Re: [PATCH v2 0/6] builtin/remote: rework how remote refs get renamed

[Junio C Hamano <gitster@pobox.com>]

Patrick Steinhardt <ps@pks.im> writes:

> There shouldn't be any textual conflicts between these two series.

What I meant was this.  This series is built on top of an older
iteration of the other series.  The other series however has a newer
iteration.  We would eventually want to both topics in the system,
so as an early preview, both would be merged to 'seen'.

The topic branch for the other series has patches from iteration vN+1.

The topic branch for this series is, since it is built on top of the
merge of the other series at iteration vN into 'master'.

We merge the former into 'seen'; we now have patches from the other
series at iteration vN+1.  We then merge the latter into 'seen'.  It
wants to _also_ merge the patches from the other series at iteration
vN, that duplicates vN+1 but in different ways.  If there wouldn't
be any textual conflicts between vN and vN+1 of the other series, it
may resolve cleanly, but is the result sane?  These two iterations
are trying to be moral equivalents, with the difference that the
newer iteration is trying to be better than the older one.

And they in practice are most likely to textually conflict with each
other.  After all they are different iteration of the same topic.