An essential priniciple of a Zettelkasten is its flexibility. From there you can derive maxims like “you should avoid proprietary software”, and “you should use plain text notes to ensure longevity”. That’s why I don’t use a fancy application with lots of spectacular features but the simple nvALT instead. Here’s how I automate note creation to speed up my workflow.
nvALT isn’t a dedicated Zettelkasten application. It doesn’t know anything about identifiers or references or topic clusters. It only manages text files, and it’s up to us to make a sense of it. We have to plug-in everything related to the Zettelkasten method ourselves.
nvALT creates new files pretty easily, though most of the Zettel markup I deem important is missing. The bare minimum of a Zettel note can look like this:
201105160958 Improve the structure of essays by rewriting Benjamin Franklin improved his writing skill by re-writing other author's articles (Isaacson 2004, p. 28): * Take notes when reading on single scraps of paper. (Like in the reading workflow for a Zettelkasten) * Shuffle the notes sometimes * When the original content is forgotten, write an essay based on your notes and compare to the original * Sometimes try to write in verse instead Cf. <http://www.earlyamerica.com/lives/franklin/chapt1/> #writing #essay #practice #zettelkasten --- References: Walter Isaacson (2004): Benjamin Franklin. An American Life, New York and London and Toronto and Sydney: Simon and Schuster Paperbacks.
According to this example, a note gets a title, a global identifier, tags, and a list of references. You could leave out a title since the identifier is sufficient to, well, identify each note. Having a title is usability sugar. So is the list of references. I think tags are important to form loosely connected clusters of notes. Let’s sketch a wireframe of a Zettel like this (
<<...>> denotes a placeholder):
<<ID>> <<Title>> (The Text will come here.) <<Tags>>
That’s what the script provides: generate a date-time based identifier, insert a title of your choice, and rename the note accordingly.
Because writing Zettel notes has to be as efficient as possible, I created an Automator Service to insert the date-based Zettel identifier into a new note:
Download the Zettel Header Service (Automator workflow) for nvALT (Mac).
To install the Service, unzip the archive and open the included Service file. Mac OS will then ask you to move the script in your
~/Library/Services folder. Agree to do so. Then, you should be able to invoke the service from the menu by selecting nvALT > Services > Insert Zettel Header.
You can assign a keyboard shortcut to this Service from your System Preferences to speed-up note creation even more. Go to your Keyboard system preferences pane and select the “Shortcuts” tab. Add an “App Shortcut”, use “Insert Zettel Header” as the menu title and select a keyboard shortcut (I use ⌘⌥⌃H). No more clicking necessary!
If you’re interested in customizing the script or if you want to understand how it works, read on.
What’s in the Script?
The script does the following:
- ask for a note title
- generate a Zettel ID for the current time (the timestamp is of the format
- rename the active note in nvALT to become
- paste the ID and title into the note
The gist of the script is about invoking keystrokes. I added comments to the script so you can make sense of the keystroke invocation.
You can copy the full AppleScript Gist from GitHub.
If you want to use a more elaborate header, add additional lines of text to the lines of
set theHeader to .... The blank line is in there to create enough whitespace to start typing. You may delete it if you don’t find it useful.
For example, I like to use MultiMardown-style headers. So in my version, the value for
theHeader looks like this:
set theHeader to "Title:" & theTitle & " Tags: #"
This adds two lines of text into the header and places the cursor right after the hash character (
#) so I can begin adding tags right away. I put my list of keywords into the header instead of the bottom of the note, contrary to the bare-minimum example from above. When you pick MultiMarkdown, this makes sense. If you stick to “just text”, I guess you better leave tags at the bottom so they are out of sight.
You can see that the function
current_date() concatenates year, month, day, hours, and minutes into a string like
201410011105 for “2014-10-01 11:05 a.m.”. You could add the seconds, too, by adding
(twoDigitDisplay(it's seconds)) to
current_date(). I found that it suffices to create Zettel notes with accuracy to the minutes. I assign IDs with accuracy to the seconds to items outside my note archive, though.1
Please share your customizations, and how you format your Zettel header!
twoDigitDisplay()stuff was necessary to pad numbers with zeros, so “1” becomes “01” to ensure consistency among identifiers. I believe this can be shortened, but I wouldn’t know how. Any AppleScript tips are welcome! ↩