riwsky

You can’t imagine placebo.

Two of the best books for software engineers are actually not specific to software engineering:

  * “The Design of Everyday Things”. Have you ever pushed on a pull door, or vice-versa, because it was unclear how it worked? Designers call those doors “Norman Doors”, now, after the author of this book Donald Norman so effectively dissected and criticized them (among plenty other objects). Too many APIs suffer similar problems—and if you don’t want yours to, this book is a great place to start.
  * “On Writing Well”, by William Zinsser. Good code and good prose have a lot in common: they know what they’re trying to do, are stripped to their clearest components, and have internal structure that supports understanding. Add to that the fact you’ll be writing lots of prose anyway (design docs, postmortems, etc), and the value of writing well should be clear.

It’s a mood.

It is, however, about turtles.

It’s a lisp, so arguably you can abstract away “actually writing the code” too, behind macros

Market forces: the ultimate garbage collector

Sure, comprehensibility is the terminal goal—but size reduction is relevant because it’s an easily measured, decently predictive proxy for that. No, code-golfing 3 lines down to one doesn’t help, but the bigger size differences are always of the “composition over XYZ” variety that change the entire complexity class of how much code you need to write (like React letting people write O(states) rendering code instead of O(transitions)).

Artificial Intelligence changes; human stupidity remains the same

We did that for tobacco, though? It was a huge public health win?

“Someone autoformatted the whole thing under new settings at the same time as introducing a new feature” is hardly a monorepo problem. That could be a pain in the ass to review even in a single file. But the flip-side, of someone cleanly wanting to a do a mass autoformat or autorefactor, is much easier in a monorepo than in split repos.

The support for remote or in-docker development has been an issue, personally. Terminal editors and VSCode are much more “it just works” in this regard, especially when joining codebases that used docker to isolate all the dependencies to actually run tests/services for local dev.

Mermaid even has experimental support for plantUML-format C4: https://mermaid.js.org/syntax/c4.html

Share button does that, at least

FromSoft would like a word.

Remember when they tried to king George Washington?

Yes. In particular, it means you can switch from depending on a concrete type (say, a whole S3Client) vs depending on just a much smaller interface (eg if you only want GetObject) without having to touch anyone else’s code. The compiler will still check that the types being assigned to the interface meet those constraints.

“jj commit” = “jj describe; jj new”. Basically, jj doesn’t force you to wait until you’re “done” with some work to write the VCS message for it.

“There should be one, and preferably only one, obvious way to do it—xor it’s packaging”

Engineering is tradeoffs; the whole concept of quality only makes sense in the context of a certain budget. If an engineer takes two weeks hand-crafting some one-off charts in d3 and react that could’ve been done uglier in some BI dashboard in a day, it might look good on a résumé—but it’s shitty engineering.

I’m aware. The context of dataflow’s original inquiry is of some mechanism to prevent a large corporation forking a codebase and running away with users; Google didn’t need to close the Chromium source to pull that off.

Darwinism, one might say.

It didn’t: that’s the joke.

I don’t think you’re pro-slavery, but I do think you picked a metaphor not for the light it sheds on the issue, but the heat—and then preemptively dismissed a strawman objection to it instead of, say, improving your own communication.

What I’m hearing is that years of improv classes, telling me to avoid asking questions but instead yes-and to establish facts, has made me autistic?

Cooking turkey in a vacuum… sous vide is your idea of easy? Props, dude!

This again?

Best practices

Statistics has a semantics all its own

Hats off to you!

That’s because you aren’t supposed to wrap your head around a hat, you’re supposed to wrap the hat around your head.