Policy on referencing/abbreviating git commit-IDs, going forward? (was: Can anyone correct the Bazaar reference "revno:111954.1.97"?)

[Joshua Judson Rosen]

Andreas Schwab <address@hidden> writes:
>
> Eli Zaretskii <address@hidden> writes:
>
> > This is bzr, not git, so the diffs aren't the whole story.
>
> There is no difference between git and bzr wrt. to the DAG.  Only the
> names they use are different.

There is an interesting `bzr vs. git' question here, though--because of
the differences in the way revisions are named: when you run into the
analogous "I can't tell what this reference is to" problem in git, how
will you deal with it? When someone writes a abbreviated git sha1sum
into a changelog / commit message because just the first 7 or
howevermany hexits was enough to unambiguously identify the referenced
commit-object in their development environment at the time when they
were committing their change, and then a later commit/merge makes that
abbreviation ambiguous, how will you trace out the now-ambiguously-
abbreviated commit? e.g.: up through July 29 2012, "7a62fbf"
unambiguously identified a commit from October 24 2007; on July 30 2012,
"7a62fbf" stopped being unambiguous.


-- 
"'tis an ill wind that blows no minds."