On Sat, Jan 14, 2012 at 6:35 AM, Edward K. Ream <[email protected]> wrote:

> On second thought, @shadow should work more like @auto than @file in
> this regard.  In other words, @shadow should not create *recovered*
> (not revered) nodes.  I plan to do this today.

Oops.  On third thought, I think creating "recovery" nodes for @shadow
files *is* the right thing to do.

My second thought was confused.  Recovery nodes aren't created for all
changed nodes, only *cloned* changed nodes.  Thus, there is a real
potential for conflict and data loss whenever a recovery node exists.

It seems unwise to ignore any situation that result in the creation of
recovery nodes: all such situations have the potential to alter data
due to the multiple update problem.

Edward

P.S.  I would like to emphasize the following fact about cross file
clones.  They aren't dangerous *if* you only change them within Leo.
All the problems arise because people change cloned data outside of
Leo.  Usually, people will change only *some* of the cloned data.  In
that case, Leo has the unenviable job of guessing what clone contains
the intended data.  In general, Leo can not guess accurately, no
matter what "rules" are put into place.  Thus, creating recovery nodes
is *always* a good thing to do.

What we are seeing is the boundary between reasonable and unreasonable
uses of cross-file clones.  That boundary is not sharp:  whether
cross-file clones work depend on the overall workflow of the people
using them.

EKR

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/leo-editor?hl=en.

Reply via email to