Re: Unnecessarily moving stiff between files considered harmful

[Karl Fogel <kfogel@red-bean.com>]

Eli Zaretskii <eliz@gnu.org> writes:
>Btw, if I'm allowed a gripe: we have too much of this in our history
>already.  Just recently I wanted to know when the assertion in
>set_blv_found was born, and was unable to determine this using Git,
>because that function was first a macro, then an inline function, and
>migrated between 2 or 3 files several times.  I eventually had to use
>ChangeLog files from old Emacs releases(!) to find out the history of
>this single line.  I'm sure Git gurus will come up with some
>convoluted way to do that with Git, and Magit gurus will tell that
>Magit already knows how to do this.  But the fact that the usual tools
>of trade fail here is for me a clear indication that we should keep
>this disadvantage in mind when making such changes: there's a non-zero
>price here.

I looked at places in CONTRIBUTE where it might work to add a sentence or two about this, but didn't see any good places to fit it in, hmm.

I agree, there is a price.  (In the case of particular change we were discussing, I thought the price was worth paying, but there are always going to be edge-case disagreements about the relative value of moving code vs the negative effects of the churn.)