Re: Advantages/Disadvantages of Open Source Software (Was Re: Package grub-xen-host breaks PV domains with 11.5 point release)
On 9/15/22 11:45 AM, Maude Summerside wrote:
>
> On 2022-09-14 23:23, Chuck Zmudzinski wrote:
> > On 9/14/2022 11:01 PM, Maude Summerside wrote:
> >>
> >> On 2022-09-14 21:45, Michael Stone wrote:
> >>> On Wed, Sep 14, 2022 at 11:16:00PM +0100, Steve McIntyre wrote:
> >>>> I'll be brutally honest: being accused of "possibly malicious"
> >>>> unwilligness is *not* a great way to convince overstretched volunteers
> >>>> to spend their time on issues.
> >>>
> >>> Especially when it's an ongoing pattern of discourse.
> >>>
> >>
> >> I think there's many barrier that discourage people from wanting to
> >> contribute to many project. I feel some developer use the community as
> >> unpaid beta tester but don't go further into accepting contribution.
> >>
> >> For sure, having managed some project, I have to say that it's hard to
> >> accept contribution that will add new functions to as software when
> >> these come from a unknown contributor. Not because of being scared of
> >> malicious intent (unless the person is really paranoid but that's
> >> another story). Simply because adding a new function means having to
> >> support it's ongoing development and there's no guarantee that the
> >> contributor will do so. Same goes on for code contributed that needs
> >> refactoring, that are badly documented, etc. But all this need some good
> >> social behavior from the project owner/manager.
> >
> > As a user of the Debian software and a user of the BTS, I am discouraged not
> > because new contributions or functions are being rejected, but because bugs
> > are not being fixed. Those are two very different things. Maybe it's just too hard
> > for volunteers to fix the bugs and make Debian better, and maybe we need to
> > pay the volunteers so they are not volunteers anymore and will be motivated
> > to actually fix the Debian software. The fact that Debian is created by volunteers
> > is probably one of the really big disadvantages of Debian software.
> >
> I think there's a piece missing hugely in *your* equation.
> The package maintainer are the LAST line of resort when there's a bug to
> fix. Sure you can report them thru BTS but they'll transmit those
> upstream to the original software developer.
Not in my experience. Most upstream projects say users should report
bugs to the distro first and let the distro's maintainers decide what to
do. The bugs I see that the Debian maintainer *should* forward to the
upstream project usually fail to do that. Of two cases of bugs affecting
my machine this past couple of years, one I reported the bug to Debian,
Debian's maintainer ignored it, I found the fix after a long bisecting process
and the fix was in the upstream part of the code. So I tagged the
bug with patch and upstream and waited for the maintainer to forward
the bug. The maintainer again ignored it so I had the opportunity to
make a contribution to an upstream project and I submitted the patch
to the upstream project myself and when it was committed upstream
I tagged the bug fixed upstream on BTS and now the bug is closed.
That is a happy ending to a bug report. The other one this year both
Debian and the upstream project, the Linux kernel, are ignoring the
bug and that is the one I described in a post earlier today to this list
when I also asked the community a question about systemd, udev,
and the coldplug all devices stage of boot where the bug happens. This
bug is still not a happy ending, at least for those who want the bug fixed.
I am not the one who reported it. I would not be surprised if the one
who reported it gave up on it and switched to Fedora or another distro
that has fixed the bug in their distro. It is the kind of bug that can be
fixed in *either* the Linux kernel upstream code or in the systemd/udev
configuration by the distro. But Debian maintainers are just volunteers
so they cannot fix it. At least that is what everyone here is telling me.
>
> What would happened if every bug was fixed by the Debian maintainer ?
> We'd end up having two different source code because at every bug fix
> there would be a different tree of source code being built.
Most users are not able to determine when they report a bug if
it is in the upstream or Debian part. I learned how to find where
the bug is because no one else in the free software world would do
it. You advocate for a world where every user can fix their own
bugs and the maintainers can complain they can't fix bugs because
they are just volunteers. That doesn't make sense to me. The BTS
is useful because users do post workarounds for the bugs that
the maintainers don't fix, but users are mistaken if they think
when they report a bug the maintainer will see to it that it
will get fixed. I also think the bot that says the maintainer
will respond to you in due course sometimes lies because in
some cases the maintainer never responds.
Best regards,
Chuck
Reply to: