I wish projects didn't use old distro releases as their primary development and gating target. It makes it really hard to advance software...
Conversation
Notices
-
Embed this notice
Neal Gompa (ニール・ゴンパ) :fedora: (conan_kudo@fosstodon.org)'s status on Wednesday, 26-Jun-2024 21:48:49 JST Neal Gompa (ニール・ゴンパ) :fedora:
- Haelwenn /элвэн/ :triskell: likes this.
-
Embed this notice
Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Wednesday, 26-Jun-2024 21:49:46 JST Haelwenn /элвэн/ :triskell:
@Conan_Kudo At the same time given how painful it is to get packaged in Debian, maintaining at least some compatibility to it makes sense. -
Embed this notice
Neal Gompa (ニール・ゴンパ) :fedora: (conan_kudo@fosstodon.org)'s status on Thursday, 27-Jun-2024 00:01:57 JST Neal Gompa (ニール・ゴンパ) :fedora:
@javierm @matt what's worse is that sometimes I can't get fixes and improvements to land because they are not testable on really old platforms.
Haelwenn /элвэн/ :triskell: likes this. -
Embed this notice
Javier Martinez C. (javierm@fosstodon.org)'s status on Thursday, 27-Jun-2024 00:02:00 JST Javier Martinez C.
@matt @Conan_Kudo it makes life of developers from distros that move faster harder, because issues that happen with newer toolchains / libraries are not found until distros attempt to package the latest releases.
-
Embed this notice
Matt Campbell (matt@toot.cafe)'s status on Thursday, 27-Jun-2024 00:02:01 JST Matt Campbell
@Conan_Kudo What kinds of advances do you think are being held back because projects develop and test against old distro releases?
-
Embed this notice
Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Thursday, 27-Jun-2024 00:04:38 JST Haelwenn /элвэн/ :triskell:
@javierm @matt @Conan_Kudo Well latest versions ought to also be tested against, specially given how easy it got to have CI for the tests and containers for development/debugging. -
Embed this notice
Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Friday, 28-Jun-2024 00:43:16 JST Haelwenn /элвэн/ :triskell:
@matt @Conan_Kudo @javierm Well I'd agree that in a lot of cases we need less churn (specially outside of system development) but there's also a lot of developers who just ignore warnings which have been there for years.
And I'd argue that we have things like containers for preserving stale software. -
Embed this notice
Neal Gompa (ニール・ゴンパ) :fedora: (conan_kudo@fosstodon.org)'s status on Friday, 28-Jun-2024 00:43:17 JST Neal Gompa (ニール・ゴンパ) :fedora:
@lanodan @javierm @matt Oh alas, CI against recent platforms is not a thing for a lot of projects. Either because GitHub doesn't offer it or because the platform the project runs doesn't do it.
-
Embed this notice
Matt Campbell (matt@toot.cafe)'s status on Friday, 28-Jun-2024 00:43:17 JST Matt Campbell
@Conan_Kudo @lanodan @javierm It sounds to me like it would help if platforms and toolchains would slow down and stop breaking things for downstream developers who don't test against the bleeding edge. Of course, I'm aware that some things do need to evolve, and volunteer developers shouldn't be expected to go through the heroics that Microsoft historically did to preserve backward compatibility in Windows. Still, it feels like there's too much churn.