jtojnar has quit [Ping timeout: 240 seconds]
hpfr has quit [Ping timeout: 240 seconds]
mkg20001 has quit [Ping timeout: 244 seconds]
kevincox[m]1 has quit [Ping timeout: 260 seconds]
worldofpeace has quit [Ping timeout: 244 seconds]
kevincox[m]1 has joined #nixos-gnome
worldofpeace has joined #nixos-gnome
mkg20001 has joined #nixos-gnome
jtojnar has joined #nixos-gnome
hpfr has joined #nixos-gnome
kevincox[m]1 has quit [*.net *.split]
kevincox[m]1 has joined #nixos-gnome
<mkg20001> the attribute spidermonkey has version 38, yet esr is now at ver 81 - I think this really needs an update
<jtojnar> we probably should not have unversioned attribute in the first place, when the library is not api stable
<jtojnar> yeah rename the uses + move it to aliases.nix for BC
<jtojnar> * yeah rename the uses + move it to aliases.nix for backwards compat
<mkg20001> unsure if we need it, 3 packages depend on it
<mkg20001> in aliases.nix
<mkg20001> seems like literally no one is using that old version anyways
<mkg20001> https://github.com/linuxmint/cjs/issues/80 so much discussion arround cjs and gjs... now they want to apparently port to mozjs68 for "some reason" instead of just straight 78... ugh...
<mkg20001> https://github.com/NixOS/nixpkgs/pull/99008 currently have this PR to get rid of mozjs52 by using a patch to use 78
<worldofpeace> merged cjs upgrade, idk, can we backport it?
<worldofpeace> mkg20001: I commented on the thread that the best option for their projects being used is to use gjs
<mkg20001> <worldofpeace "merged cjs upgrade, idk, can we "> sure, just needs spidermonkey backported - unless that requires any upgrades during backport
<worldofpeace> but I don't know why gnome can't be more sensible! like, we do no specs, we break everything for third parties every release, and broken apis... I felt that comment because pantheon has to get by with that
<mkg20001> reminds me of GCE apis... heard people are switching just because of that... I think it's the "code fast, break things" approach that everybody nowadays seems to be taking. very seldom I see any backwards compat in software, most I get is migration guidelines
<worldofpeace> qt has some pretty good backwards compat IIRC
<worldofpeace> lol, people probably go to plasma for these reasons
<worldofpeace> yeah, I think it might even be a culture problem. like in a really traditional setting it's ur manager "do this and I don't care if it's a garden built around a giant thorn"
<worldofpeace> and then people burnout...
<mkg20001> another thing is continously refactoring for the sake of it.... I usually stop when it's sane. Dunno what causes it exactly in FOSS tho.
<mkg20001> when the API is sane*
<mkg20001> there's a suggestion to use a patch so instead of CJS GJS gets used
<mkg20001> currently upstream is still discussing but we could implement this unless there's anything that CJS does that GJS doesn't
<worldofpeace> I saw a "pandora's box" comment so...
<worldofpeace> <mkg20001 "another thing is continously ref"> some people value correctness very much (and a changing sense of correctness) but in the end if they all do the same thing why just waste ur time.