student protestors looked bad. They provided a right wing judge with an an opportunity to obfuscate, avoid getting questioned and pressed on important points where he's genuinely vulnerable, and they took from the left the opportunity to raise awareness of really serious implications of what I consider to be a radical court. Worst, they gave him the opportunity to paint his opponents the worst possible light. I think they gave the right wing a real gift that day.

So yeah, I think that everyone on the moderate left knows this is becoming a problem.

I’d consider that issue distinct from wokeism. I agree that there are anti-productive leftist behaviors. My original point was not that this didn’t exist. BUt rather the same people who often complain about wokeism or CRT are those whose behavior perpetuates the issues.

I think you're using the term "inhabitants" loosely here. The US doesn't have much of a social safety net, which means that any city that attempts to build one may suddenly find that "its inhabitants" include the 330 million people living in other parts of the country who need help. Or who would just like some money and an opportunity to do meth and fentanyl and shoplift and commit property crime in a place with very little law enforcement. I personally would include that in the definition of "needing help." Any version of "help" that will help in SF will have to be coercive, which is another thing SF's political culture is having tremendous difficulty accepting.

But really, if you took Stockholm out of Sweden and plunked it down into the middle of the US, how long do you think the social safety net would last before it went bankrupt? And this is assuming you retain the european willingness to be coercive about mandatory rehab and confinement (something US based progressives are reluctant to note is a feature of the european social safety programs they otherwise praise).

Nah, you're not. I often find myself wanting more data and stats when I read the interviewing.io blog as well.

Good post, though with the graphic, I do think this is one of those cases when the y-intercept should start at zero. The graphic does make a 7% difference look like a 3x difference. Same thing for the previous interviewing.io post on whether to list certs on LinkedIn.

And when it is done there should be some sort of visual indication anyway. I was taught to squiggle the bottom of the axis line to show it's non-linear/skipping a bit. You could also rule e.g. every 50, so here starting at 40 you'd notice the uneven rules. (Much like how nobody uses a log scale and then rules only the decades.)

I enjoyed this article and find the service this company offers intriguing. I've watched a bunch of videos and see a lot of value in paying for a very professional interview with actual feedback, something that is immensely difficult to get from "real" tech interviews.

There's one thing that concerns me, it's the line: "We make money in two ways: engineers pay us for mock interviews, and employers pay us for access to the best performers."

I definitely don't want to equate this with the legal but perhaps disreputable "pay-to-play" or "double-dipping" practices of charging both job seekers and the companies trying to hire them. This looks like a good company, and it makes sense to offer a paid service for people who want to practice but aren't looking for a recruiter. Still... I see a potentially uncomfortable situation where someone who wants a job is paying $150/hr or more to take a mock interview for a recruiter who would then market the candidates who are able to pass.