Convincing myself of the software agnostic approach

grayen

It is interesting to see how most note-taking software seems to have a huge influence on how notes are structured. To give a few examples:

• Emacs's org-mode: Start with an outline and fill in content underneath each section. Due to it being easy to hide away the contents from view, splitting up notes in small pieces seems less common.
• Notion: Due to a lot of features being surrounded around tables, notes will be organized around them.
• Roam Research: Only having bullet notes makes all notes more structured than maybe need be, and makes almost every note look like a structure note as mentioned here in the forums. The use of bullets also seems to lead to more succinct notes.
• The Archive: Unstructured by default, but there is a heading by convention, and with the use of Markdown extra structure can be given. Though the use of Markdown does seem to have the influence that you end up with a note that could just as well have been a small Word document.

So after reading about a few experiences of people on the forum moving to different software for their Zettelkasten, I think it is important to note that being software agnostic, like the approach taken by The Archive, is about the resulting notes not being dependent on the software that produced them, unlike e.g. Notion, but that does not mean they are not opinionated, that they do not have their own structure, which for example, would not fit well with Roam, so they are not agnostic in that sense. This means that whenever you choose to move to a note taking application that is not software agnostic, you will run into the same issue, although probably to a much lesser degree. The point is making the Zettelkasten future proof in terms of technology, not being forced to fragment your Zettelkasten due to a change in software, thus giving security in that sense. Am I correct in my understanding of this?

Nick

Yes that is the idea. Luhmann used his Slipbox over a 40 year period(two slip boxes). Because digital software is so fickle, people look for an implementation that will stand the test of time.

cobblepot

@grayen , I'm not quite sure what you mean by "software agnostic". Are you talking about software interacting with the standard file format, such as plaintext files, or are you talking (instead or in addition) about note content or the structure of the text in a note?

grayen

@cobblepot I was referencing to the approach taken when they developed The Archive:

Remember that we follow a “software-agnostic approach”. That means our app will be great, but it will still be replaceable. Because we don’t want to lock people in with things they couldn’t do by any other means. The workflow and the method are important, not the tools!

I was talking about both I guess. The file format determines whether it could software agnostic, but also the note content and structure.

Any note-taking software that requires you to export in order to get access to a note as file, would be using a software specific format internally, whether that be some file format or a database. An example of this would be Notion.

And the format of org files from Emacs's org-mode would be an example of note content that can only be fully leveraged when combined with the original software, even though it might be usable as plain text to some extent. That would be an example of not being software agnostic in the note content.

cobblepot

@grayen said:
And the format of org files from Emacs's org-mode would be an example of note content that can only be fully leveraged when combined with the original software, even though it might be usable as plain text to some extent. That would be an example of not being software agnostic in the note content.

Thanks for explaining. I think the clearest way to demarcate whether software is "agnostic" about notes is whether it uses an open standard, such as Markdown or plain text.

But it is interesting to think about your comment that org-mode only allows you to fully leverage the note content when combined with the original software. On the one hand I see what you're saying, but on the other hand if you want to avoid that limitation then you can only use functions that exist across all software that reads your notes. Even with something like markdown, if you use software that takes advantage of non-Commonmark syntax, then those files don't allow you to take "full advantage" of the notes when opening the files in software that doesn't support that syntax

Sascha

but that does not mean they are not opinionated, that they do not have their own structure, which for example, would not fit well with Roam, so they are not agnostic in that sense.

This is very true. Even screen size effects your note taking and the overall make up of your Zettelkasten. In the book, there is some elaboration on the different software environments with some advice how to modify Zettel and workflow. Two examples:

1. If you use a wiki a purely atomistic approach (one thought per Zettel) will slow you down. With a wiki software you will have the ability to refer directly to subsections.
2. Wikis are slower than text editors (because you can't edit right away but need to switch to an editing mode). Therefore, working for longer time sections on one page should be more productive. If you think and craft a lot outside the Zettelkasten and lean towards just storing results the slowness of a wiki might not impact you. A friend of mine is a mathematician. She mostly thinks on paper and stores the results. (Can't remember the software). This would be a good example of not needing a nimble software like The Archive. I personally think a lot in my Zettelkasten and therefore like Wikis just for aesthetics.

So, the software influences your workflow and your workflow allows or forbidds some software solutions. Both influence your Zettel style. The general principles may remain untouched but the realisation of them differ

The software agnosticism comes into play that you apply techniques that are translatable between software with minimum friction. Only software that operates with plain text qualify as canditates.

grayen

@cobblepot said:
But it is interesting to think about your comment that org-mode only allows you to fully leverage the note content when combined with the original software. On the one hand I see what you're saying, but on the other hand if you want to avoid that limitation then you can only use functions that exist across all software that reads your notes. Even with something like markdown, if you use software that takes advantage of non-Commonmark syntax, then those files don't allow you to take "full advantage" of the notes when opening the files in software that doesn't support that syntax

In programming this problem of "you can only use functions that exist across all" is called the Lowest Common Denominator Interface problem. In the case of taking notes, taken to the extreme, you might end up with only allowing to write a sentence or even a word... or a little more useful, just plain text without any special support for any of the notations you might use within it. The nice thing about Markdown however is, that even when restricted to its raw form, plain text, even than it has its value. This is a nice property of Lightweight Markup Languages. It even has its origin in exactly this:

Lightweight markup languages were originally used on text-only displays which could not display characters in italics or bold, so informal methods to convey this information had to be developed.

About the example of Emacs's org-mode, I would say it is much more software agnostic, than say Notion, because the internal structure used by org-mode is the plain text file. The syntax used for their files is also mostly Lightweight Markup, but less so than Markdown. The big problem in terms of being software agnostic for org-mode lies in the workflow with those files. You can work on them outside Emacs, but in practice, to work well with them requires Emacs.

@Sascha said:
Even screen size effects your note taking and the overall make up of your Zettelkasten.

That is something I have been thinking quite a bit about when I was designing an editor for my Zettelkasten. I thought maybe the best would be to go with A5, allowing me to print them well too, but it seems bad to optimize for printing rather than writing.

Your book is not available as ebook, is it? I have some basic understanding of German (I am Dutch, so our languages are quite close + have learned some German in school) and when I tried reading bits and pieces on the Google Books preview of your book I could mostly follow your meaning, but having it in digital format would allow me to easily use translators when I would get stuck.

cobblepot

@grayen wrote:

The big problem in terms of being software agnostic for org-mode lies in the workflow with those files. You can work on them outside Emacs, but in practice, to work well with them requires Emacs.

Can you elaborate? I see why this might be true for non-ZK elements such as scheduled agenda items, but I don't see the issue with permanent ZK notes that are primarily short pieces of text.

ctietze

@cobblepot You end up with .org text files with lists; but the mode of org-mode adds e.g. folding of outlines, so you can keep your sanity while you browse a huge document. (My to-do list for The Archive clocks in at about 72KB, and while I can read it in another text editor, I don't want to work my way through the 1,5k+ lines of text I'd have separated the file into multiple files in any other editor.)

Without Emacs, even more automation and functionality from e.g. org-roam features would be gone.

You can write perfectly fine .org files with Notepad on Windows, but your result will certainly be different than what you come up with in Emacs org-mode with all it's functionality tailored around manipulating outlines, tables, and the like.

grayen

@cobblepot said:
Can you elaborate? I see why this might be true for non-ZK elements such as scheduled agenda items, but I don't see the issue with permanent ZK notes that are primarily short pieces of text.

I could not have explained it better than @ctietze above did, but maybe it is useful to list the various ways software can cause a lock-in:

• The worst offenders have full data lock-in, only being able to access the data through their software. Although most will still allow you to copy text, doing this manually would not be reasonable.
• Or it might not be a full data lock-in, but could still get your data out through some export mechanism. However things will always get lost in translation. A random example on top of my head would be if your software allowed you to color code table cells and you then convert to Markdown, which as far as I know does not support this, so this information can not be exported to Markdown and thus will be lost.
• The software you use is likely to have various features helping you with your workflow, some of those are generally available, while others might be highly specific to the software you use, those features might cause feature lock-in. Where by leveraging those highly specific features you will have to keep using the software to use those features. Since it is at odds with usability (those features enrich your experience), I think this is the most common type of lock-in.
• There is also format lock-in where the format used limits the software you can use. An obvious example is using .docx would limit you to software like Microsoft Word, Open/Libre Office, and other such Word processors. Just like feature lock-in, unless you stick to pure plain text, some lock-in of this kind is inevitable, it is therefore more about the degree to which the format causes lock-in.

I think org-mode falls under both feature and format lock-in. Some features are essential to use the format as intended, for example the TODO lists in org-mode cannot be reduced to just their plain text format, but also the workflow defined within Emacs to work with the TODO lists.

Talking about format lock-in, I was curious how some Markdown features would across different implementations and found this: https://johnmacfarlane.net/babelmark2/

Sascha

Your book is not available as ebook, is it? I have some basic understanding of German (I am Dutch, so our languages are quite close + have learned some German in school) and when I tried reading bits and pieces on the Google Books preview of your book I could mostly follow your meaning, but having it in digital format would allow me to easily use translators when I would get stuck.

No, it isn't.

grayen

@Sascha Do you think the second edition will be published soon, or might this possibly still take more than a year? I ask because I am interested in buying it, but I don't mind waiting a few months to get access to the latest edition.

After skimming the book some more on Google Book Preview on Amazon, I notice I like your writing style, as I am able to understand most of it (70% or so, I think), which is definitely not always the case for German in general for me. What better way to learn some more German then by reading a book I am highly interested in .

Sascha

@grayen said:
@Sascha Do you think the second edition will be published soon, or might this possibly still take more than a year? I ask because I am interested in buying it, but I don't mind waiting a few months to get access to the latest edition.

It will not take more than a year. I am writing consistently for it. Which means that I have the draft ready in the first half of the year. If you already have started a Zettelkasten I recommend you to wait for the second edition. I think the improvement is quite high.

After skimming the book some more on Google Book Preview on Amazon, I notice I like your writing style, as I am able to understand most of it (70% or so, I think), which is definitely not always the case for German in general for me. What better way to learn some more German then by reading a book I am highly interested in .

Thanks. I am "known" (IF people know my writing they often tell me that) for my simple style. I hope I improved my writing since then. If my improvement adds 10% of understanding you can make the last 20% happen to understand 100%.

grayen

@Sascha said:
It will not take more than a year. I am writing consistently for it. Which means that I have the draft ready in the first half of the year. If you already have started a Zettelkasten I recommend you to wait for the second edition. I think the improvement is quite high.

That's great to hear! I indeed had the impression the second edition was not simply fixing some mistakes but includes quite a few new insights gotten since, so I will be looking forward to its release!

Thanks. I am "known" (IF people know my writing they often tell me that) for my simple style. I hope I improved my writing since then. If my improvement adds 10% of understanding you can make the last 20% happen to understand 100%.

I appreciate the simple style. Isn't the goal of a book and most writing to convey its intended meaning, and simple writing, when done right, leads to clear texts. When people go for unnecessary complicated styles, I am sometimes left to wonder what they actually meant and really want to just ask them, and then shout, why then not just put it like that!