<gchristensen>
weird... kswapd is taking 99% cpu but I don't have swap
<andi->
that is likely purging and reloading pages from disk? Did you consume all your RAM?
<gchristensen>
not that I know of
<gchristensen>
where on my disk is it putting pages?
<andi->
nowhere
<gchristensen>
hah
<andi->
from what I've seen (very unscientific) it is removing pages of libraries/applications in order to make space for others when you are running out of RAM (without swap).
<gchristensen>
ah
<andi->
Always nice to see that my home automation is working. Staying up late to listen to the debate and that single piece of automation is turning my amp off :D
<hexa->
when kswapd beats the debate
<gchristensen>
$750 filing fee lmao
<bqv>
cole-h: there already is one, fwiw
<bqv>
neovim nightly
<bqv>
i saw it in passing
<cole-h>
Not flakes enabled, I thought
<bqv>
oh, maybe not
<andi->
Whats the gain of using neovim nightly? Honest question. Thought about it as well but then I failed to see the gain.
<bqv>
i thought i'd seen it with the word flake but i'll defer to you
rajivr has joined #nixos-chat
<cole-h>
andi-: Nightly has built-in LSP support
<cole-h>
Fledgling, admittedly, but better than 0
<andi->
I thought 0.4 already had that?
<cole-h>
I don't think so
<andi->
I think it has some but it isn't as complete yet as some of the plugins
<andi->
but they might have postponed the integration
<andi->
Some day I'll sit down and cleanup my mess of VIM configuration....
<gchristensen>
meh
<gchristensen>
who cares
<samueldr>
is it managed with nix?
<andi->
of course it is..
<samueldr>
then who cares :)
<andi->
biggest closure on the system...
<samueldr>
it took me quite a while until I got mine in nix
<andi->
I failed to get it under any kind of proper version control before using Nix... All those VIM plugin managers and dotfile tools were not working out for me.
<gchristensen>
(it is a good thing sway's developers don't want sway to crash...)
<samueldr>
until it does?
<samueldr>
or should have?
<gchristensen>
it crashes all the time but they don't want it to, so it is fine
<samueldr>
right, I'd hoped it wasn't "crashes all the time" territory
<andi->
what do you do to crash it?
<gchristensen>
just about anything :P
<andi->
Must be that weird package manager you are using. This nix thingy.. have you tried alpine yet?
<abathur>
in Nix/OS's defense, I feel like it has a whole genre of it-is-great-but-annoying and it-is-annoying-but-great posts that most cults can't muster
<andi->
more of this cursed stuff! Back in the old days we just had our user ;)
<gchristensen>
:D
<andi->
abathur: that was a quote of someone that also doesn't like functional programming an might or might not be involved in the development of sway ;)
<pinpox>
Just wanted to know what happens when you self-karma
<pinpox>
Guess, infinisil thought of that too.
<etu>
pinpox++
<{^_^}>
pinpox's karma got increased to 0
<etu>
You shouldn't have to be below 0 :)
<pinpox>
thank you etu
dadada_ has quit [Quit: WeeChat 2.7.1]
dadada_ has joined #nixos-chat
avn has joined #nixos-chat
<ornxka>
the nixpkgs.chromium package was not updated in time for the latest chrome 0day and now when i turn my computer on it looks like this https://i.imgur.com/OVdv4Uf.jpeg what do
<viric>
is that a joke?
<ornxka>
yes
<viric>
ok
<gchristensen>
cool
<gchristensen>
I chuckled at cult o'frazer is
<gchristensen>
s/ is//
disasm has quit [Read error: Connection reset by peer]
disasm has joined #nixos-chat
rajivr has quit [Quit: Connection closed for inactivity]
cole-h has joined #nixos-chat
__monty__ has joined #nixos-chat
<cole-h>
Ever since upgrading my computer, 1) my keyboard doesn't randomly disconnect; and 2) my system never hangs after waking from sleep :D
<lassulus>
sounds good
<lassulus>
I have firefox weirdly crashing for me since yesterday with: (firefox:32749): Gdk-ERROR **: 22:10:28.131: The program 'firefox' received an X Window System error.
jared-w_ has joined #nixos-chat
aria_ has joined #nixos-chat
rizary_ has joined #nixos-chat
Valodim- has joined #nixos-chat
drakonis has quit [Ping timeout: 260 seconds]
aminechikhaoui has quit [Ping timeout: 240 seconds]
lopsided98_ has joined #nixos-chat
DavHau[m]1 has joined #nixos-chat
DavHau[m] has quit [*.net *.split]
tilcreator has quit [*.net *.split]
crazazy[m] has quit [*.net *.split]
aanderse has quit [*.net *.split]
CRTified[m] has quit [*.net *.split]
arcnmx has quit [*.net *.split]
lopsided98 has quit [*.net *.split]
colemickens has quit [*.net *.split]
aria has quit [*.net *.split]
rizary has quit [*.net *.split]
jared-w has quit [*.net *.split]
etu has quit [*.net *.split]
f0x has quit [*.net *.split]
Valodim has quit [*.net *.split]
ninjin has quit [*.net *.split]
Valodim- is now known as Valodim
jared-w_ is now known as jared-w
rizary_ is now known as rizary
aria_ is now known as aria
<hexa->
lassulus: on wayland?
<hexa->
LTO was disabled yesterday to fix issues with firefox on wayland
<lassulus>
no on X11
<lassulus>
xmonad
<lassulus>
but on 20.03
arcnmx has joined #nixos-chat
crazazy[m] has joined #nixos-chat
CRTified[m] has joined #nixos-chat
etu has joined #nixos-chat
tilcreator has joined #nixos-chat
f0x has joined #nixos-chat
aanderse has joined #nixos-chat
colemickens has joined #nixos-chat
aminechikhaoui has joined #nixos-chat
ninjin has joined #nixos-chat
drakonis has joined #nixos-chat
samueldr_ is now known as samueldr
samueldr has joined #nixos-chat
samueldr has quit [Changing host]
MichaelRaskin has joined #nixos-chat
<samueldr>
ugh, it's a bit painful when freenode reconnects you
<samueldr>
its nickserv identification at server connect is not instant, so your channel rejoins can happen when you're not +R'd yet
<gchristensen>
how about using SASL?
<samueldr>
exactly
<gchristensen>
:x
<samueldr>
that's what I'm using
<gchristensen>
...
<samueldr>
which means you might be trying to join channels before being identified
<samueldr>
which isn't all that bad... you have a list of channels to join later
<FireFly>
huh... that's weird, I believe with SASL the identify should happen before joins are issued
<samueldr>
you also are allowed to join channels that bans nick changes for unregistered users!
<samueldr>
FireFly: I believed so too
<samueldr>
but clearly that doesn't work
<samueldr>
and today I have tangible proof
<samueldr>
oh right
<samueldr>
RIGHT
<samueldr>
I know what failed
<samueldr>
I was logged into nickserv properly
<samueldr>
but using a wrong nick
<samueldr>
because "I was still there" in that netsplit
<samueldr>
so, that's part of the same problem, identity
<samueldr>
but yeah
<samueldr>
I couldn't even change my nickname!
<joepie91>
IRC \o/
<samueldr>
until I left 5 more channels
<FireFly>
weird, I don't think the nick should matter but I might be wrong :\
<joepie91>
samueldr: tip, on some networks (incl. Freenode) you can identify to a given username even if you don't have its nick
<samueldr>
as long as it won't force the nickname for the connection
<samueldr>
nah, I was wrong, as I said
<samueldr>
it's not that there was a delay
<samueldr>
it's that I wasn't +r'd because I was using a nick that wasn't part of the group
<cole-h>
Oh, I see.
* samueldr
adds samueldr_ to the group
<samueldr>
I might missing from channels, due to additional possible issues with joining too many channels :/ if you see me not joined to a channel I should be, tell me
<cole-h>
I see you in all the (important) ones I'm in
<samueldr>
I really regret not pushing, about 10 years ago, to build a "presence-first" IRC server, which acted like the IRC server was a BNC for anyone
<__monty__>
But that implies logging everything.
<sphalerite>
uuuh so like matrix? :D
<__monty__>
Tbf Matrix is intended to be much broader than just text chat.
<samueldr>
I don't see how that implies *logging* everything
<samueldr>
but it could be done yes
<samueldr>
if you want to also have the backlog
<__monty__>
samueldr: Then what part is the BNC for everyone?
<samueldr>
the server
<samueldr>
you log in into your account to access the server, so you could be logged-in from two irc clients at the same time, or none
<samueldr>
into the same username
<__monty__>
Oh, I'm not sure that's a common desired feature though?
<__monty__>
Afaict most people run a BNC for the backlog.
<samueldr>
I would say half and half
<samueldr>
a venn diagram with a huge common section
<__monty__>
Automatically ghosting a nick upon authentication of said nick would be enough, right?
<samueldr>
but yes, logging too
<samueldr>
nope
<samueldr>
not enough since you can't be connected on multiple devices *at the same time*
<samueldr>
connected on your phone and computer at the same time
<samueldr>
or even through a webchat for quick peek during courses on a school-owned computer
<samueldr>
anyway, I don't have to convince you
<samueldr>
I know it would have been a game changer
<samueldr>
couple that with a good webchat interface
<cole-h>
lol
<samueldr>
which was planned at the time!
<samueldr>
and *maybe* slack and discord wouldn't be the same
<samueldr>
>> the source code expressly suggests its use to copy and/or distribute the following copyrighted works owned by our member companies:
<samueldr>
eek
<samueldr>
I wonder what _actually_ is in the source
<FireFly>
uh yikes
NinjaTrappeur has quit [Ping timeout: 260 seconds]
<samueldr>
can the RIAA file DMCA notices on behalf of youtube even??
<__monty__>
Isn't it on behalf of their labels?
<samueldr>
it seems they're using the anticircumvention violation here
<samueldr>
>> Anticircumvention Violation. We also note that the provision or trafficking of the source code violates 17 USC §§1201(a)(2) and 1201(b)(1). The source code is a technology primarily designed or produced for the purpose of, and marketed for, circumventing a technological measure that effectively controls access to copyrighted sound recordings on YouTube
<samueldr>
>> In light of the above noted copyright infringements and anticircumvention violations
<samueldr>
I don't think there is actual copyright infringement (while IANAL)
<samueldr>
so that's weird
<samueldr>
>> Under penalty of perjury, we submit
<samueldr>
as if that meant anything here :)
<MichaelRaskin>
Right, I think the only thing that is actually under penalty of perjury is holding copyright approximately to anything at all
<__monty__>
Maybe no more youtube-dl would be good for me. So much less time wasted watching youtube.
<MichaelRaskin>
We need fetchmega
hmpffff has quit [Ping timeout: 272 seconds]
hmpffff has joined #nixos-chat
maxdevjs has quit [Ping timeout: 256 seconds]
maxdevjs has joined #nixos-chat
maxdevjs has quit [Remote host closed the connection]
maxdevjs has joined #nixos-chat
pumpy has joined #nixos-chat
pumpy has left #nixos-chat [#nixos-chat]
<gchristensen>
looks like like I'mma be writing some Go for the first time in nearly an eternity ...
<samueldr>
maybe you can write some Rust and say "ah, it's a new fork of Go"
<gchristensen>
haha
<sphalerite>
or write a rust-to-go transpiler?
<gchristensen>
the trouble is the really nice cloudy Go libraries
<samueldr>
make Go bindings for Rust?
<gchristensen>
hah
<samueldr>
I know, the Go ABI is like... nothing else
<gchristensen>
I see we're in a "bake an apple pie from scratch" mood
<samueldr>
I call that "everyday"
<gchristensen>
:D
<gchristensen>
my "ehh.." towards Go and my "I will regret this dearly" w.r.t. Bash has me slightly paralyzed and I might just go play with Tycho instead :P
<sphalerite>
gchristensen: sounds like the option that has the most value in the future as well :)
<gchristensen>
I wonder how Simon feels about crashes. does he want sway to crash?
<samueldr>
nah, it shouldn't
<andi->
I'd rather have it crash then execute random instructions :)
<colemickens>
Maybe instead of openapi2nix, I want openapi2nickel...
<colemickens>
given Nickel's stated goals...
<colemickens>
is there a #nickel I wonder
<andi->
there is tooling for openapi2dhall IIRC
<colemickens>
hm, I see something, and then there's dhall-kubernetes that I believe is generated from the openapi specs.
<andi->
yeah
<andi->
what are you trying to use it for?
<colemickens>
Azure IaaS APIs have OpenAPI schemas for their ARM resources.
<colemickens>
I want to specify my IaaS resources in Nix and then have it generates the raw JSON documents to POST.
<andi->
sounds like a job for dhall. I recently wrote a few documents with dhall-kubernetes and they weren't even valid afterwards. Apparently those openapi specs aren't "complete" for the simplest examples of running a hello world webserver.
<colemickens>
I am quite excited about Nickel's stated goals. Having typed modules for something like this, and then letting the user specify untyped (or inferred, I guess) input would be highly ideal.
<andi->
I would love to see Nix used instead by ripping out the store from the evaluator. Users, that aren't by heart functional programmers, would only have to learn a single language and not two, three, …
<joepie91>
samueldr: the takedown notice is a huge crock of shit
<joepie91>
"expressly suggests its use to copy and/or distribute the following copyrighted works" refers to the URLs to some of 'their' videos being part of the test suite
<samueldr>
joepie91: I know
<joepie91>
likewise, I'm pretty sure that the observation of "you can't actually demand takedown this way for circumvention" is correct
<joepie91>
as the process is only intended for actual infringements
<joepie91>
by all appearances this basically looks like intimidation
<samueldr>
it _is_, but it could be with legit or ill intent!
<joepie91>
it's the RIAA
<joepie91>
I don't think there's any question about intent
<samueldr>
joepie91: I was using legit as "respecting the current legal framework", without casting judgement about it :)
<joepie91>
samueldr: considering it's an organization that's 50% lawyers by volume, I strongly doubt that it was an innocuous mistake, so I don't believe that they were intending to respect the legal framework
<samueldr>
ah, I meant that in the larger sense; they always do intimidation, sometimes respecting the framework, sometimes not
<drakonis>
colemickens: sadly not yet
<drakonis>
i'm looking at the commit history and i'd honestly be surprised if it didnt replace nix lang
<colemickens>
no, I feel like I've read (anecdotally) that it didn't work as well as wlr-obs, but I've not tried it myself.
<colemickens>
(and I've been sorta having weird issues with xdg-portal lately, though it is working again now)
<ashkitten>
according to #wayland xdg-portal should work to capture individual windows whereas that is not supported in wlrobs
<ashkitten>
(i haven't gotten that working in firefox though, only entire screens... maybe that's a firefox bug?)
<colemickens>
I think that would have to be supported by xdg-desktop-portal-wlr and my understanding was that it didn't (yet) support per-window selection.
<ashkitten>
ah
<colemickens>
they don't even have output selection yet
<drakonis>
nickel looks incredibly promising as a clean slate nix replacement
<drakonis>
given a potential way migration route into it
<colemickens>
ashkitten: imo, it's a big feature cap in this year of our lady coronavirus. I hope it's something someone decides to champion and push through.
<ashkitten>
mhm
<andi->
Migration would be nice if Nickel would support 100% Nix ;)
<andi->
Not looking forward to having a split between the languages and constantly having a wrapper/language barrier between the two
<drakonis>
i'm looking forward to nickel all the things and do away with dependencies on third party languages
<ashkitten>
btw, colemickens, do you know why we're supposed to set XDG_CURRENT_DESKTOP=sway for the portal to work? i feel like if it's important then sway should be setting that itself
<colemickens>
I think that there's a binding somehow between sway->wlr portal. I asked about this a long while back. Let me see if I can find anything to substantiate that.