<aanderse>
grahamc: we're at the point where everyone who is listed as a maintainer should be part of the NixOS project, right?
<lovesegfault>
Restarted https://hydra.nixos.org/build/108681230 to see if it will pick-up the previous restart (which succeeded since it was a spurious failure)
<lovesegfault>
aanderse: I'm a maintainer but not part of the NixOS project
<aanderse>
lovesegfault: is that by choice?
<lovesegfault>
aanderse: Nope :P
<aanderse>
grahamc: if lovesegfault is a maintainer and not in the nixos organization is this on purpose?
<aanderse>
lovesegfault: maybe we didn't get around to adding everyone yet
ris has quit [Ping timeout: 258 seconds]
<lovesegfault>
aanderse: I don't think everyone was added; it's a lot of people
<aanderse>
yeah i mean with 750 people in the organization i figured that was it...
justanotheruser has quit [Ping timeout: 248 seconds]
<aanderse>
i guess we have more than 750 people who are involved or signed up as maintainers
<aanderse>
wow :)
<lovesegfault>
It's nuts
<lovesegfault>
Although I wonder what the active number is
justanotheruser has joined #nixos-dev
<drakonis>
that's too many people tho
<drakonis>
where do we draw the line for being part of the project or just a maintainer?
<qyliss>
What does "part of the project" even mean to you?
<qyliss>
Part of the GitHub organization?
<drakonis>
as far contributions go, not being a one off
<drakonis>
someone that continually does significant work
<qyliss>
Well, being listed as a maintainer is a promise to keep a package up to date
<qyliss>
But I'm wondering, what do you think it's too many people _for_?
justanotheruser has quit [Read error: Connection reset by peer]
<drakonis>
never too many maintainers
justanotheruser has joined #nixos-dev
justanotheruser has quit [Client Quit]
bhipple has quit [Ping timeout: 265 seconds]
bhipple has joined #nixos-dev
<worldofpeace>
If you have a contribution to NixOS you've given a part to the project
<worldofpeace>
There's always different kinds of "memberships", this is just one with a label for what it is. A maintainer.
<drakonis>
fair enough
drakonis has quit [Quit: Leaving]
justanotheruser has joined #nixos-dev
justanotheruser has quit [Ping timeout: 268 seconds]
pie_ has quit [Ping timeout: 260 seconds]
bhipple has quit [Remote host closed the connection]
zarel has quit [Ping timeout: 258 seconds]
zarel has joined #nixos-dev
zarel_ has joined #nixos-dev
zarel has quit [Ping timeout: 260 seconds]
__monty__ has joined #nixos-dev
<yorick>
gchristensen: do you do anything special to configure mcelog?
<yorick>
apparently rasdaemon is the new mcelog
<yorick>
but it doesn't support soft-offlining
<aanderse>
gchristensen: when you're around ... how should we go about adding people to the nixos-maintainers team if they were somehow missed?
<gchristensen>
aanderse: ah, I can do that today. I suspect I didn't get it running on a schedule.
<aanderse>
gchristensen: cool! sounds good. so your magic should automatically pickup everyone who is listed as a maintainer, right? like @lovesegfault and @emmanualrosa
<gchristensen>
yeah
<gchristensen>
I would just go ahead and run it now, but it would sort of kill my motivation to make it automatic :P
<aanderse>
right?
<aanderse>
good call then :)
<aanderse>
well... thank you very much :D
* gchristensen
heads to the office
pie_ has joined #nixos-dev
{^_^} has joined #nixos-dev
<arianvp>
gchristensen: so you did tpm stuff on nixos?
<gchristensen>
is this the way to teach packages about static-ey-ness?
pepesza has quit [Read error: Connection reset by peer]
drakonis has joined #nixos-dev
<gchristensen>
okay so before we can get this really going nicely, the maintainer-list.nix needs to be validated by ofborg because wow
<Taneb>
I think the .travis.yml for nixops is out-of-date, my PR is failing because we're trying to create /etc/nix in the before-script, which already exists
psyanticy has joined #nixos-dev
<infinisil>
gchristensen: How about just a Nix function for validation
<gchristensen>
that might be good
<infinisil>
Like `let maintainers = { ... }; validate = ...; in validate maintainers
<gchristensen>
some problems: githubId being missing, githubId being a string, github (name) missing
<infinisil>
gchristensen: Oh, how about it automatically errors with the correct githubId if it's not set
<infinisil>
That would be fancy
<gchristensen>
that would be cool
yegortimoshenko has quit [Quit: killed]
Ox4A6F has quit [Quit: killed]
worldofpeace has quit [Quit: killed]
domenkozar[m] has quit [Quit: killed]
jtojnar has quit [Quit: killed]
jonge[m] has quit [Quit: killed]
ma27[m] has quit [Quit: killed]
Ericson2314 has quit [Quit: killed]
roberth has quit [Quit: killed]
mkg20001 has quit [Quit: killed]
thefloweringash has quit [Quit: killed]
timokau[m] has quit [Quit: killed]
vaibhavsagar has quit [Quit: killed]
abbradar[m] has quit [Quit: killed]
rnhmjoj has quit [Quit: killed]
layus[m] has quit [Quit: killed]
aanderse has quit [Quit: killed]
Nyanloutre[m] has quit [Quit: killed]
arcnmx has quit [Quit: killed]
rycee has quit [Quit: killed]
nh2[m] has quit [Quit: killed]
alienpirate5 has quit [Quit: killed]
bennofs[m] has quit [Quit: killed]
vaibhavsagar has joined #nixos-dev
ma27[m] has joined #nixos-dev
bennofs[m] has joined #nixos-dev
jonge[m] has joined #nixos-dev
rycee has joined #nixos-dev
arcnmx has joined #nixos-dev
mkg20001 has joined #nixos-dev
jtojnar has joined #nixos-dev
Ericson2314 has joined #nixos-dev
abbradar[m] has joined #nixos-dev
dywedir[m] has joined #nixos-dev
Ox4A6F has joined #nixos-dev
dtz has joined #nixos-dev
domenkozar[m] has joined #nixos-dev
yegortimoshenko has joined #nixos-dev
worldofpeace has joined #nixos-dev
Nyanloutre[m] has joined #nixos-dev
nh2[m] has joined #nixos-dev
layus[m] has joined #nixos-dev
alienpirate5 has joined #nixos-dev
rnhmjoj has joined #nixos-dev
thefloweringash has joined #nixos-dev
roberth has joined #nixos-dev
aanderse has joined #nixos-dev
timokau[m] has joined #nixos-dev
ris has joined #nixos-dev
psyanticy has quit [Quit: Connection closed for inactivity]
pie_ has quit [Ping timeout: 258 seconds]
ixxie has joined #nixos-dev
pie_ has joined #nixos-dev
<worldofpeace>
hmm, I think I've been guilty for a few githubid is a string cases 😁 I assumed ofborg validated it in the first place
ixxie has quit [Ping timeout: 268 seconds]
ixxie has joined #nixos-dev
tilpner_ has joined #nixos-dev
tilpner has quit [Ping timeout: 268 seconds]
ixxie has quit [Ping timeout: 265 seconds]
<gchristensen>
FYI it appears spam bots are practicing again, so we may see some activity
<infinisil>
I have another idea I'm investigating for potentially rescuing NixOS eval performance. I'll quickly mention the gist of it in case somebody sees an obvious problem with it
<MichaelRaskin>
Switch to actual datastructures in Nix eval code?
<infinisil>
Nah :)
<infinisil>
The idea is to have a dependency system between modules, where you can optionally declare the dependencies for a module. If all modules do that, only those modules get imported, otherwise it falls back to all of them
<infinisil>
So the tradeoff here is that in order to get the benefits, all nixpkgs modules need to be changed
<infinisil>
And people need to opt-in their configuration.nix by declaring their dependencies
<infinisil>
So this is backwards compatible, but for the ones who want the faster eval, they can have it
<samueldr>
how do you define "a module"?
<samueldr>
as of right now, any file can add options anywhere in the options "tree"
<infinisil>
samueldr: Basically a module as we know today, so something you can put into `imports = [ ... ]`
<samueldr>
so you would define a dependency on the path of said module?
<infinisil>
The options tree is irrelevant in this, modules can still add them anywhere they want, but they can only define and use options from the dependencies they declare
<infinisil>
samueldr: I'm not entirely sure how it would work, but something like that yes
<samueldr>
my thinking is how you _identify_ a module, more than define, I think
<samueldr>
and I was talking about the options tree since that's not a good way to identify a module :)
<infinisil>
I'm hoping for a syntax like `_module.dependsOn = [ "base" "nginx" ]`
<infinisil>
And defining a modules name like `_module.name = "foo"`
<infinisil>
(but that won't work because then we'd have to evaluate the file to figure out its name)
<infinisil>
Maybe a central file mapping names -> files
<infinisil>
With this in place, this could potentially be extended to limit what modules can do. E.g. if you have `{ _module.canSet = [ "base" "nginx" ]; imports = [ ./bad/module.nix ]; }` in your configuration.nix, it could limit what bad/module.nix can set (recursively)
<infinisil>
gchristensen: Is that an ofborg check now?
<gchristensen>
not an ofborg check
<gchristensen>
this is metrics coming out of the tool used to do the syncing
<gchristensen>
(reload first) if you click the (i) on the top left of the "invalid maintainer.nix file" graph there is a tool tip about what that means
<infinisil>
I see, I'd think just a check in the file itself would be better still though
<gchristensen>
I agree
<gchristensen>
but this isn't instead of that
<gchristensen>
these metrics out of this process is trying to answer questions like "when did it run last and why isn't it working"