All this uproar about using Leo without Leo has stimulated some wild
thoughts.

Yesterday I watched Tim Berners-Lee's TED talk:
http://www.ted.com/index.php/talks/tim_berners_lee_on_the_next_web.html

Not convincing.  All these links to "data" are dubious.  What happens
when the data changes?  It's all very well to provide breakable links
when you don't care much about the data's accuracy.  But what if the
accuracy matters?  And what about privacy?

Inspired by such sleazy links, I have been wondering if Leo could live
with breakable clone links.  Suppose we dispense with gnx's, and
simply use some form of headline searching to resolve clone links.
I'm thinking that a headline of the form A (B) would match any
headline A, so that we could put stuff in parens without breaking the
link.

If we could remove gnx's, we could remove most sentinels from external
files.  If we could remove most sentinels, we could, in effect, use
import logic for most external files.

So, for example, there might be little difference between @auto-rst
x.txt and @thin x.txt.

The idea is that sentinels would mostly be hints to the read/import
logic.  If we think "import" instead of "read" most sentinels might
just disappear.  Section references do require sentinel pairs, but
they are rare.

Blue sky stuff.  It might be worth a prototype, though.

Edward

P.S.  I still hate the thought of messing with the leoAtFile read
code.  The write code for @auto-rst should, instead, insure a space at
the end of each node, that is, before the ....@-node sentinel.  That
should be enough, I am guessing.

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