Arguments for 'connected note-taking'
I'd be interested to hear if there are any compelling arguments for using a connected note-taking tool like Roam/Logseq/Obsidian, assuming you still do the work of making your own connections rather than letting the software do it for you.
@sfast @ctietze do you see a benefit to the (potentially unexpected) connections between information that these tools can surface, which in turn might help you connect knowledge?
One benefit I've seen in these tools is that for some reason they do make it easier to just jot down lots of thoughts. It's probably because the Daily note eliminates the need to think about where notes should be saved.
Howdy, Stranger!
Comments
I used Roam for 3 months, putting 100,000+ words into it. I really enjoyed my time with it, but I ended up returning to a more zettelkasten approach in plain text.
If you make your own connections instead of letting the software do it, I don't think that there is any difference between using Roam and using something in plaintext, other than the difficulty of getting your notes out of Roam. That said, I think you would have to be very disciplined to use a tool like Roam and not take advantage of its features. It didn't take long before I was linking between blocks, linking topic words without writing notes in that topic's page, nesting information into all sorts of branched hierarchies without splitting it up into atomic components, etc. Roam makes it very easy to put all of your notes and thoughts down in a tangled web of connections. I definitely wrote more when I was using Roam.
In ZK, you are deliberate with the contents of your note, you think about what context you want to find the note in, and you have to deliberately place links to other notes and contexts. The time it takes to do this, while not excessive, is enough to push you towards curating the notes you take. I find that working in my ZK can push me towards a flow state. Roam never gave me a flow state, and I think what was was missing was the friction. It was too easy to put my thoughts down, too easy to connect to any little bit of previous text. Which is not necessarily a bad thing. I really enjoyed the frictionless writing and linking.
In this way, I think that Roam (and I guess Logseq, which I haven't used) is the flipside approach to ZK. It gives you a place to put all of your thoughts and connect those thoughts. It has some aspects of ZK, but it also removes the friction enough that you can more effortlessly capture any stray thought or bit of information you come across. It isn't a filing cabinet like Evernote, and the linking ensures that you will come across that information again, so I don't think that it strays too far into the collector's fallacy. It does allow you to more quickly collect information, though. I think that it is an approach to dealing with and trying to tame the firehose of information that we're confronted with in the information age. With ZK, you have to curate the information that you take down. With Roam, you can capture anything that catches your eye. This isn't necessarily good, nor is it necessarily bad. It's just a different approach. Do you find it more valuable to be constantly collecting and interacting with knowledge, or do you find it more valuable to carefully build your notes? With Roam, I felt immersed in my notes. It was an application that I never closed and was always adding to. With ZK, taking notes is a different mental state that I have to put myself into.
That said, Roam is very hard to get your notes back out of. Maybe it's easier now with Logseq, but the nested hierarchy of the outline structure makes it hard to convert your notes into plaintext while retaining the context implied by the hierarchy, not to mention that the block-level linking only works so long as you have a program that can read Roam's internal structure. This is ultimately what pushed me to leave Roam. Until I feel like I can reasonably expect to be able to store a local copy of my notes, and be able to interact with those notes in perpetuity, I can't comfortably trust my notes to Roam. That's just me, though. I know that many people trust that Roam will be around for a long time, or that the benefit they're getting from Roam is worth the risk of losing their notes, or that other tools with definitely be developed to read their notes-backups if Roam ever shuts down. I wish that I had that trust, because I really really love working with Roam, but for now I'll stick with plaintext.
As an aside, I didn't touch on Obsidian here because I don't think that Obsidian belongs in the list with Roam and Logseq. Obsidian takes markdown notes in plaintext. It does have some proprietary syntax you can take advantage of, but it in no means forces you to use that syntax, nor is that syntax so useful that you'll feel like you're missing out by avoiding it. You can easily use Obsidian like a normal ZK. When I tried using Obsidian like I was using Roam, it was not pleasant. Obsidian has the same sort of friction that forces curation that something like The Archive has.
Edit:
Not to add further to a wall of text, but I see you commenting elsewhere that backlinking is one of the features that you see as being potentially problematic about Roam, and this is why you've included Obsidian in this list. I don't think that I'm particularly persuaded by the arguments I've seen on this forum against using backlinks. Backlinks are still a link that you made. Fundamentally, you had to make a forward link for there to be a backlink to view.
I think that an an argument could be made against the use of backlinks as tags, definitely. If you put brackets around any topic word that you even halfway think might be useful in the future, and rely entirely on the population of the backlinks tab for adding "structure" to your notes, then you're bound to dilute the potential information stored in your links if you aren't careful. But I don't see this as much different from using an over-large number of tags.
The other caution you should have when using backlinks is that backlinks typically show the context around the linked word, but not the context in terms of the linked-to note. In other words, you can capture Note A -> Note B. The backlink tab still shows Note A -> Note B, even if you are on Note B. You may be missing information if you don't also explicitly write down Note B -> Note A. You also may not be missing information. Sometimes A -> B is meaningful, but B -> A is not. This is a situation where backlinks allow you to view useful information that is harder to find without backlinks. You can see that A links to B, and the context around that link (since you can see a few lines on either side of the link). You can also see that B -> A isn't necessarily meaningful, but you can still decide if you want to swim upstream, so to speak.
This comes up a lot in science. Lots of different topics take advantage of the Bose-Einstein distribution, the Fermi-Dirac distribution, the Arrhenius equation, etc. These are equations and topics that appear in all sorts of other situations. Often, I'll see an equation and instantly realize that it's built upon one of these foundations. It is useful, in this instance, for me to link that equation to one of these foundational equations. If I always did the reverse--linking the foundational equation to the topic built upon it--I would end up with an overwhelming number of links that don't contain much information other than the fact that the foundational equation is... foundational. On the other hand, I can look in the backlinks to see which other notes point to the foundational note, and I can swim upstream to look at some of these notes if I want. This allows for the links that I do make in these foundational topic notes to not be diluted by an endless list of all of the other science topics that were built upon them.
ORCID: 0000-0003-2213-2533
@jameslongley Since you pinged me: I don't know, and I don't want to comment or speculate on paid services as long as we have something to sell on this site. That'd be dishonest. See: https://zettelkasten.de/our-software-review-policy/
Author at Zettelkasten.de • https://christiantietze.de/
Sorry, can't help with "compelling arguments for using ..." @prometheanhindsight clearly spells out the reason having the software think your connections for you is a bug rather than a feature. I agree with @prometheanhindsight's view that user-generated backlinks have a missing value that system-generated ones don't. The goal is to have a mass of well-thought-out links that connect two or more ideas, created by the user, for the user's future self.
I've been experimenting with Daily Logs to "Accumulate insights." in The Archive with some success.
Will Simpson
My zettelkasten is for my ideas, not the ideas of others. I don’t want to waste my time tinkering with my ZK; I’d rather dive into the work itself. My peak cognition is behind me. One day soon, I will read my last book, write my last note, eat my last meal, and kiss my sweetie for the last time.
kestrelcreek.com
I've been playing around with NotePlan. It has general notes (in which you can set up work projects or knowledge categories or whatever; could also use it for a ZK) and it also has a daily note. The daily note is integrated with what is in your Calendar (same view), so I keep that open and enter whatever thoughts come up during the day. That includes:
NotePlan handles tags - it has two kinds, one preceded by a "#" and one by an "@". It also links to other notes (using the note title; I don't think it links to a specific location within a note - similar to The Archive). It also lets you assign a particular bullet point or task to another day (using the format >2021-02-08) so that the task or bullet point you've just written shows up in your daily note at some time in the future.
After working in NotePlan for a couple of weeks, I find I've got it open and am using it frequently during the day; it's become a regular tool. You may find it has some advantages for maintaining "daily notes" rather than just throwing those directly in your ZK. However, I still prefer using The Archive for my ZK itself.