jamesliudotcc

There is this free book, https://www.obeythetestinggoat.com/, which is also available in print for O'Reilly. Manning has https://www.manning.com/books/unit-testing.

The link is wrong. It should be https://www.reuters.com/technology/atlassian-lay-off-about-1...

@dang

As for Calibri, even Microsoft has moved on. They designed a successor font family for even better readability, and that happened before the State Department changed to Calibri. It also has serif variants.

https://microsoft.design/articles/a-change-of-typeface-micro...

No knock on Calibri, which I admire.

(edited to put the link on its own line.)

The worst is Times New Roman text with Calibri page numbers. A sure sign somebody never learned how to use Microsoft Word.

The Warren Commission report was set in Century Schoolbook, the Supreme Court's typeface of choice. The appendices are photo reproductions of originals produced on typewriters, so they are in something monospaced.

You can see for yourself. They certainly would not have used Times New Roman.

But I suppose interoffice memoranda are meant to be skimmed, not read, so TNR or Calibri are both fine.

It's worse than that.

> After the earthquake Google's researchers changed the algorithm, and simulated the first earthquake again.

> This time, the system generated 10 million Take Action alerts to those at most risk – and a further 67 million Be Aware alerts to those living further away from the epicentre.

So, basically, Google learned from the data, improved the algorithm, and they are blamed for not having done that before the earthquake.

The headline makes it sound like Google had the good model before, and something went wrong in the notification system so nobody got the alert.

Not out of the question after a week of Cursor just absolutely torching goodwill

Nothing for certain, but yes. The IP payout is to give the investors something.

It's a local NYC news channel. They are subject to this effect: https://saulsteinbergfoundation.org/essay/view-of-the-world-...

Except they don't even see that far, since not even neighboring New Jersey seems to be in view.

Only a draft chapter for now, but this could be very interesting.

Or, consider that the causation could have been the other way around. The EPO could have made him more susceptible to cancer.

There is a mountain of evidence that the drug cheating was systematic. You can read The Secret Race, or draw your own conclusions from the $5 million false claims act settlement.

HtDP was written to address difficulties with learning from SICP https://cs.brown.edu/~sk/Publications/Papers/Published/fffk-...

I have gotten much farther (and accordingly learned more from HtDP). It is accurate to think of it as an on ramp for SICP.

In the Flatpak episode, he rolls his eyes and states, "I'm not taking advice from a cartoon dog!" The episode is mostly about how Bluey takes after the parents in a game where she's raising Bingo through adulthood while they also go from fishes through dinosaurs to modern humans, er, dogs.

This show is so good on so many levels.

Asus makes a gaming tablet with a 13 inch screen with a 128GB of unified RAM on the new AMD Strix Point.

I agree that MacBook is not an option. I have a work MacBook and it has better silicon, ok, but I hate being on a Mac. Just not for me.

I'd wait until more manufacturers announce what they will do with that AMD SOC.

I feel what you're describing very viscerally. I have tried so many times as an adult to finally get linear algebra. Worked my way through Strang to eigenvalues and eigenvectors repeatedly. Still feel like I am failing to see something.

---

I hear Apostol is hard. Like Spivak-level hard.

Maybe your problem is Stewart? I used that textbook and was successful, but it's not for everyone. For example, beginning calculus with limits is another bit of misguided conventional wisdom. I still don't get limits, really. Serge Lang's calculus book takes the approach to just roll with an intuitive notion of limits, saving rigor for analysis. Which seems better.

Gilbert Strang has a textbook, also more intuitive and applied. Free PDF provided by MIT. Sylvanus Thompson's book is recommended here, again, intuitive, applied.

Other comments here, 3 hours isn't enough, use Math Academy, nobody gets it on the first approach, all seem relevant. One of the textbooks recommended here says in the preface that it's for a second course in Linear Algebra. Analysis is just calculus the second (or third) go round, and it's said to be the hardest class in a math major.

I am in your boat, but about linear algebra instead of calculus. This is what I try to get myself over the hump.

No, of course not, but credentialism.

> According to a new Gallup poll, Americans are losing the thread with higher education. Confidence in college has taken a nosedive, with one out of three poll responders claiming they have “little or no confidence” in higher education. This contrasts sharply with a 2015 poll, when 57% of those surveyed claimed to to be fairly or “very” confident in the old hallowed halls.

This paragraph shows why. 1/3 having little or no confidence now compared to 57% fairly or "very" confident. Without knowing the magnitude of the middle choice, is this even comparable?

It doesn't help that this is two paragraphs down:

> 3) an unpromising job market with diminishing returns for un-STEM professions and 3) the many tortured varietals of “free speech” discourse.

Ok, it's probably a typo, but still, not confidence inspiring.

Also, why is "very" in quotes, but "fairly" not?

It's too bad they don't hire proofreaders anymore. That used to be a job you could get out of college.

And there is plenty of bad philosophy out there. At a first approximation, avoid anything to do with the continental side of the analytical/continental divide. I say this as someone with a BA in philosophy from an analytic institution (Chicago) and an MA from continental (New School).

Another reason engineers should study philosophy is that philosophy departments take logic seriously. Talking with CS majors, their logic background is a few weeks in a discrete math class. I had two quarters, covering Godel's proof of the soundness and consistency of first-order logic (so 20 weeks total). If I had the appetite, I could have taken another two, for the incompleteness theorem, and modal logic.

If you are looking for a place to start, I recommend Grayling's Philosophy 1. It's an edited collection of introductory essays written to be understandable by beginners, but deep enough for philosophers in other subdisciplines. The best way to read it is to skim through the essays until you find one that is interesting to you (they don't much depend on each other) and then read that one carefully, and if it still holds your interest, chase down the references. Think of it as building a heuristic for A* search. Philosophy only gets good once you dig deep into a subject, and so you really need to find something in it that speaks to you.

Specifically for engineers, Philosophy 2 has an essay in it titled Philosophy of Psychology, which covers logic gates and Turing. Philosophy 1 is not a hard requirement for Philosophy 2, though I recommend at least skimming the Philosophy of Mind essay in 1.

I've used it on Mac on my work machine for a year-plus now. Now the wait is finally over for personal. Thanks for all your hard work on this!

Fluent Python is superb.

Definitely not the Intros for a desk reference.

You have to know yourself a little to decide between free Youtube university and bootcamp. I think of myself as a self-starter, but at some point, I had plateaued in my learning and needed some fresh momentum.

Could I have kept learning on my own? Sure. But the butt-in-seat for 9 hours a day at a bootcamp accelerated my learning quite a bit. Was it worth the $14,000? For me, yes. I got to being a hireable junior developer many months faster, so just in terms of runway, it was worth it. You'll have to make your own assessment about how much shorter of a runway you'll have, and how much quicker you'll get to takeoff with a bootcamp for yourself.

Still, it's tough to recommend a bootcamp in 2024. I did some writing about it. https://james07.bearblog.dev/how-to-keep-coding-bootcamps-re.... It isn't so much, will gen-AI take all the junior jobs (though, to some extent, yes, it has)? It is more, only gen-AI-wielding juniors will be worth hiring. At the very least, the bootcamp you're considering should have their story straight about how they will help you learn with gen-AI. They've had a year to figure this out, or not. If they seem lost about it, stay far away.

On par with US wages.

My company uses [Remotely](https://www.remotely.works/). They source talent across Latin America (mostly Brazil, but other countries as well). Expect to pay well for the best talent (obviously). My coworkers sourced through Remotely are all extremely good.

The time zone difference is not a major problem for us, and I am in PST (the US part of my company is mostly EST & CST). There is a large actual time zone difference with Brazil, but all of our contractors tend to be night owls (we are clear about where we are, if they weren't at least slightly night owls, they wouldn't choose to work with us). If you keep core working hours in US EST, you hopefully have morning larks among your Washington state employees as well to make it easier.

Things also depend on the specific foreign country, and what other business your company wishes to transact in that country. My company has (I believe) direct employees in many jurisdictions, including in Europe. But for Latin America, I believe (almost) everything is through Remotely.

Yes. Five years ago, I tried to use an eink tablet with Android to consume the O'Reilly Android app. It tried to animate the page transitions, so it would refresh multiple times for a page flip, making it unusable. It would take most of a second painting one intermediate screen after another. They've since moved away from the page-flip skeuomorphism, but scrolling might still require multiple refreshes. I'm not sure. Using the browser was slightly better, since the browser included a page-down function.

The irony is that the page flip skeuomorphism works for eink because when you flip a page, you expect to go to the origin of the next page once the flip completes. E-readers do this, and it works as long as you don't try to animate it because of how slow eink is. The pgdown/pgup mechanism leaves the last few lines on the screen, which works well when you can animate the scroll action, since your eyes can follow it. But again, eink is too slow for this to work well.

I complained to O'Reilly about this, but they gave me a boilerplate response that they take user feedback seriously. I don't know if it's been fixed.

What I do now is use a Python package that converts O'Reilly learning books to epub. You need a login to use it (of course), and it would certainly be beyond acceptable fair use if you distribute the generated epubs beyond your own devices.

I've known two people who have suffered from actual carpal tunnel syndrome, as opposed to milder forms of RSI, and both were in food service--a chef and a barista. Picture flipping food on saute pans five hours a night or pulling portafilters out of espresso machines five hours a day. You get the idea. The chef was also beginning to suffer from elbow tunnel syndrome.

Folks typically don't get actual carpal tunnel from keyboards, especially newer (here, newer than the 90s) keyboards with much lighter key presses. Touchpads are much more resistant to even milder forms of RSI than mouses.

That and better keyboards and mouses are an easy and relatively cheap way to mitigate, so those who suffer from RSI pain at work can ask for ergonomic devices. It is hard to argue that they aren't "reasonable" accommodations under the ADA, so firms accept that they are required to purchase them.

Yup.

But what is a reasonable percentile for someone to "comfortably" afford the monthly payment for a median new car? Maybe 80 is on the high side. But it certainly isn't 50%.

10% of income as a heuristic for a car payment isn't too terrible.

It should have been clear enough from reading the abstract of Khan's law review note https://www.yalelawjournal.org/pdf/e.710.Khan.805_zuvfyyeh.p.... Quote: "This Note argues that the current framework in antitrust policy--specifically pegging competition to “consumer welfare,” defined as short-term price effects—-is unequipped to capture the architecture of market power in the modern economy." In other words, the current law is wrong, and it should be this other way instead.

Courts, especially U.S. District Courts, apply the law as it is, not the law as the FTC chair wishes it to be. If I was a hedge fund manager with a law degree, I would draw the same conclusion, and make similar bets.

It feels like a very truncated version of the translator's preface. Maybe someone will post a review.

I am partial to Pevear and Volonkhonsky. I find the English fine, despite everyone complaining it is too close to the Russian. I don't read Russian, so the closer the translator can get me, the better.

What happens when they lose this one too? Just because the commissioner has a law review Note setting forth a legal theory does not mean the law has changed.