<samueldr>
(though I am applying your suggestions)
<gchristensen>
I am concerned that might become a hard blocker
<gchristensen>
but we will see
<samueldr>
I still think it is a usability issue not to list them in some way, considering some people depend on unfree packages, but I agree in not advertising them, and also in neutral-to-negative phrasing about those
<gchristensen>
hmmm
<gchristensen>
samueldr: could you write up a paragraph or two about that?
<samueldr>
gchristensen: paragraphs like in the inital PR body text?
<samueldr>
(does anyone read those?) :)
<gchristensen>
no, unlike your paragraphs
<gchristensen>
what you said is that you can remove the feature, and that you're in favor of showing unfree software. your comment about it being a usability issue, and many people depending upon unfree packages (like the foundation does) goes unsaid, but is, I think, the most important part
<samueldr>
right
<gchristensen>
also I think it is good that you clearly say your personal perspective on software freedoms
<gchristensen>
in the hopes that it helps go from "this is someone who has no moral position and will go at this from whatever angle" to "this person has a moral position on software freedoms which generally agree with mine, and is proposing reasonable arguments in favor"
<__monty__>
Has there been discussion about this before? Pretending unfree doesn't exist doesn't seem like a helpful stance to me. I for one, and I think most, don't like free software just because it's free but because it's better (and being free "libre" is definitely part of why). But I'm still pragmatic, sometimes you just need (or want, that's equally valid) some unfree software, tools shouldn't judge you
<__monty__>
for that. I'd much rather people use some foss because it's just plain better than unfree alternatives than have them begrudgingly use foss in spite because it's hard not to.
<gchristensen>
lots of discussion
cjpbirkbeck has joined #nixos-dev
orivej has quit [Ping timeout: 272 seconds]
pie_ has quit [Ping timeout: 252 seconds]
ixxie has joined #nixos-dev
orivej has joined #nixos-dev
orivej has quit [Ping timeout: 272 seconds]
drakonis has joined #nixos-dev
orivej has joined #nixos-dev
Synthetica has quit [Quit: Connection closed for inactivity]
tilpner has quit [Quit: WeeChat 2.4]
drakonis has quit [Read error: Connection reset by peer]
<jtojnar>
worldofpeace: so should I use `if withQt then mkDerivation else stdenv.mkDerivation` or always `mkDerivation` and add `wrapQtAppHooks` to inputs?
<worldofpeace>
jtojnar: The package should be called with `libsForQt5.callPackage` and use `stdenv.mkDerivation`. It should then conditonally add `wrapQtAppsHook` for withQt etc.
<jtojnar>
worldofpeace: done
<jtojnar>
hmm, pushed to a wrong branch
<jtojnar>
btw apparently 3.34 will bring the systemd sessions in gnome-session
<emily>
there's no downside to using the qt5 mkDerivation if you set dontWrapQtApps, right?
<worldofpeace>
jtojnar: yeah I read that, it still has compat without the systemd sessions?
<jtojnar>
emily: dontWrapQtApps only affects the runtime closure, qt deps are still pulled in for build
<emily>
even if you only conditionally add them to the build inputs?
<jtojnar>
emily: the hook adds them unconditionally
<emily>
ah :/
<emily>
guess I should fix nextpnr
<worldofpeace>
I'm particularlly nervous about when elementary's going to decide to finish their port to 3.34 jtojnar
<{^_^}>
elementary/gala#566 (by tintou, 8 weeks ago, open): Adapt to mutter 3.33
<jtojnar>
worldofpeace: the comments I read are were not very conclusive
<jtojnar>
worldofpeace: what is their release cycle, do they still follow LTS Ubuntu?
<worldofpeace>
jtojnar: they make a major release for the next lts, though they did do a point release for 0.4. They pretty much say they keep everything rolling
<worldofpeace>
In the sense that when Ubuntu LTS is really near they actually start a feature freeze
<worldofpeace>
So currently we're seeing releases for every new feature and they try to maintain backwards compat in most instances