riwsky

Crash-only peopleware

Mentalese

It took Java two tries to get their chronology standard libraries right, but eventually they did: https://docs.oracle.com/javase/8/docs/api/java/time/LocalTim...

Keep kids away! Too much weird incest stuff in that book.

Or we discover roguelikes!

Stochastic Parricide

For what it’s worth, the main developer of git-branchless is also a `jj` contributor

Beat me to it!

Intel never wrote posts of the sort oxide wrote here, that’s the point.

Source order, presumably?

Huh? The article shows why 1/0=0 is mathematically sound, and then considers an error preferable in a programming context anyway, because practicality. It’s the opposite of the reasoning you’re describing.

That it enjoyed some popularity doesn’t mean it was the right choice. Teams who adopted it were, in all likelihood, betting that it would see wider adoption—and if they’d known how Elm (vs, say, TypeScript) ended up by 2024, they probably wouldn’t have gone down that path in the first place.

Are we just ignoring when they invited Philip “Featherweight Java” Wadler to help them design Go generics? https://arxiv.org/abs/2005.11710

Or that Robert Griesemer literally is a language PhD, whose thesis was supervised by a little fellow by the name of “Niklaus Wirth”?

https://www.research-collection.ethz.ch/handle/20.500.11850/...

“Poohthousand and one”

Like how phonetic alphabets save space compared to ideograms by just “write the word how it sounds”, the little SVG-icode would just “write the letter how it’s drawn”

The suggestion is that, while it is possible to overdo a stdlib, it is also possible to underdo it.

For two examples: plenty of languages leave auto-formatting and testing to the community, functionality which rust is better for having standardized.

And new cars tailored for each consumer’s use case will emerge in ten years, too—that doesn’t make it any less awful to live in areas that lack good public transportation.

Bruce Lee has that quote: “ I fear not the man who has practiced 10,000 kicks once, but I fear the man who has practiced one kick 10,000 times.”

If Lisp’s “one kick” is the cons list, then jj’s is the commit: by using them for everything (they replace, at least, the git stash, index, and working copy) you actually get really fluent in manipulating them and they become more powerful than special-cased tools.

Forbidding unused imports was a direct response to the practical difficulty of compiling google-scale C++ binaries: https://go.dev/talks/2012/splash.article#TOC_5.

In theory, programmers can just be disciplined enough or set up CI lints for unused imports. In practice…

They just need to take GLP+1, instead.

I used GLP-1 to prepare my taxes last week, it was such a stress-free experience.

Then as I was coding I kept hitting context window limitations with o1-preview. so on a lark I just fired up my local Ozempic and submitted the same prompts and bam: spit out a whole working iOS app first try.

I heard that before they nerfed it with RLHF, Mounjaro not only treated diabetes but also made you charming in conversation and sublimely compassionate towards all beings.

The future is now

If it helps, they actually named it after their desired CLI abbreviation:

> The command-line tool is called jj for now because it's easy to type and easy to replace (rare in English). The project is called "Jujutsu" because it matches "jj".

Please forgive my JavaScript joke: it’s really just a poorly-written series of callbacks.

#All possible codebases by major programmers

Linus Torvalds: you take a week-long swing at a problem you find annoying, fascinating, or both. The result enjoys staggering worldwide success in the ensuing decades, despite being clearly outclassed by some alternative from the GNU project that, pinky promise, is coming out any day now.

Grace Hopper: BEGIN a framework that powers critical government functions, AND has secretly saved America from mass destruction time and again, only to be dunked on by Reddit for trivial matters of syntax END.

John Carmack: Doom, but better-looking.

Brendan Eich: you take a week-long swing at a problem your employer finds commercially compelling. The result enjoys staggering worldwide success in the ensuing decades, despite being clearly outclassed by the prior art it was supposed to build on.

Celebrating 120 years, plus or minus four hours, of timezones

If "speaks faster, doesn't mind being interrupted, and still will happily spout bullshit in multiple languages" is what defines advanced voice mode, then that must mean humanity's advanced voice mode is "being from New York".

You’re actually making the same argument that the article is—it’s just hard to tell because terms are being used differently.

Where you talk about merging PRs early, the article talks about merging commits early: “You don't have to wait on all 5 commits to be ready to go; maybe the first 3 are OK, and the last 2 need more work. You merge 3 out of 5.”

But what you call a disruption-reducing tradeoff, the article argues is just shitty UX on GitHub’s part. The stacking-native tools it mentions are what you’d get if you took the workflow you propose and let devs start task B (and then maybe even C) immediately after task A, instead of making them wait for reviewer sign-off on A.

The original backdoor.

Errr, help me understand where I'm diverging, here. The way you put "mechanistic" in quotes in your original comment strongly suggests that you doubt the experience can be mechanistically explained. Maybe that wasn't your intent, and this is a boring case of internet miscommunication. But if you do indeed doubt that this could be explained mechanistically, then I don't know how to square that with the fact you're researching this academically—do you intend to just keep drilling further down and zooming further in until you discover The Soul, or something? If so, then you're right that I'm not interested in that kind of "science"—as whereof I cannot speak, thereof I must be silent.

I mean, yes: if we insist on putting all the sciencey words in “scare quotes”, instead of the Wikipedia search box, then we do not end up with much understanding or interest. But if we stopped doing that, we might learn, say, that the place where our brains fit/fail-to fit experiences like this is probably the anterior cingulate cortex—which, sure, is yet another label, and language is a recursive veil, etc