Strawman: eliminating debdiffs

[Bryce Harrington]

On Thu, Oct 09, 2008 at 01:59:25PM +0100, Colin Watson wrote:
>   * Sometimes people send a perfectly good patch in an Ubuntu bug report
>     and then somebody says "in order to get this included you should
>     attach a debdiff instead". This is *a complete and utter waste of
>     time*.

I'd definitely agree here.  If someone identifies or creates a patch,
the packaging work to incorporate it is usually pretty straightforward.
Finding the right patch is usually[*] the lion's share of the work.

But I definitely appreciate it when a contributor posts their patch as a
debdiff, for the simple fact that it shows that they've ensured that 
a) the patch applies, and b) probably gave consideration to any
integrational impacts.

Bryce


* - the most common exception I've found is with SRU proposals for Hardy.
Often a patch to fix an issue is known, but the delta between where that
patch came in upstream and what we ship in hardy is too large to bridge
so easily, and so a debdiff is going to be hugely important in this case.
But of course this case doesn't fulfil the "perfectly good" criteria.