Greg A. Woods wrote :
|| [ On Tuesday, February 15, 2000 at 16:17:32 (-0500), John Macdonald wrote: ]
|| > Subject: Re: CVS File Locking
|| >
|| > Detecting that a goof has happened, while useful, is far less
|| > valuable than preventing it in the first place.
||
|| People have tried to make the same claim about the copy-edit-merge
|| paradigm too. Fortunately they've been unable to prove their claim.
That's not all of it though. Staring with the Berliner report, there
has been proof that copy-edit-merge has (1) many operations that have
no conflict at all, (2) the vast majority of conflicts can be
automatically resolved with essentially no error, (3) the vast
majority of conflicts that cannot be resolved automatically can be
resolved easily manually. The remainder (4) is so much smaller than
the potential for gain from concurrent activites, that the value is
quite evident.
With unmergable files, you have case (1), no case (2) and a far more
equal split between (3) and (4). (The manual resolution is easy only
if you can pick one or the other - usually you have to duplicate the
effect of one of the changes over top of the other.)
There is a real difference here.
--
Anyone who can't laugh at himself is not | John Macdonald
taking life seriously enough -- Larry Wall | [EMAIL PROTECTED]
