Since Leo files are so flexible, they can be considered as many
things, I towards seeing them as collections of data.
A node is a key:value pair, headline and body.
A subtree is a graph of key value pairs, a Leo file is
a graph of graphs, usually on a topic.

key: value examples:

Classname: code graph
Methodname: method code
@auto: disk file
@command: leo script
FAQ: answer to the question
...

I currently have over 80 Leo files on my hard drive:
Lets call it surfeit of curiosity instead of deficit of attention.
I would be interested in a store of all the nodes in all the files,
allowing me unified access to their contents.

I have put lots of notes, hints, reminders etc. in Leo I'd like to
be able to marshal my thoughts on a topic, across the span
of those Leo files. I'd like to identify duplication and popularity.

I currently have _many_ places to stash points of interest, reminders,
tips, facts, answers, ideas, etc.

I would like to consolidate to Leo, but that means breaking out of the
file-wise restriction. It requires a 'base' or 'store' for the data.

The file-wise aspect is great for authoring the data, but doesn't work
for analysis and retrieval, that's what databases are for.

Thanks,
Kent

On Sat, Mar 30, 2013 at 4:56 PM, Edward K. Ream <[email protected]> wrote:
> We've gotten a bit far afield with recent discussions of DB's.  I'm
> responsible for the digression.
>
> Obviously DB's are useful, but for whatever reasons, I just am not
> interested in them.  Consider it a failing of mine, if you like. Having said
> that, I wouldn't mind Kent or Terry showing me how they are beautiful.
> That's probably a conversation best left for happy hour :-)
>
> So let's refocus on the future of Leo.  What problems with Leo (or .leo
> files) would DB's be likely to solve?  This would be a good topic to discuss
> at the sprint.  Your comments, please.
>
> Edward
>
> P. S.  I would answer this question as follows.  Imo, the biggest problem
> with both Leo and .leo files is that the rest of the world isn't attuned to
> the Leo way.  What can we do, here and now, to ameliorate that situation,
> without waiting for others to "get" Leo?
>
> For example, bzr understands neither .leo files nor Leo sentinels in
> external files.  As I use Leo, bzr commits don't always involve .leo files,
> but they almost always do involve external files.  So it seems to me that a
> useful tool would be to represent a bzr diff as (yet another) Leo outline.
> Leo's present conflict nodes are an example of this idea, as is Leo's
> compare-outlines command.  Generating outline-oriented diffs is actually a
> *very* easy problem, because gnx's allow fast and foolproof determination of
> which nodes have been inserted, deleted and changed.  It is then trivial to
> use difflib to report diffs in changed nodes.
>
> EKR
>
> --
> 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?hl=en.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

-- 
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?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to