On Thursday, February 13, 2020 at 10:58:01 PM UTC-5, andyjim wrote:
>
> A node for each zettel? Thousands of them?
>

Well, yes, that's what I've been been envisioning.  Having thousands of 
them - whether they are represented as Leo nodes or some other way  - will 
require us to be smart at coming up with ways to work with them, and to 
find interesting related ones as well as the ones we've been working on 
recently.  That would be true for the old paper system, too.  With a 
computerized system, at least searches will be much faster.

My experience with my bookmark management system makes me think that the 
zettels should be collected under headings that relate to the reason the 
zettel was written in the first place.  The headings can be informal and 
changeable, and I found there is very little cognitive overhead in "filing" 
under one or another heading.  In Leo, the headings could be implemented by 
creating organizer nodes.  This does not commit you to a particular view of 
the world, but only gives you some mnemonic help in finding related 
thoughts.

Leo itself has something like 300 or 400 files of code, most of them 
contained in one Leo file.  Each code file would be equivalent to maybe 
dozens if not hundreds of individual "notes", if you wanted to think about 
them that way.  That might in total be several tens of thousands - in the 
rough ballpark of the number of zettels we are imagining . It's not too 
unwieldy to work with them.

Come to think of it, my bookmark management system, which I've mentioned 
before, has some 25,000 web pages, with more than a hundred top-level 
headings, not including all the subheadings.  The cross-links are all 
between headings, though, not between individual pages.  So the size and 
complexity of the bookmark collection is also in the ballpark of what we 
are looking at for the zettelkasten.

What the Leo code doesn't have is the extensive system of cross links that 
we expect to have in our zettelkasten.  That will be one of the key things 
we have to work out - not how to accomplish linking technically, but how to 
make it easy to work with. Just to make this seem a little more real, one 
thing you can do with Leo right now, without adding anything new, is this:  
you can "mark" any number of nodes, and then activate a single Leo command 
that makes clones of all of them and puts them under a single organizing 
node. So you could do a search, and then scan through the search results, 
mark the ones you want, and automatically collect clones of all of them in 
one place. I see this capability as a way to collect various thoughts into 
a project to work on.

In addition to finding the notes to work on together, another challenging 
user interface issue will be to display a number of them at once, and in a 
readable (and maybe editable) way.  That's never been easy on a computer. 
Leo does have the ability to open multiple editors, but the way that works 
won't, I think, be quite right for what we need to do.  I've read that with 
the old paper system, people used to physically remove a number of zettels 
from their filing case (the *kasten*) and spread them out on a desk. When 
they were done, they'd have to get them all (with any new ones) back in the 
right place or they would never be able to find them again.  We can do 
better than that, anyway.

One approach that I think will be useful in finding zettels related to the 
ones you are thinking about is sometimes called a *Concise Bounded 
Description*.  Basically, given a node, this collects all references of a 
chosen depth of links in one subset.  See 

https://www.w3.org/Submission/CBD/

I think this is likely to lead to the other zettels one is most interested 
in. I'm fairly sure that a CBD won't be hard to implement.  I've done it 
before.   One way to display them would be be with a mind map - I'm very 
fond of mind maps, myself.  Making a mind map that can interact with the 
rest of our zettelkasten system would be challenging.  So a read-only mind 
map would be what we could expect for quite a while.

Well, there I go getting into details of the user interface when we are 
still working on the requirements at a higher level.  Fun and useful, but 
we shouldn't get carried away too soon!

-- 
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 view this discussion on the web visit 
https://groups.google.com/d/msgid/leo-editor/bf5b429e-049b-45db-8693-3f2127b4561b%40googlegroups.com.

Reply via email to