Re: enable-clash-detection
>>>> "Jan" == Jan Vroonhof
<vroonhof(a)math.ethz.ch> writes:
Jan> Hrvoje Niksic <hniksic(a)srce.hr> writes:
> > Clash detection is an incredible fucking drag in an
environment such
> > as the one here. (...) It's really hard to explain to the spoiled
> > bastard that has a 4 CPU ultra that he has to wait for an NFS server
> > to listen to him before he can to a find-file on a local disk path!
> >
> > Sorry, but I feel pretty strongly about this one.
>
> OK, what you say makes sense. I say we keep it disabled by default,
> and roll Jan's patch into 21.1.
Jan> Please guys, could you read the patch again. The new clash detection
Jan> does NOT use a global locking directory. It makes a symlink in the
Jan> same directory as the file is.
Those interested in this should read
FSF Emacs 20.2 Info: (emacs) Interlocking
as I just did.
I did not realize that the global locking directory has been
eliminated. This is GOOD. Nevertheless, we should still be cautious
about enabling this by default. In particular, this paragraph from
the info file makes me worry:
-----------------------------------------------------------------------
If Emacs or the operating system crashes, this may leave behind lock
files which are stale. So you may occasionally get warnings about
spurious collisions. When you determine that the collision is spurious,
just use `p' to tell Emacs to go ahead anyway.
-----------------------------------------------------------------------d
I vote for experimentally turning clash-detection back on by default,
with the new FSF-Emacs-compatible implementation, early in the 21.1
development cycle.
Martin