💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › kisslinux-2022-10-12.txt captured on 2023-01-29 at 03:45:03.
-=-=-=-=-=-=-
[2022-10-12T03:30:25Z] <wael_> Hi [2022-10-12T03:30:41Z] <ioraff> hi [2022-10-12T04:41:27Z] <testuser[m]> Hi [2022-10-12T04:54:04Z] <saturn[m]> hello [2022-10-12T07:16:56Z] <rohan> hi [2022-10-12T07:17:12Z] <rohan> how can i disable iwd verbose on init? [2022-10-12T07:17:16Z] <rohan> and busybox verbose [2022-10-12T07:23:46Z] <testuser[m]> GalaxyNova: hi [2022-10-12T07:30:01Z] <virutalmachineus> hi [2022-10-12T07:38:08Z] <wael_> rohan: have a system logger [2022-10-12T11:43:17Z] <midfavila> there's a kernel parameter you can set to change the minimum urgency level required before debug messages are printed to the console, rohan [2022-10-12T11:43:24Z] <midfavila> i can't quite recall which, though... [2022-10-12T11:43:49Z] <wael_> loglevel=4 [2022-10-12T11:43:51Z] <wael_> or quiet [2022-10-12T11:44:08Z] <midfavila> yeah so there you go [2022-10-12T11:44:11Z] <midfavila> no system logger necessary [2022-10-12T11:44:28Z] <wael_> doesn't it print to stdout and not kernel log [2022-10-12T11:45:11Z] <midfavila> iwd spams both in my experience [2022-10-12T11:45:21Z] <midfavila> you can also just direct stdout to null or something [2022-10-12T11:45:56Z] <testuser[m]> so bad [2022-10-12T11:46:03Z] <testuser[m]> make logger [2022-10-12T11:46:30Z] * midfavila shrugs? [2022-10-12T12:37:52Z] <testuser[m]> It's not in GNU's interest to break my program, but it's also not not in GNU's interest to break my program. GNU wants us to write programs for GNU. A proliferation of programs that only work under GNU makes it more attractive to install GNU, and less attractive to install alternatives. [2022-10-12T13:05:58Z] <testuser[m]> phoebos: any idea why man3 is not installed in man-pages? it doesn't conflict with anything [2022-10-12T13:06:03Z] <testuser[m]> from linux pages [2022-10-12T13:34:12Z] <phoebos> probably because man3 is glibc related [2022-10-12T13:34:19Z] <phoebos> man3p covers basically all of musl [2022-10-12T13:36:35Z] <testuser[m]> liek getopt_long [2022-10-12T13:36:37Z] <testuser[m]> musl implements extensions too [2022-10-12T13:50:41Z] <testuser[m]> frexpf [2022-10-12T13:51:48Z] <testuser[m]> getentropy [2022-10-12T13:52:55Z] <testuser[m]> malloc_usable_size [2022-10-12T13:54:11Z] <testuser[m]> qsort_r , was recently added [2022-10-12T13:54:15Z] <testuser[m]> to musl [2022-10-12T15:41:16Z] <phoebos> i have no objection to adding them but i'm guessing that's what dylan was thinking [2022-10-12T16:44:55Z] <testuser[m]> _<PID>_GNU_nonoption_argv_flags_ This variable was used by bash(1) 2.0 to communicate to glibc which arguments are the results of wildcard expansion and so should not be considered as options. This behavior was removed in bash(1) version 2.01, but the support remains in glibc. [2022-10-12T16:45:20Z] <testuser[m]> phoebos: ok [2022-10-12T17:11:56Z] <midfavila> https://inv.riverside.rocks/channel/UC7-CyoYfsrVI-dsuHRQx0IQ [2022-10-12T17:12:06Z] <midfavila> ^lectures to accompany SIT's APUE course [2022-10-12T17:12:14Z] <midfavila> figured you guys might find them useful [2022-10-12T23:18:00Z] <GalaxyNova> standards are actuallly pretty bad [2022-10-12T23:18:16Z] <midfavila> that's actually a stupid take. [2022-10-12T23:20:00Z] <GalaxyNova> well somewhat [2022-10-12T23:20:08Z] <GalaxyNova> my statement was a bit exagerated [2022-10-12T23:21:30Z] <GalaxyNova> standardization isn't inherently bad [2022-10-12T23:22:03Z] <GalaxyNova> but most of the standards we have today are filled with decades worth of cruft and historical garbage that is kept around only for compatibility with older versions of the standards [2022-10-12T23:22:36Z] <GalaxyNova> over time this causes standards to become larger and larger [2022-10-12T23:23:23Z] <midfavila> legacy compatability is one of the main advantages of standards [2022-10-12T23:23:38Z] <midfavila> compatibility* [2022-10-12T23:24:00Z] <midfavila> obviously if you look at stuff like what GNU does, where they throw out new ideas every five minutes and then deprecate them in another five, that causes issues [2022-10-12T23:24:48Z] <midfavila> but idk, you look at, say, POSIX - sure, there are a handful of functions kept around for legacy support exclusively, but it's trivial to just... ignore those, or rewrite programs that use them [2022-10-12T23:25:23Z] <GalaxyNova> you can say that for anything [2022-10-12T23:25:37Z] <midfavila> yes, that's... kind of the point [2022-10-12T23:25:47Z] <midfavila> your statement can be applied to anything, but so can its inverse [2022-10-12T23:25:49Z] <GalaxyNova> well, there's a reason people use C over something like C++ [2022-10-12T23:26:03Z] <GalaxyNova> when they could easily just ignore the features C++ gives them that they don't need [2022-10-12T23:26:23Z] <GalaxyNova> but the point is that it dirties the whole ecosystem [2022-10-12T23:26:46Z] <midfavila> sure, if you're a compiler writer (in the case of C++) [2022-10-12T23:27:02Z] <midfavila> so... don't write C++ compilers, or only implement a subset of it [2022-10-12T23:27:17Z] <midfavila> just because some standards are handled poorly doesn't mean all standards are, or even most [2022-10-12T23:28:56Z] <GalaxyNova> legacy compatibility isn't as important as people make it out to be [2022-10-12T23:29:02Z] <GalaxyNova> compatibility layers are a thing [2022-10-12T23:29:08Z] <midfavila> you uh [2022-10-12T23:29:29Z] <midfavila> you *realize* compat layers implement part or all of the abstract spec you're trashing, [2022-10-12T23:29:38Z] <midfavila> *in addition* to a bunch of other shit to translate it to underlying system calls? [2022-10-12T23:29:55Z] <midfavila> anyway, "legacy support" is critical [2022-10-12T23:30:06Z] <midfavila> it implies that the standard gives you time to actually use it [2022-10-12T23:30:20Z] <GalaxyNova> it's still better than trashing the base OS [2022-10-12T23:30:26Z] <midfavila> it's literally worse [2022-10-12T23:31:02Z] <midfavila> if you're going to implement the standard anyway, and you're going to use it in a large number of system programs (or even a moderate amount), you should just... implement the standard [2022-10-12T23:31:43Z] <GalaxyNova> most people that use OSS software don't have a use for legacy compatibility, so they won't need the compatibility layer [2022-10-12T23:32:06Z] <GalaxyNova> since most software is just recompiled or patched when the OS breaks compatibility [2022-10-12T23:32:41Z] <GalaxyNova> and also standards like POSIX and ISO C are not free [2022-10-12T23:32:54Z] <midfavila> if the standards that those programs adhere to change, because "muh legacy", then you need to rewrite the programs [2022-10-12T23:33:11Z] <midfavila> end result? you spend a fuckton of wasted time rewriting things every time the standards change [2022-10-12T23:33:48Z] <midfavila> and I fail to see what you mean by "free" in that context [2022-10-12T23:34:17Z] <GalaxyNova> https://www.iso.org/standard/74528.html [2022-10-12T23:34:26Z] <GalaxyNova> notice how it costs hundreds of dollars to buy the standard [2022-10-12T23:35:30Z] <GalaxyNova> it's all copyrighted and locked down behind made up IP laws [2022-10-12T23:36:07Z] <midfavila> uh, just download it off libgen [2022-10-12T23:36:11Z] <midfavila> if the money is the problem [2022-10-12T23:38:47Z] <midfavila> http://www.open-std.org/jtc1/sc22/WG14/www/docs/n1256.pdf [2022-10-12T23:38:56Z] <midfavila> seems like you can get a copy of the C standard here [2022-10-12T23:39:06Z] <GalaxyNova> that's a draft [2022-10-12T23:39:27Z] <GalaxyNova> I'm pretty sure there's some copyright law involved in actually implementing the standard [2022-10-12T23:39:34Z] <GalaxyNova> that requires you own a copy of it [2022-10-12T23:40:39Z] <GalaxyNova> POSIX is open though, i think [2022-10-12T23:40:47Z] <midfavila> the standard is, certification isn't [2022-10-12T23:41:21Z] <midfavila> i doubt the bit about them trying to force you to buy a copy of the standard to implement it is enforceable