[2022-07-23T01:41:13Z] illiliti: have you been sending patches to the muon list without git-send-email? [2022-07-23T01:49:40Z] nevermind. https://lists.sr.ht/~lattis/muon/patches/34099 [2022-07-23T04:05:48Z] Hi [2022-07-23T04:37:26Z] hi [2022-07-23T05:09:44Z] hi [2022-07-23T12:52:09Z] Hi everybody [2022-07-23T12:52:46Z] Does any of you have done a package for clang-tools-extra ? [2022-07-23T12:52:57Z] (I try to install clangd) [2022-07-23T12:57:35Z] https://codeberg.org/git-bruh/kiss-repo/src/branch/master/overrides/llvm/build [2022-07-23T12:57:43Z] Sanitizers clang tools everything [2022-07-23T13:05:13Z] Ty <4 [2022-07-23T13:05:17Z] <3* [2022-07-23T15:56:04Z] ioraff: thanks for the patch. did you test it with foot-pgo? [2022-07-23T15:56:13Z] and i use git-send-email+msmtp if you still interested [2022-07-23T16:15:04Z] illiliti: yes [2022-07-23T16:24:17Z] do you know when your mesa MR is supposed to appear in a release? [2022-07-23T16:30:46Z] Which [2022-07-23T16:31:32Z] https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/15715 [2022-07-23T16:37:24Z] hey nerds [2022-07-23T16:39:16Z] midfavila: hi [2022-07-23T17:12:56Z] what did you just call me [2022-07-23T17:13:32Z] Nerd [2022-07-23T17:14:57Z] o/ [2022-07-23T17:29:54Z] what have you all been up to [2022-07-23T17:47:14Z] you fucking heard me, noocsharp [2022-07-23T17:47:21Z] what're you gonna do about it, [2022-07-23T17:47:22Z] n e r d? [2022-07-23T17:47:48Z] phoebos, I haven't been up to much, myself. Working on a shell client for invidious, got a new mathematics book, etc. [2022-07-23T17:47:51Z] you? [2022-07-23T17:59:58Z] midfavila: not much either. need to have a project but can't be motivated [2022-07-23T18:36:30Z] phoebos: how's your javascript? [2022-07-23T18:37:28Z] ioraff: no idea tbh [2022-07-23T18:38:17Z] haven't touched it for like 3 years, why [2022-07-23T18:38:30Z] used to do stuff with canvas [2022-07-23T18:44:13Z] i started working on a little collaborative text editing server [2022-07-23T18:45:13Z] it works over websocket, so i have a really simple browser editor [2022-07-23T18:45:42Z] trying to escape google docs basically [2022-07-23T18:46:55Z] cryptpad [2022-07-23T18:49:46Z] how far have you got [2022-07-23T18:50:36Z] better idea: an ed server that works like irc [2022-07-23T18:50:52Z] log in, send commands, other people can see what you did [2022-07-23T18:51:14Z] server forks on new connection and sends everything to all the file descriptors [2022-07-23T18:51:22Z] first file descriptor is connected to ed [2022-07-23T18:54:48Z] testuser[m]12: that actually looks cool [2022-07-23T18:56:31Z] phoebos: it's not collaborative editing if only one person will ever use it [2022-07-23T18:58:07Z] if two people who could use ed met irl they might use it together! [2022-07-23T18:59:30Z] haha good one [2022-07-23T19:00:53Z] but honestly, i have no idea how to do that stuff in javascript [2022-07-23T19:01:15Z] sounds like the kind of stuff eMploYeRS want though [2022-07-23T19:03:02Z] it's pretty simple, it just reads data from a websocket connection and modifies a textarea [2022-07-23T19:04:56Z] can i look at your code? [2022-07-23T19:05:41Z] but also, fuck you, i'm gonna make this ed thing and you and i *will* use it [2022-07-23T19:08:39Z] if you say so [2022-07-23T19:09:51Z] here's the js code: https://www.nihaljere.xyz/files/main.js [2022-07-23T19:09:58Z] obviously unusable since you don't have the server [2022-07-23T19:10:44Z] where reactjs expressjs serverless mongo [2022-07-23T19:12:42Z] cryptpad is written in js btw [2022-07-23T19:13:34Z] ik [2022-07-23T19:13:53Z] > let cid = raw[0] + (raw[1] << 8) + (raw[2] << 16) + (raw[3] << 24); [2022-07-23T19:14:06Z] uint64_t cid = (uint64_t)(*raw); [2022-07-23T19:14:26Z] s/64/32/ [2022-07-23T19:14:27Z] g [2022-07-23T19:15:28Z] well you're assuming the host is little endian there [2022-07-23T19:16:06Z] ntohl or something [2022-07-23T19:16:16Z] isn't *raw coming from your server [2022-07-23T19:16:27Z] yeah [2022-07-23T19:16:53Z] so explicitly send it as LE [2022-07-23T19:16:59Z] i do [2022-07-23T19:17:22Z] oh [2022-07-23T19:17:31Z] but js byte order isn't specified i don't think [2022-07-23T19:17:48Z] ach [2022-07-23T19:18:34Z] the browser virtual machine is certainly nicer than X to code for in some ways [2022-07-23T19:21:29Z] i watched a talk recently about how the v8 people got browser js to run half as fast as C which is very impressive [2022-07-23T19:21:50Z] well c++ actually, doing prime number generation [2022-07-23T19:25:31Z] bleh, that doesn't bode well [2022-07-23T19:28:25Z] bode well for what [2022-07-23T19:29:06Z] PC software as an ecosystem [2022-07-23T19:29:20Z] if JS runs faster, then that makes the whole "JS is slow as shit" talking point less valid . -. [2022-07-23T19:29:51Z] makes it less likely people'll develop desktop applications where performance is a factor, I bet. [2022-07-23T19:30:06Z] js is shit [2022-07-23T19:35:01Z] you can write poorly performing code in any language [2022-07-23T19:35:50Z] just stick to programs written by people you trust to not write bad code and you'll be fine [2022-07-23T19:36:49Z] and js will always be slower than c/c++ [2022-07-23T19:37:23Z] sure, but my point is that if it's "good enough", then there's not going to be as much of a reason to write software natively [2022-07-23T19:37:50Z] which means doing stuff outside of the browser becomes increasingly difficult to sustain, in some cases [2022-07-23T19:43:15Z] i mean browsers are mostly user facing software anyway, anything that actually requires performance will still be written natively [2022-07-23T19:43:26Z] that is, the stuff backing the user facing software [2022-07-23T19:45:17Z] at the very least, you can see exactly what code in the browser does which lets you replicate it natively [2022-07-23T19:48:07Z] is there any browser design that doesn't suck? i mean the concept is very prone to over-engineering and bloatware [2022-07-23T19:48:20Z] mothra [2022-07-23T19:48:22Z] * midfavila nods [2022-07-23T19:48:46Z] links is also based [2022-07-23T19:49:09Z] links is cool but actually crap [2022-07-23T19:49:17Z] the code is a big fat mess [2022-07-23T19:49:23Z] i use lynx to read html email [2022-07-23T19:49:45Z] howso, phoebos? do you think it's salvageable? [2022-07-23T19:50:14Z] noocsharp: me too, because links -dump doesn't print the urls along with hyperlink names [2022-07-23T19:50:29Z] midfavila: didn't you play with the code once? [2022-07-23T19:50:33Z] there's a lot of hard-coded stuff [2022-07-23T19:50:39Z] eg. keybindings [2022-07-23T19:50:51Z] only enough to kick a patch I found into working order [2022-07-23T19:51:01Z] i've considered forking links for a while now [2022-07-23T19:51:25Z] try to make it leaner and easier to integrate into a system [2022-07-23T19:51:42Z] be cool to optionally embed a scheme for such purposes [2022-07-23T19:51:46Z] some of the elinks/lynx programs are forks of links i think? [2022-07-23T19:51:55Z] elinks is a fork, lynx and links are unrelated [2022-07-23T19:51:58Z] ok [2022-07-23T19:52:04Z] have you used nyxt [2022-07-23T19:52:11Z] yes [2022-07-23T19:52:16Z] when it worked it was fairly comfy [2022-07-23T19:52:36Z] i was going to say that nyxt gets an honorable mention, but it uses common lisp and is a pain in the ass to build aside from that [2022-07-23T19:53:05Z] ...of course, common lisp is *mostly* fine, but they rely heavily on SBCL [2022-07-23T22:38:31Z] I got gimp, inkscape and kicad working. But at what cost? D: [2022-07-23T22:44:33Z] I saw someone patched out atk from gtk. A mega-patch. Must have taken a lot of effort.