Skillmon

@John sorry to hear

@John the fact that new code is constantly written in expl3 should tell you that it's the more convenient coding style...

@John this is an incredibly error prone way to communicate. I have no idea what the rotating TeX-thingy is supposed to mean in this context.

@DavidCarlisle says the one who frequently mistakes operating systems for text editors.

(except that I really enjoyed Tangled more, but that's a matter of taste, any argumentation was not to be taken serious)

@John I thought this was obvious but in case it wasn't: I was not serious with any of the Frozen vs Tangled discussion.

@samcarter yeah, a human could say it as well while sweating like crazy. But you wouldn't say it when you're trying to say that you speak a language fluently.

@DavidCarlisle there's a correlation between things I said and things that are true.

@DavidCarlisle of course you can, it's futile, but you can

@DavidCarlisle or that is the case :P

@DavidCarlisle you wouldn't say it like this in German, while you're speaking a language fluently (eine Sprache fließend sprechen), you're not "fließend". It's only used in this context as an adverb not an adjective.

@DavidCarlisle "Who needs AI, when we're flowing by nature"

@cis someone did very likely, yes. But I doubt that anyone in this chat did

Yay, I managed to listen to a few minutes of Frank's talk (jumped in during the screen reader example), but have to leave now :( I'd have loved to meet you guys during TUG. Maybe, hopefully, perhaps next year.

@Witiko two years ago we created an extra room for TUG on SX

@cfr the idea is that internal commands don't have optional arguments.

@cfr not sure, everyone did when I started. No one ever questioned the practice

@mickep yes, and we found 3 of them, but some evil mamoth and its friends destroyed them

@mickep I can blame the melons

@mickep but whom can I put the blame on then? Do you volunteer?

@mickep wait, if I'm using ConTeXt I can't blame her?

@cfr because that's what TeX people do, isn't it?

@cfr we could blame @UlrikeFischer

@cfr :D

@cfr then I misunderstood that comment:

@cfr \pgfpath... commands don't use pgfmath. They only accept \pgfpoint-like arguments, and \pgfpoint takes TeX dimensions.

@cfr so we're comparing \pgfmoveto with draw_path_moveto:n and friends

@cfr the PGF basic layer that is. And why do you think so? PGF's number parsing is actually quite fast (l3fp might be amazing, but fast is none of the adjectives I'd use to describe it -- at least not without the addition of "for what it's doing")

@cfr l3draw is slower than PGF, and boxing is cheating :P

@user2379123 it's reopened now.

@JosephWright yes, but not until tomorrow.

@cfr see here: tex.stackexchange.com/q/745988/117050

There's a question about that on the main site, let me dig

@cfr ^^^

@DavidCarlisle not sure, but I know that l3draw doesn't like being boxed. Use xsavebox instead.

@DavidCarlisle I think he's also convinced that it's not fair.

@DavidCarlisle not fair!

@John still your fault for using the repackaged TL instead of upstream.

@John that looks very graphical, so your fault for using a GUI when there's a perfectly fine TUI.

@HenriMenke yay, a Henri is in chat! Hi, how are you?

@MaxChernoff you don't have a serial port connected to your stove?

@MaxChernoff and I'd never do this in C :P

@cfr my only reason would be so that every L3 piece of code looks the same, making any future maintainers' life easier. But if it's just about indentation/whitespace and not about serious things then an automated tool could do those changes...

@cfr it doesn't and you're right :)

@cfr as I said in the earlier comment: You're close enough to not nag (and I really wouldn't have if @John hadn't brought up the space-matter)

@cfr I indent semi-manually, yes.

@cfr but since we started this: You're allowed to not put spaces in {#1} (if the braced group is only around a single parameter), but using { #1 } is fine as well. "Official" style indents the opening { by two spaces and the contents by 2 more spaces (instead of starting on the same indentation level as the line above). I'd not introduce spaces in g__ foo_ and in _ box, so I'd use \box_new:c { g__foo_ \int_to_roman:n {#1} _box } (but I don't know any official rule about this).

@cfr you're close enough to not nag about it, imho (and I don't abide by it 100% as well)

@John there are a few spaces I'd not use, but only inside the c-arguments, the rest looks pretty much like idiomatic L3-code.