Re: summarizing patches for review (was: Gets rid of length in the docs. (issue 4965053))

[Graham Percival]

On Tue, Aug 30, 2011 at 11:25:27PM +0100, Trevor Daniels wrote:
> 
> Graham, you wrote Tuesday, August 30, 2011 2:43 AM
> 
> >LARGE PATCHES
> >
> >SHORT PATCHES
> 
> I think just CODE PATCHES for these.  It's
> hard to think of a meaningful difference.
> Number of affected files, number of changed
> lines, lines of changes C++ and/or scm?

Combination of all the above?  I don't want to have any criteria
that would take a non-programmer more than 10 seconds to decide
which a patch falls into.  But I'm fine with just CODE PATCHES.

> Or maybe split into ENHANCEMENTS and BUG FIXES.
> Enhancements generally will require more
> thought and discussion.

Will it be obvious to a non-programmer (i.e the patch meister) to
decide which is which?  OTOH, maybe we should just require that
any patch clearly states which it is, and immediately reject any
patch whose commit message wasn't clear enough for a
non-programmer.  I'm not too fond of that last idea, though -- I
don't want to burden really technically-skilled programmers with
lots of red tape.  We want to encourage those people to be
producing patches, not discouraging them with extra non-technical
work.

Cheers,
- Graham