01:37 <phoebos> hi claudia02

03:03 <noocsharp> i saw you on lobste.rs phoebos

06:16 <testuser[m]> Hi

08:17 <kwlw4> Hi

08:18 <kwlw4> testuser[m]: is it possible to use sbase with glibc?

08:18 <kwlw4> and nvidia drivers

08:22 <kwlw4> I'm just trying to make my system as minimal as possible

09:14 <stjohn> kwlw4: You can definitely compile sbase with glibc, but I don't know if you can compile anything else with sbase. I've tried it and immediately run into weird problems that come from the fact that sbase hasn't implemented something yet or has implemented it differently. Their tar and sed are particularly troublesome. If you want to use sbase, the safest thing to do is compile everything else first with busybox, t

09:14 <stjohn> en you're done.

09:17 <kwlw4> stjohn: I think I'll stick with busybox, thanks :)

13:03 <phoebos> noocsharp: i'm new there ^^

13:13 <claudia02> hola phoebos

15:07 <riteo> hiiiiiiiiiiiiiii!

15:13 <riteo> just sent what nvidiaLinuxUser asked me on shithub: https://github.com/kiss-community/community/pull/1139

15:15 <riteo> how do you guys handle bleeding edge packages? The latest kakoune release is wayyyy too old and we're loosing lots of goodies like a hare highlighter; should I make it git or just track the current HEAD every once in a while?

15:16 <riteo> the first approach should be the "freshest" while the second one should be the most stable I think, since the package is pretty much garaunteed to compile (ignoring compiler issues such as version changes)

15:18 <illiliti_> how about asking mawww to cut a release first?

15:21 <riteo> I read a bit and it looks like they want to be absolutely sure that it's stable or something like that

15:21 <riteo> and it's still getting actively updated anyway, so something might be in the works

15:22 <riteo> I read that the goal was to release monthly, but mawww got busy with life

15:22 <illiliti_> i would ask anyway

15:22 <riteo> perhaps we could package our own monthly releases?

15:22 <illiliti_> no

15:23 <riteo> yeah it would be more a fork than anything in that case

15:24 <riteo> there was this issue already asking for it last year: https://github.com/mawww/kakoune/issues/4290

15:26 <illiliti_> it's time to remind about release again

15:27 <riteo> *sigh*, I don't want to be *that* guy asking for ETAs, but I think that asking mawww's point of view in general shouldn't be that bad

15:32 <illiliti_> yep

15:32 <illiliti_> simple question wouldn't hurt. no need to even mention etas

15:33 <riteo> done, on their IRC channel

15:38 <testuser[m]> I find it great that people who don’t come from a computer science can learn and become so knowledgeable about programming. The rust community is so great and welcoming

15:53 <stjohn> testuser[m]: In my experience people can walk out of a computer science degree and still not know the first thing about programming. They're different disciplines, even if they do overlap.

15:54 <testuser[m]> It's a copy pasta

15:55 <illiliti_> it's not limited to programming btw

16:00 <stjohn> Oh? How's that?

16:26 <illiliti_> i've met plenty of people who has degree of something and they have absolutely no idea how things are working in their profession

16:26 <illiliti_> dunno, maybe it's just experience matter or bad education(most likely, at least in ~~my~~ country)

16:26 <illiliti_> also they are not willing to improve, progress

16:27 <illiliti_> basically their progress has stopped when they graduated from some college it seems

16:27 <testuser[m]> https://github.com/hyprwm/Hyprland

16:36 <illiliti_> it seems it doesn't support xdg spec as well as posix TMPDIR

16:36 <illiliti_> but interesting

16:38 <stjohn> illiliti_: You're probably right, part of it is a matter of lacking real-world experience. But some people go into computing just because it pays well, and then they're totally unable to learn anything new when they get a real job. They can pass exams, but they can't actually be a programmer. It can be spectacular to watch them crash and burn --- provided you're not the one who has to clean up the wreckage :P

17:20 <riteo> news from #kakoune: mawww's working on breaking changes, but if people ask enough he might make a release

17:21 <riteo> someone said that "master is almost always releaseable", so I'm tempted to just either randomly pin HEAD every once in a while or to go straight to a `git` version

17:27 <riteo> guys, random question: are `\n`s line separators or terminators?

17:28 <riteo> because I heard that stuff subtly breaks if there's no EOF newline, but most software visualizes newlines as spaces

17:29 <riteo> like, I don't know if you noticed, but Dylan made the help pages with 80 column wide headers with an extra newline at the end, which breaks on an 80 column terminal on busybox `less` but not `more` nor `cat`

17:37 <riteo> ok, POSIX defines a line as: "A sequence of zero or more non- <newline> characters plus a terminating <newline> character.". Is thus an 80 column line 79 characters + a newline or 80 characters + a newline?

17:40 <testuser[m]> Probably the former

17:40 <testuser[m]> Btw use `less -S`

17:41 <riteo> well that hides the issue, doesn't it?

17:41 <testuser[m]> Yeah

17:41 <testuser[m]> But u should use -S anyways

17:42 <riteo> Why? Can I scroll right?

17:42 <testuser[m]> Yeah

17:42 <riteo> oh, nice

17:42 <testuser[m]> It's much less messy

17:42 <riteo> I'll change my PAGER env var then

17:43 <riteo> umh

17:44 <riteo> `/usr/bin/kiss-help: line 35: less -S: not found`

17:44 <riteo> am I dumb or is this a kiss-help bug

17:44 <testuser[m]> it's executing "$PAGER" instead of $PAGER

17:44 <riteo> bruh

17:45 <riteo> probably blind shellcheck compliance, dunno

17:45 <riteo> oh well

17:46 <riteo> talking about the line thing, the most annoying part is that more and cat blindly output stuff and it works fine, fills the whole terminal. While less and everything else put that space to represent the newline

17:46 <riteo> so there's a pretty obvious inconsistency

17:57 <riteo> I mean, POSIX defines an empty line as: `A line consisting of only a <newline>`

17:57 <riteo> so I think that actually the right definition is 80 columns of chars + newline, and not the more common 79 columns of char + newline

17:58 <riteo> and that is a huge mess

18:07 <riteo> also a blank line is "a line consisting solely of zero or more <blank> characters terminated by a <newline>"

18:11 <riteo> wtf, newlines are also in the space character class

18:11 <testuser[m]> Yeah

18:12 <riteo> but it also says "any characters included in the class blank are automatically included in this class."

18:13 <riteo> I'm severely confused rn

18:21 <riteo> well, terminals surely interpret newlines as 0 column wide

18:21 <riteo> it's those pesky editors/viewers that for some unkown reason visualize them as 1 column wide spaces

18:35 <riteo> fuck, control characters seem to have a 1 column width

18:35 <riteo> in unicode world, I mean

18:38 <riteo> I think that that's the issue, unicode defines stuff differently from terminals

18:39 <riteo> well, gotta go for now, although I'll still read the logs if you want to scream at random tools with me

18:39 <riteo> cya!