On Tue, 2006-11-21 at 11:31 -0600, Rex Dieter wrote:
Ralf Corsepius wrote:
> On Tue, 2006-11-21 at 09:29 -0600, Rex Dieter wrote:
>> I consider ABI compatibility as just one part
>> of what defines a stable distro, but, imo, there are certainly cases
>> where breaking ABI is justified (for essential features, bug fixes, and
>> yes, stability sometimes).
>
> Please ask RH how they have been handling Core, so far.
>
> I don't know how many times I've been told: "No API-changes, no ABI
> upgrade, no feature upgrades, often not even bugfixes (aka
> FIXEDRAWHIDE)"!
When it comes to breaking API/ABI, I'd say it's primarily the package
maintainers' call to make.
I am inclined to agree in those cases, where a package is of limited
importance, has a very limited number of dependencies and/or a small
userbase, but I can't avoid to disagree in general.
But what would you think of a kernel, GCC, Glibc, Gtk/Gtk or Qt/KDE
maintainer, who breaks things midst of a distro's life time?
Have a look at FE: A classic breakdown is maintainers not paying
attention to SONAMEs/ABIs/APIs and them inadvertently breaking something
by not so.
Most maintainers, after having gone through a learning curse, will try
to circumvent such issues, either by providing compat-packages, by
trying to inform their users in advance, or ... to resort to refraining
from their plans.
Ralf