I'll be taking a break for the next day or so, paying bills and doing other 
chores that have accumulated while I have been working so intensely on this 
project.

Here, I'll summarize my overall thoughts about what has been done, and what 
remains to do.

1. This may be the most important work in Leo's history.  It removes the 
fundamental barrier (sentinel comments) that up until now have prevented 
people from using Leo in cooperative situations.  It's impossible to know 
how many people might give Leo a new look as a result.  It is the solution 
to a problem that has vexed me for at least 20 years.  It still seems like 
magic.  Well, it *is* magic.

2. @auto-view nodes themselves, packaged in .txt files, say @file 
at_auto_views.txt, provide a way for Leo users to share outline structure 
without involving non-Leonine users in any way.  This is one example of why 
@file will never go away.  unitTest.leo also requires @file, imo.

3. None of this would have happened without recent discussions among the 
usual suspects.  Zoom.Quiet got me thinking, once again, about 
cooperation.  Terry, Kent et. al. *finally* convinced me that people would 
(justifiably!) accept *no* changes to their source files.  Terry's crucial 
remark that bookmarks will not usually break in practice thus came in 
prepared environment.

4. Terry asked (in the #leo irc) whether it would it not be safer, at least 
initially, to use a different directive (say @new-auto) and leave the 
existing @auto behavior in place.  Yes, it would be possible, but there is 
a simpler way: just revert to the initial imported tree if the "before" and 
"after" trial writes don't match. This sacrifices all the data in the 
@auto-view tree so as to preserve the external file.  This should be plenty 
safe enough.

I expect that the code will quicky become rock solid, making this question 
moot. We really can reasonably expect *exact* matches in all trial writes. 
Indeed, most mismatches will arise from incorrectly moving an imported 
(non-organizer) node.  Such mistakes are possible, but they are *far* less 
likely than the myriad potential problems that exist in the 
character-oriented importer code.

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/leo-editor.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to