Two questions about build-path reproducibility in Debian

[Chris Lamb]

James Addison wrote:

> None of the remaining thirty-or-so (and in fact, none of the 66 updated so far)
> are usertagged both 'buildpath' and 'toolchain'.
>
> I would say that a few of them _are_ 'toolchain packages' -- mono, binutils-dev
> and a few others -- but for these bugs the buildpath issues are internal to
> each package at build-time and do not affect the construction of other
> packages in their ecosystem.

You are absolutely right to distinguish between a package that is
itself unreproducible and a package that is causing other packages
to be unreproducible. These are very much orthogonal concepts as you
imply, and a package can certainly be in both categories at once.

What might be confusing to folks is that our "toolchain" usertag in
the Debian BTS does not refer to a toolchain *package* in the usual,
Debian sense, i.e. Mono, libc, Bison, documentation generators and
so on. But rather that (loosely speaking) "if this usertag is applied
to a bug, its denoting that that particular *bug* is affecting the
reproducibility of other packages."

Unfortunately, the tag is actually an excellent example of that
general trend in tech where something was badly named in the spur of
the moment, and then the name just sticks around forever due to some
combination of muscle memory, inertia and, frankly, priority: as in,
this metadata is not *all* that visible nor A++ important to begin
with… outside of threads like this. :)


Best wishes,

-- 
      o
    ⬋   ⬊      Chris Lamb
   o     o     reproducible-builds.org 💠
    ⬊   ⬋
      o