Bug#1031325: e2fsprogs 1.47.0 introduces a breaking change into Bookworm, breaking grub and making installations of Ubuntu and Debian releases via debootstrap impossible

[Sam Hartman <hartmans@debian.org>]

>>>>> "Theodore" == Theodore Ts'o <tytso@mit.edu> writes:

    Theodore> On Wed, Feb 15, 2023 at 01:17:38PM -0700, Sam Hartman wrote:
    >> 
    >> I.E. I think your question of "for how long" has a very simple
    >> answer based on our history: if we care about stability in this
    >> instance it's for +/-1 Debian release.
    >> 
    >> I'm struggling trying to figure out whether we should commit to
    >> that stability.

    Theodore> I recogniuze that there are precedents that go in both
    Theodore> directions.  We have *never* required that shared library
    Theodore> linkages created in Debian N work in Debian N-1.  Sure,
    Theodore> there are workarounds that you can use (e.g., compiling
    Theodore> with -static), but I've listed workarounds for mke2fs as
    Theodore> well.

For what it's worth, I don't think the shared library situation is at
all analogous.
We've basically decided that we care about shared libraries as they
interact with packages, and we've invented a whole bunch of dependency
logic to deal with them.
Which is to say we've explicitly turned shared libraries into a special
case.

You argue about shared libraries for non-packaged binaries.
I think we mostly don't care about that, and again, I think that's at
least a generally recognized thing that came out of our focus on
packages and package dependencies.

Which is to say that I think shared libraries are such a special case in
Debian you cannot use them to argue for or against anything else.

You make some good arguments based on other things.  I just don't want
us using shared library handling as a precedent for anything other than
shared libraries, so I am arguing against it.