🚀 go-pugleaf

Karl

Thanks for the patch to filelock.c.  This has worked very well for me.
I have seen the discussions about this and from my perspective support
the decision to place the lock file in the same directory for the
following reasons.

- We have a large network with a few different types of machines on
	them.  Depending on where you are the same file can have many
	different names.  eg:

	/dme/user/jaws/.bashrc (Solaris)
	/tmp_mnt/dme/user/jaws/.bashrc (SunOS)
	/export4/cscf/jaws/.bashrc (The SunOS machine the file is 
					physically mounted on)
	
	There are ways of resolving these but they get messy and take
	time.  As our network has been around for a while, and there
	have been changes in thinking, we also have the problem that we
	have the same directory mounted under two different names.  I
	realise that this indicates a bit of a messy network and there
	should be no real need to program for this, but we have left
	some of the old names lying around so because removing it would
	break too many things.

- The main reason I took an interest in emacs File Interlocking is that
	our section is dominated by scum of the earth vi users (Funny,
	they call me a parasite emacs user!).  This alternative scheme
	made it a lot easier to write a wrapper script for vi to use
	the same locking as emacs.  It also means that if there is a
	stale lock file lying around it is a lot easier for people to
	find them and blow them away.

I can see the problem that not having write permission in the directory
posses.  Luckily this doesn't affect our operation.  From my point of
veiw it is nice to have a system that works for most situations rather
than one that doesn't work at all on a network.

Just my two pennies worth
James