423 karma · joined February 9, 2022
IMHO, best modern tools for lern2code would be a tie between Lua/Love2d & p5.js.
And for 3rd place... you're going to think this is WAY OUT THERE... C--spare them the gcc esoterica of course--with sample project they can use as a scratchpad. Having types, but without all of the meta complexity of anonymous funcs, OOP, closures, dynamic typing, etc, takes a lot of the complexity (and frustration!) out of learning to program.
At the older & more social ages, the old web APIs were a wonderful stomping-ground. That avenue has since been narrowed-down & greatly so, for obvious reasons.
The Papert/LOGO/PARC stuff should be thrown into the fire. A military-industrial experiment on children; how quickly can we transform the random prole into a technician? Not that any of the researchers were "bad people" or even aware of it. It was a simple necessity of the times & typical "forest for the trees" situation. I think the numerical/visual/spatial primacy of this branch--trying to give kids a visual "feel" for mathematical representations before their brains are ready for the "real deal"--diverts focus away from many of the more "human" uses of the technology.
In something like Go/Pascal/Java/Rust/C, if I introduce an impedance mismatch during a re-factor, the compiler catches it in seconds or less, and can tell me every file & line to adjust. In classic Lisp, this becomes a runtime error, and it can be days/weeks/months/years (if ever!) before the bugs are shaken out. In typed Lisp, it's a split between the two.
Splitting the difference for the sake of a legacy codebase (like PHP->Hack), or because you've had a language thrust upon you (Javascript->Typescript) is a reasonable choice. To double-down on a predominantly type-declarations-optional language in 2022 is nuts. Lisp is good, but it's not that good.
It's the same reason why people are flocking to TypeScript.
And I like Lisp--as a not-systems language.
Can the Mossad hack me? Sure. Would Lisp stop them? Probably not. Ron Minnich gave a talk a few years back titled, "If You Trust Your Computer, You're Crazy!" How many of Ron's points do Lisp & Rust fully address? By my count, zero.
My words were "slap on the wrist." I think only having to pay $1B instead of $46.5B meets the criteria.
> something he should have argued in his letter
That's Levine's analysis of, "if the banks don’t put up their $13 billion then he doesn’t have to put up his $33.5 billion".
Going by Levine's "Exhibit E", I find it hard to believe that the banks generated this much verbiage without a single escape hatch:
https://www.sec.gov/Archives/edgar/data/0001418091/000110465...
> gets a class designated of all Twitter holders
Yes. I understand how class action works. If Twitter were a normal company, that's what one would (correctly) expect to happen. But it isn't. I think Twitter would take on a class action before airing its dirty laundry.
It's the difference between having a plain cadaver, completely drained of its life fluids, and having one where every empty tube, regardless of size, has been labeled with something like "blood, to heart" or "lymph". If you aren't intimately familiar with human anatomy, the latter is an immense time saver.
Most software that has been continuously developed over a long period of time is either C or C++, even though Lisp predates both by over a decade. There's a reason for that.
Per Levine:
But it’s messy, and you can sort of see a path to “Musk says the deal is off, so his banks walk away, so his financing isn’t available, so he doesn’t have to close the deal and can get away with just paying $1 billion.”
And the "Musk says the deal is off" part doesn't even have to enter into it. The banks could come, independently, to a last-minute change of opinion on his collateral--whatever that was for this deal.
When it comes to "sued into oblivion by their own shareholders", who exactly are they? ETFs & mutual funds, of which Twitter is only a fraction of a sliver of their portfolio? CIA/DOD "dark money", stirring up foreigners (& domestics!) against "rogue" regimes (i.e. "arab spring")? Large shareholders (like Saudi, or hedgies), who were the ones giving the board up-to-the-minute instruction on how to (mis)handle the Musk offer in the first place?
Twitter has been a propaganda asset for a long time. The large stake-holders surely understand that they have far more to lose in discovery & cross-examination than Elon's offer. I am talking Snowden-tier revelations here.
If you put two economists in a room, you get two opinions, unless one of them is Lord Keynes, in which case you get three opinions.
If it's a choice between "inequality", or reducing it by minting more parasites, I think I prefer inequality.
Ex-wife is an English professor. Department-wide "purges" were not unheard of. She was spared, though this was probably because she was a lowly adjunct.
> Twitter says an “error” introduced in Q1 2019, resulted in the overstatement of its monetizable daily active usage or users (mDAU). According to the company, this went on undiscovered for three years.
Three years, and they figured this out mere days after being strong-armed into accepting the Musk offer they wanted to evade? I have a bridge to sell to anyone who believes Twitter didn't already know.
https://nairametrics.com/2022/04/28/twitter-says-it-inflated...
edit: Not an Elon fan BTW. Just don't appreciate this Levine's axe-grinding masquerading as impartial legal analysis.
POSIX: man 7 re_format
PCRE: man pcrepattern
PCRE2: man pcre2pattern
PCRE may be a "works on my machine" thing, but POSIX should be there.
In *nix land, a decent reference is built-in: `man awk` and jump to the "Regular Expressions" section.
https://en.wikipedia.org/wiki/Operation_Mockingbird
I'm old and from the city. The newspaper has always been a puppet of hospitals, real estate brokerages, & car dealerships. Part of how these companies get away with murder is by putting the entirety of local media, print & broadcast, in their back pockets.
Unless you're paying something in the ballpark of Bloomberg terminal prices, you're the product.
I started one 10 years ago. It's not google pay, but it's still very nice, and at this point in my life, they couldn't pay me enough to put up with all of the bean-counting & back-biting in corporate software development.
> answer to your next question since i'm rate-limited:
College is crucial for W2s--if only to get through the HR filter (why I asked in the first place)--but outside of that, people don't really care as long as you get the job done. Nobody (at least nobody with any sense) is going to ask John Carmack where he went to school and what he majored in. The work speaks for itself.
Place an ad, follow the leads, take care of business, and you'll have more work than you want to deal with. "Somewhere else" businesses don't have FAANG-tier problems. Someone with a decent GPA in physics can probably handle it.
And there's your answer. Practice your leetcode, go to Seattle or SF, profit!