Re: [MENTORS] Potential issue with GitHub releases

Hi!

apologies for chiming in late to the thread, but it was my impression
that GH releases
were something akin to some sort of convenience binaries area.
Wouldn't a note stating
so + some links for every release would be enough?

Would anyone mind some link pointing to the appropiate resources on
why they shouldn't
be used? the link provided by Jarek yields a 404 for me, even after
login in as asf committer

(benefits of keeping suscribed to incubator, you always learn new things :-))

best regards,
juan pablo

On Sat, May 2, 2026 at 9:59 PM Jarek Potiuk <[email protected]> wrote:
>
> I started a discussion in Airflow and I hope we will just remove them and
> stop releasing them [1].
>
> We publish some releases on GitHub, but, to be perfectly honest, I never
> saw value in them and that nagging feeling always bothered me a bit :).
> This thread made me realise they might be actively harmful.
>
> So ... it happens even for well-established projects.
>
> Thanks for raising the subject, Justin, PJ, Sebb, Dave, Piotr,
>
> Sometimes you just do not know why something is nagging you, and then
> suddenly someone explains it clearly... and then you know.
> I agree there is absolutely no reason to publish GH releases and it's
> actively harmful. We should all stop doing it.
>
> Maybe we should have an automated check run by someone (infra/tooling?)
>
> It always also bothers me when you say "things should never happen" but
> fail to take action to ensure it does not happen and ask people to fix it.
> That sounds a bit like "I have strong opinion, but I know it does and I
> can't do anything about - so Iit am closing a blind eye on it",
>
> Is there a way we would like to enforce it? Who can do it? Or do we not
> care that some projects do it?
>
> [1] https://lists.apache.org/thread/wc49rx24j09r87mgs2t6vpf1rtbtr6gf
>
> On Sat, May 2, 2026 at 8:33 PM Piotr Gankiewicz <[email protected]> wrote:
>
> > Thank you for the info, I've updated all releases under ASF with the
> > missing incubating disclaimer.
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [email protected]
> > For additional commands, e-mail: [email protected]
> >
> >