Let me know if I should put this in an attached file instead, and I'll do 
it. Pretty long.

1. No Lockin
agreed

2. Text Files
agreed

3. Simple Section Demarkations
For distraction-free creative/productive activity I'd like to be able to 
optionally see only the text body, or, (and maybe the following belongs 
under Display) in a collection/stack of zettels, or only the first lines, 
or only the titles, or only the links, or only the citations or other 
reference types.

Editor:
In my ideal world I might add a category to this list: Editor. My idealized 
editor displays in rendered text, controlled more or less as a word 
processor does, e.g. Ctl-I for italic (though I don't need and don't want 
the accompanying bells and whistles of a word processor, 'just the rendered 
text please').

This, I think, would significantly enhance creativity because with no 
intervening (and distracting) step I get the immediate benefit of the 
clarity that enhanced formatting gives, plus it eliminates from the display 
all distracting markdown/coding, plus it eliminates me having to think in 
markdown and type in markdown (distractions, distractions).

YET, this wonderful interpretive, self-rendering editor stores the file in 
text/markdown so it's transferable, or forward-compatible (as I have come 
to like to put it). So when you read the file in a plain text editor (and 
my ideal editor optionally displays and operates in plain text/markdown as 
well), the markdown is there. But when rendered in the editor it's in 
rendered mode and you edit directly in rendered mode with word-processor 
style key commands.

Ok, here is a wild thought: Leo has a real time renderer, right? All we 
gotta do then, in my ideal world, is rig key commands and allow the user to 
work directly in the render pane, with the ability to toggle with the 
regular body pane, either overlaid or side by side.
Easy for me to say from my vantage point as a non-programmer. My 
imagination lives in joyous, unfettered freedom by the simple device of not 
knowing what can't be done.

4. Unique Identification
agreed

5. Hierarchical Navigation
The way I see this playing out for me: I will seek a minimal set of 
'highest level' or 'encompassing' headings, the least number under which I 
can class everything. Every zettel must (eventually) class under at least 
one 'top heading' (maybe the default could be 'unclassed' if I don't class 
it). But the acyclic graph principles apply. Even the encompassing headings 
can be anyone's children.

Each zettel, then, maps into the whole shebang by means of:
 . headings attached to it (as parents, children siblings)
 . zettels attached as parents, children, siblings (Luhman's system does 
this, right?)
 . direct links with other zettels
 . tags, which creates a peculiar, separate but interwoven mapping, which 
has even fewer restrictions than the acyclic graph. I want to be able to 
look at all tags 'associated' (through the zettels but not directly) to a 
given heading or set of headings.
 . organizing nodes  Luhmann used indexes, for one, and I have not explored 
Leo's concepts on this. I expect there is a lot one can do with organizing 
nodes
 . other organizing features not yet mentioned or conceived...

It almost seems as though the acyclic node-mapping system should hardly be 
called hierarchic, as the very word implies binding limitations, which 
clones and acyclic graph bring escape from. And yet a Leo file is a 
hierarchical outline (or perhaps I don't yet grasp how clones change the 
apparently strict hierarchical display of the Leo outline).

I guess this is one reason you like mind maps. I do too, even though I've 
given up on them a couple of times already and don't currently use them. 
You probably enhance the mindmap idea with semantic processing concepts 
that would be great to  have in zettelkasten.

I can visualize opportunistically traversing an integrated mindmap of all 
the above-listed mappings, just in search of inspiration, no particular 
purpose or topic in mind. Each node in such a mapping would access all the 
headings, tags, organizing nodes, etc that intersect it; all the types of 
handles integrated in one map. Or equally imaginable, traversing such a 
mapping with a particular purpose very much in mind. It becomes a capable 
search tool as well as a rich browsing context. Wild, I know, and quite 
possibly too much mess in practice to be worth anything. Yet the concept is 
intriguiing to me. Yep, wild.

6. Link Navigation
And these links are (or can be) embedded in text, right? Do they, or can 
they, also link directly to a specific place in the target note, or simply 
to the note as a unit? Another way to say that: can the target of a link be 
an embedded marker in the target note? I can easily imagine, e.g. in the 
case of more developed zettels such as chapters or textually developed 
outlines, wanting to link to a particular place in the target zettel. I can 
also imagine wanting to clone selected material from another zettel into my 
current zettel. 

7. Time Stamp
agreed generally. I do want date of origin, and I would like modification 
dates as well. The vast bulk of my zettels will be extracted from archived 
files, so I will have to manually enter those dates of origin as I prep 
those files for the parser. And I want to be able to search the 
zettelkasten by time stamp as well as (and in combination with) other 
handles. If the system-applied UID is time-based (YYMMDDHHMMSS, since I 
might create more than one zettel within a minute), then the UID could 
double as the timestamp. But, for my zettels drawn from archive I would 
have to append additional identifying characters after the YYMMDD portion 
of the (user-created) UID, since many of those zettels will have been 
created on the same day but without HHMMSS data available.

But once again I want all this necessary and highly desirable baggage to 
disappear when I'm writing. Get outta my sight and leave me alone until I 
want ya! But when I do want ya I want ya right now, with no fuss and 
bother! I'm hard to get along with.

8. Tags
And tags form an interwoven and searchable mapping, as above.

9. Backlinks
agreed, and as above can inter-zettel links go from text point to text 
point? as opposed to from zettel unit to zettel unit? (preferably both 
types)

10. Display Styles
In Leo just as in MacOS Finder (I don't know about others), you can arrow 
through a file list viewing the files in another pane. I like that a lot 
and it's way better than not having it, plus with Leo you can toggle right 
into the file, which you cannot do in Finder.
And as you say you can create ephemeral subsets to peruse and play.

I don't know how the user's mechanisms by which to discover and make such 
collections are to play out efficiently. 

But the enhanced display capabilities you mentioned are highly desirable as 
well, along with ideas I mention above and in an attached file a few days 
ago.

I need to better understand organizing nodes and how one creates them, how 
they are used, how they interact with the other organizing/navigational 
elements we're discussing. For example, can one easily, quickly, 
intuitively compose an organizing node somewhat as we do a term-based 
search? And perhaps thus the organizing node itself becomes a sort of 
search engine? And then one can perhaps reorganize the results as desired?  
And then how about a mapping of organizing nodes? What would that be like?

Blessed ignorance. It allows my imagination to go where none have gone 
before (not likely, that).

I don't know LeoVue at all. Interested in how it could enhance our system. 
I have sometimes wondered if my ideal system would have use for an on-board 
server, though I have no knowledge in that area. How much computer does it 
take?

-- 
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/6bfff1eb-edd9-49dd-aefb-1a66f0e220ae%40googlegroups.com.

Reply via email to