The issue of lacking a work bench
This discussion was created from comments split from: Folgezettel is Not an Outline: Luhmann's Playful Appreciation of (Dys)function.
Howdy, Stranger!
Comments
The drawback is that the original running text is often interrupted by hundreds of notes, but if you are methodological with the numbering...
Now in my experience, I think the Folgezettel/Timestamp/Structure note wars miss the point. I have another problem--which could be due to my workflow, but maybe not. I am missing a visualization feature that Luhmann had (and that Nabokov had, for that matter): the ability to spread note cards out on a table.
Confusion
Reconciling similarities among notes had become a bottleneck. Some notes could be combined and condensed; others could be split into several notes. Switching from one note to another in Zettlr is inefficient for comparison. The ability to spread cards on a table for comparison is one of the advantages of physical Zettelkasten over digital Zettelkasten. I thought about printing notes and marking them up or jotting down IDs and titles of notes that I wanted to edit, and indicating edits on paper, but I kept putting that off.
It was hard to take stock of what I had. My Folgezettel IDs were too abbreviated for this. Structure notes could help up to a point, but they had to be written and modified in addition to the notes already present, and they couldn’t substitute for the kind of overview possible by spreading notes out on a surface. The structure note presupposes having in advance such an overview; e.g., an outline or an ordering. The "having an overview in advance" prior to the creation of a structure note, combining or simplifying notes is at issue. And this prior issue cannot be resolved by saying "use Folgezettel or use Structure Notes."
What was lacking? A means of visualizing, adding, removing and re-ordering notes prior to re-combining and synthesizing notes into a new note (possibly a structure note), and prior to simplifying existing notes.
The Scrivener Corkboard
Scrivener 3 could be used to impose some additional structure on the Zettelkasten. Scrivener has a "binder" feature, through which documents can have sub-documents that can be visualized on the Scrivener corkboard. Because Scrivener represents text files in rich text format, the workflow has to include importing markdown files of the Zettelkasten into Scrivener, working with them in the corkboard, and then exporting them. Scrivener will only allow references to media files, but not to markdown or text files, so importing and exporting markdown is unavoidable.
Against that, Scrivener will work with pandoc, and a ruby gem called pandocomatic. Pandocomatic will read the pandoc export files that I defined with Zettlr for LaTeX and PDF LaTeX, and the modified latex template for pandoc, and will perform the same export function from Scrivener to LaTeX and PDF latex that I have for Zettlr.
A solution: dissolve the problem by postponing it
The plan now is to leave the Zettelkasten as it is and add Scrivener for projects that draw on the Zettelkasten. Any repetition can be resolved within Scrivener and fed back into the Zettelkasten. The technical issue is to ensure that Pandocomatic will read the same configuration files that Zettlr reads (or a computed function of them), to be used with Scrivener. This will give me the corkboard feature of Scrivener when it is needed. And it will tend to emphasize document production as opposed to "tending the garden."
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
Oh, I don't know, maybe it is. (Just thinking aloud no reply needed...)
predecessor | / \ / \ neighbors -- child child -- neighbors \ / \ / | siblingsOf the nomenclature I've been mulling over, Chirality[1] keeps popping up. But the thing is, merely appearing chiral doesn't necessarily imply that a given relationship &/or pedigree holds, eg - two unrelated structures could be just that - unrelated. Its a silly thought I've been wrestling with & can't wrap my brain around it just yet.
DOM[2] has long since went down this road, XML too. A nifty tool for 2D UML[3] modeling can found at the [4]th link. I used it to build an asynchronous message-pump[5] many moons ago.
Back to work for me...
https://en.wikipedia.org/wiki/Chirality
https://en.wikipedia.org/wiki/Document_Object_Model
https://en.wikipedia.org/wiki/Unified_Modeling_Language
https://www.cadifra.com/
https://en.wikipedia.org/wiki/Event_loop
Tinybase: plain text database for BSD, Linux, Windows (& hopefully Mac soon)
I use Scrivener a lot for work projects and that is exactly what I do regarding interactions with my ZK. Good advice!
Now you know how long it has taken me to discover what you already knew. It's certainly worth knowing that I'm on the right track (or a right track)--at least that I haven't run off the rails.
Do you use Scrivener + Pandoc as well? Scrivener + LaTeX?
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
I develop the overview I think you are talking about on the Structure Note. It is actually the canvas on which I develop the higher level patterns (aka new knowledge). So, it is not a presupposition for me at least.
But perhaps the difficulty stems from the mathematical content?
Do you need to edit the individual notes while working on the synthesis (~structure note?)? If not you can use Scapple. You can drag and drop content freely to it as a canvas and then play with it. Perhaps, it is the canvas on which you can play.
I am a Zettler
I've looked at Pandoc; it seems really powerful. However, I mostly export directly from Scrivener to PDF, Word or Open Office, and have experimented with direct export to .html and .epub, just to see the product. I don't have any more "exotic" needs (at least, not yet). One of my daughters is a writer and uses Scrivener extensively - I'll have to ask her what her workflow is after that, to get to epub or for "real" book publishing.
There is the option for exporting from Scrivener using Pandoc -> Word (or Pandoc -> epub). I've played with that a bit but somehow don't get a very consistent product.
I assume you are using Pandoc because you want to convert to a file format that isn't available directly from Scrivener?
LaTeX, PDFLaTeX. But also, I wanted to locate the configuration files (with my modifications) that Zettlr uses to produce Pandoc output, and reuse them with Scrivener and Pandoc (or pandomatic or scrivomatic). As of a day ago, given a Markdown file (produced by Zettlr, but it doesn't matter from where) and using only Pandoc and the configuration files that Pandoc uses, I can reproduce exactly the output that Zettlr export produces.
Here's how it's done:
First set symbolic links to some of Zettlr's default files in the pandoc data directory. This was initially empty on my machine, incidentally. Creating this subdirectory and adding symbolic links to Zettlr's export files (which are Pandoc defaults files) does not change the behavior of Zettlr or Pandoc.
Call pandoc on a markdown file, specifying the defaults files, in addition to the defaults files that Zettlr calls
export.outputtype.yaml, specify the JSON CSL biblography exported from Zotero using --biblography media\MyLibrary.json.This completed without warnings or errors. We may have solved our postprocessing step already! This should apply to Scrivener as well.
Check with WinEDT (my favorite LaTeX editor). The output compiles. Great. This makes Zettlr a removable component--we could use Scrivener -> Markdown -> Pandoc -> LaTeX or PDFlatex
Likewise, we can produce PDFLaTeX (which runs PDFTeXify) as if the export were run from
Zettlr. Same process.
We can now reuse the same pandoc configuration files within Scrivener (as a postprocess). Pandoc
s files include YAML pandoc defaults files, a LaTeX template (a lightly modified LaTeX template for pandoc), and a JSON CSL bibliography file exported from Zotero and imported into Zettlr. This means that we can continue to use Zettlr for the Zettelkasten, draw from this as needed for projects written in Scrivener, and process those using the same configuration files for output that Zettlr uses. This streamlines the workflow and the configuration.
This is the workflow I am working toward:
A. Fleeting Notes → A | B | C
- B. Zotero → A | B | C
- C. Zettlr → A | B | C | D | F.n
- D. Scrivener → A | B | C | D | E | F.3
- E. Scrivomatic → A | B | C | D | E | F.n
- F.1. LaTeX → WinEDT → PDFLaTeX →
- F.2. PDFLaTeX → PDF →
- F.3. Markdown →
That should explain what I am up to. An attempt to simplify the workflow through config file reuse.
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
Then you must have a pretty wide monitor.
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
Haha! You are way ahead on the curve from me, but I will consider what you are doing and try to draw out some hints for my own use case. Thanks for all the detail!
No.
I just use my laptop monitor, even ditched the external monitor half year ago. So, I have 13'' and my trusted note block to the right. 
Keep in mind that math, physics and similar formal languages put a much higher strain on your working memory. The canvas as an aid for the working memory is way more important to you (math) than for me (words).
I just hold anything I need to know ready to use in my mind until I am finished. I couldn't work like that if I'd have to deal with math.
So, an infinite canvas like scapple could be the solution for you.
Additionally: My current hypothesis is that for math/physics it is even more beneficial to use the ZK more for the retrieval part and paper for the thought development part. Methodically, there is no difference if the Zettel you edit is physically in front of you while the supplemental material is on the monitor. Practically, the difference is that you'll end up with a zetteldraft which you'd process back into the Zettelkasten.
I am a Zettler
...
I was making a joke, but you saw that.
I cry Uncle! I'm using Scrivener anyway--I might as well try Scapple. It has one advantage over IDs: arrows. I'm more interested in what links to what than the specific ID, though I have a soft spot in my head for the rough classification of Folgezettel. Maybe the prions will attack that portion of the brain.
I have an additional method when confusion sets in, which is to do something else. When I was younger I would do nothing else. Confusion often means my approach is wrong. Another is to attempt to form longer-term memories, so that I can keep more in working memory.
@GeoEng51 and I share a similar workflow, which is to use Scrivener, especially the corkboard. This is almost kicking the can down the road, which in this case is a good idea. Like @Sociopoetic, the result can be fed back into the ZK. That's why I wanted to reuse in Scrivener the user-configurable Pandoc defaults files and the Zotero JSON bibliography that Zettlr uses. This is doable. It helps to have different views of the same data.
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
Having tried Scapple, I am uncertain of the result.
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
I am, too. I thought that you could split your screen and dumb all the notes you need into scapple and develop the note in your app of choice.
I am a Zettler
@Sascha said:
"I thought that you could split your screen... "
Well why didn't you say so ...
"and dumb all the notes you need..."
My notes are dumb enough already
Actually Scapple grows on a person. Unlike hair.
GitHub. Erdős #2. Problems worthy of attack / prove their worth by hitting back. -- Piet Hein. Alter ego: Erel Dogg (not the first). CC BY-SA 4.0.
Without evidence I will state confidently: 95% of all troll faces are posted by you.
I am a Zettler