Atomicity principle and note titles

d503

Long time lurker here...

I've been struggling with the atomicity principle. In short, I noticed I naturally tended to name and create notes about topics. I will use an example from current work.

I'm reading on the securitization of European borders. I can make a note called 202011120357 Securitization of migration. Its name suggests a topic. The note should include what is, what is not and examples of securitization of migration.

After reading the recent introduction to the method, I got very aware of the difference between topics and thoughts. So I went back to my working note and renamed and reframed what I had so I have notes like 202011120409 Borders control the movement of people based on their origin or 202011120410 Securitization of migration means framing migration as a security risk. Those are assertions that need some kind of support. A sort of as-complete-as-possible answer as to why that is true, a need for context or relevance. And that should be placed in the content. It is clear that I can place the IDs in other text as a way to call the whole explanation for that assertion.

If this all makes sense, I've been wondering:

• How meaningful do you think titles for content/permanent/thought notes are? How do you title the notes? It seems that the title is just a quick way to know what the note is about but it shouldn't limit the content like, say, the title on a Wikipedia article. In short, if there's not short title one can come up with, you could use the first X words of the note to make the title. In any case, the title shouldn't be a topic when it contains a thought.
• How do you manage to work on both the thought and the topic level? If one wants to create a new Zettelkasten and reads any article or book, it is going to formulate the thoughts but also will recognize topics. In many examples given in the forum I see topics and assertions as titles of notes. Do you write both types of notes at the same time or do it in separate moments? What's the use of a strict separation and is it worth to note it in the ID or the title of the note? I find it difficult to work separately on thoughts, then topics or structure, then linking around and also keeping track of the whole argument in short for the bibliographic note.

I guess every Zettelkasten has its own knowledge entities but the forum kind of agrees on structure, content/permanent/thought and bibliographic notes. Maybe we can find other collections that use others according to the interest of the person creating it. In any case it could be useful to deepen the catalog of possible contents like Sascha did in Three Layers of Evidence.

Thanks!

learning_ran

Since I am still with experimenting Zettelkästen, I am also interested
in this ideas.

My interaction with Zettelkästen beginning with the title
fuzzy search result. I tend to write title with long sentence and with
multiple keywords that I thought the best for describing the content
of this note. This is critical part, since most of time, I start to
interact with Zettelkästen using this approach.

Another approach to acess Zettelkästen is to view from structured
notes. In such context, the note title can be changed according to its
content. If I need to find a note with a nil searching result, I would
like to think about the most related structure that it could be
located in.

ctietze

I think your examples, 202011120409 Borders control the movement of people based on their origin and 202011120410 Securitization of migration means framing migration as a security risk are useful titles because they reveal so much about the content (if the content matches! ). I would personally really like to see these kinds of titles when I work with notes on the topic of migration, I think.

But I don't understand the actual question you're asking. These assertions should be backed by something, otherwise they are just unsupported claims; is something wrong with the "assertiveness" of the titles?

How meaningful do you think titles for content/permanent/thought notes are? How do you title the notes?

For navigation and getting an overview, very important. I often catch myself glancing past note titles I came up with years ago -- because they are not helpful. That's an opportunity to rephrase. Again, I can imagine your example titles to be rather useful, but that of course depends on what's inside and what you want to do with the insides: A note like "X is actually Y" should contain support for the claim, but if you look for all sources that talk about "Y", that probably has to live somewhere else.

In short, if there's not short title one can come up with, you could use the first X words of the note to make the title.

That sounds a bit strange, even though some of my more content-revealing titles look like the first sentence summary.

In any case, the title shouldn't be a topic when it contains a thought.

Can you explain this further? I kind of think I know what you mean, but just to get clear, so we can discuss the following question:

How do you manage to work on both the thought and the topic level?

d503

@ctietze said:

is something wrong with the "assertiveness" of the titles?

I think I wanted to point out that, if the smallest unit, the material of the web we're building, is a thought, an assertion like those in the examples is what we're after when we read and process notes. I guess one can ask what's the grammar of thoughts, how should they be expressed and how titles are clues about what the note contains.

Can you explain this further?

It seems that we can have these distinct elements on a note: (1) an UUID (no questions here about it), (2) a title displayed next to it in the file name and (3) a title inside the note. If we want to make the most out of titles, specially of (2), then it is useful to have some kind of rule for them. Beyond using symbols to identify types of notes, a thought/content note should have a title according to its contents.

The thought, as the smallest indivisible unit we want to keep in the collection, is difficult to title. If the title declares what is it about, then it can potentially contain more than the thought we want to capture because other thoughts speak to it too. So, taking it to the extreme, the best (not the most practical) title for these notes are their content itself or just the first X number of words in it.

A topic, then, is indicated by a title that can contain several thoughts or answer various questions. In my example, securitization of migration can contain (intentional, contextualized links to) the other notes that are formulated as assertions. I think the trap then is naming a thought note with its topic. In an effort to be practical, the atomicity of the note could be clouded by the title. And then it is possible, as it has happened to me, to make the mistake of capturing too much on a note that started as a thought because of its title.

@learning_ran said:

the note title can be changed according to its content.

I just checked a couple of notes on Luhmann's Zettelkasten and they don't seem to have a title. Maybe titles are only appropriate on structure, outlines or topic/hub notes. Just an idea. For all the rest, UUIDs should be more important than the title. I mean, it's not that his Zettelkasten is dogma but it made me think that we might be using titles because we can afford them.

Splattack

@d503 said:

202011120409 Borders control the movement of people based on their origin or 202011120410 Securitization of migration means framing migration as a security risk

Hello, my strategy with filenames is to keep them as short as possible always using only these characters in brackets [aA…zZ0…9-_]. By short I mean less than 50 characters. Each title always has a unique identifier (UID) preceeding. Inside the note I use full-length titles, and subtitles, and tags. (Same with tags, short, few, and very specific tags, not broad words).

If I take your zettels and put them in my zettelkasten they'll become;

20201112a04-borders-control-migrations and
20201112b04-securitization-migration-risk

Knowing than inside the notes the full-length titles still intact or a search will find the notes I prefer short filename titles.

Then, in my convention for time a04 means it was the first note written at 4 am while b04 is the second note written at the same time. At any given hour I can write as many as 26 and capitals if you want to extend to 52 notes per hour. (I've never written more than 10 permanent notes per hour. I imagine that doing more than 10 will be collecting quotes more than writing with intention). By writing a…z you can visually look at the time stamp better and you save 3 characters.

When I link a note inside another note it will look something like;

• For more on migrations see also Borders control the movement of people based on their origin. A one sentence description of the note. [[20201112a04-borders-control-migrations]].

Lastly, for non permanent notes I have a tag #0fn (fleeting notes, and the zero to keep them ordered alphabetically), that indicates me that its title and filename aren't permanent.