[Pkg-gnupg-maint] Bug#760273: Bug#760273: gnupg.info: typo: Ceation -> Creation

[Thijs Kinkhorst]

On Wed, September 3, 2014 17:21, Daniel Kahn Gillmor wrote:
> On 09/03/2014 04:13 AM, Thijs Kinkhorst wrote:
>> On Tue, September 2, 2014 15:49, Daniel Kahn Gillmor wrote:
>>> Thanks, i've sent a patch upstream for this, and will drop a patch in
>>> the debian packaging to fix it in the meantime.
>>
>> For GnuPG 1 I've tried to maintain a "minimal patches" policy. Patches
>> have a cost, is my experience in a package that I got which had about 50
>> patches for bugs of wide variety, indeed, quite a few also for such
>> documentation typos as above.
>>
>> Given that upstream is normally very responsive and receptive to fix
>> such
>> issues, I don't see much need to patch non-critical issues in Debian
>> locally. This issue in particular should of course be fixed upstream but
>> I
>> think it has near zero impact on the usability of GnuPG in Debian, so it
>> can easily wait to be fixed when we import the next upstream, no?
>
> I think we've got different instincts here; i'm willing to do things the
> way you describe if you feel strongly about it, but let me briefly
> explain why i would tend to lean the other way:
>
> For patches that are likely to be applied upstream as-is (or have
> already been applied but not released), i don't think i see the
> disadvantages of including them in the debian packaging while the issue
> remains unfixed upstream -- when a new version is released with the fix,
> we can just delete them from debian/patches/
>
> This also provides a way to keep the open bug count down (which makes it
> easier to see what issues are truly unresolved), and makes tracking the
> bugs eventually fixed upstream simpler: We don't need to comb through
> the list of outstanding bugs at each new upstream release and figure out
> which ones were actually fixed -- we can just delete the patches that no
> longer apply and move on.
>
> Anyway, i'm fine going with your preferred approach if you feel
> strongly, and i definitely don't want to carry any heavy divergence from
> upstream.
>
> What do you think?

I'm not strongly opposed, but the case cited really doesn't seem worth it:
one missing letter in a info page.

For real bugs, I'm more inclined indeed apply the fix in Debian if the
chance is indeed high that upstream will accept the fix. Not so much if
upstream is unlikely to accept the fix.


Cheers,
Thijs