Some questions I have are keeping me from committing to my Zettelkasten
Hi, this is my first post, and I want to thank everyone who can help me reduce my resistance and avoid failure. Starting a Zettelkasten digitally and future-proof is not something easy to do.
In my case, I want to create some guides and rules to guarantee some success (or headaches in the future).
And there is some of the question that are keeping me from committing to my Zettelkasten:
- I start using file name as
yyyyMMddhhmm Title of the note.md
- Should I avoid spaces in the files names?
- Should I use Folgezettel? Or using
yyyyMMddhhmmis enough to guarantee the evolution and future prof of my notes?
- Using Obsidian and The Archive as my main tools for Zettelkasten, I started using wikilinks
[[yyyyMMddhhmm Title of the note]]to connect notes
- Should I use Markdown links
[yyyyMMddhhmm Title of the note](/path/to/the/file)or Wikilinks is something standard for markdown?
- Since Obsidian supports properties using YAML and The Archive is just a markdown editor, I using YAML with this format (
202501091122 four values of the agile manifesto)
I don't have tags for now on YAML because The Archive don't search on click when using tags: [tags1]
- Should I use tags at all?
- If yes, should I put tags on YAML or in the body of the note using
#tags1?
Ok, I have more questions, but those are the most important ones for me now.
Can I have some feedback on that from you?
Howdy, Stranger!
Comments
Hi,
Using folgezettel or not is a personal thing.
I don't use it, but having made this choice I believe it is very important to use another method that takes its place in some way.
I need (and in the future you will probably need, too) a model that allows me to express sequences of ideas, collections of ideas, trains of thought, things like these. I use higher order notes (notes that contain links to other notes) for that purposes, instead of folgezettel.
The possibility of browsing with eyes sequences or path of notes already taken in the past is one of the most important features of the Zettelkasten.
Regarding tags, very broad topic. Very personal thing.
I don't use them for subject classification, i use links to "subject notes" for this purpose.
I have tags for status modeling, progress modeling, annotating some notes for some operations, but I rarely use them in practice once I've put them. I tend to make my cluster of notes collecting links in specific notes. For some purpose I prefer to use metadata with Dataview, but this is an Obsidian use case outside the my Zettelkasten space.
I can't say much about the other things, I don't know how you plan to use the combination of both Obsidian and The Archive together.
Hello and welcome! You're asking good questions. You're right that starting a Zettelkasten (ZK) that's future-proof isn't easy at first—it was not intuitive to me. Once you get the hang of it though, the Zettelkasten Method (ZKM) is wonderfully simple.
If you're hesitant to start your ZK, I recommend beginning right away. I delayed starting my ZK because I wanted to learn as much as possible beforehand to avoid common pitfalls/mistakes. However, I found I learned the most from practicing the ZKM itself and making those mistakes anyways. If I could do it over, I would start practicing the ZKM right away and learn from my mistakes much earlier. I also would have created more notes.
I will do my best to answer your questions, but then I encourage you to commit to starting your ZK.
Not unless you're a CS expert with a good technical reason not to. And luckily bulk file renaming is technically very simple. So if you ever needed to, for example, replace all the spaces in the future, you could easily do so.
Like @andang76 said, Folgezettel or not is a personal choice. Personally, I use date-based UID, as I feel it is simpler than Folgezettel, especially for digital ZKs. To directly answer your question,
yyyyMMddhhmmis plenty sufficient to future-proof your notes.I'll address the remaining questions together, as they all pertain to your reliance on using Obsidian and The Archive to interact with your ZK.
>
Like @andang76 said, "I don't know how you plan to use the combination of both Obsidian and The Archive together." I suggest you don't rely on either. I think that the software-agnostic approach that Sascha and Christian emphasize is essential to keeping a digital ZK. Here are two forum threads related to software-agnosticism worth checking out:
Software-agnosticism has a significant impact on using a ZK, which makes it very relevant to your remaining questions.
The Archive was designed to be software-agnostic, and switching from Obsidian to The Archive showed me just how important that is. In Obsidian, many parts of my workflow were tied exclusively to its features. When I eventually decided to adopt a software agnostic approach, I realized how unnecessarily dependent I had become on Obsidian's features.
If I haven't convinced you to use The Archive, hopefully I've convinced you to work towards software-agnosticism in your workflow. However, I think you'll find it difficult to have a software-agnostic workflow while continuing to use proprietary software features like those in Obsidian.
If you were just using The Archive, this is how I would answer your remaining questions:
The Archive uses Wikilinks to function as search criteria. Use Wikilinks to link to another note, but only for the UID (e.g., [[yyyyMMddhhmm]]). And be sure to provide a good link context.
The Archive doesn't use YAML but it does use tags to search. I honestly don't really use tags. Like @andang76 said, tags are a broad topic and much of it comes down to personal preferences. There is good advice on these forums about preferring ontological tags. Others prefer links over tags all together. You might get started without tags and see how far that gets you.
@ivomota I'll jump in with a brief comment on the use of tags. I really like them and use them a lot. I use tags to associate a zettel with one or several topics, to show the status of a zettel (e.g., completed or not, linked or not), and to connect zettels to the names of individuals (either the zettel is about an individual, such as a personal history note, or an individual is the author of a quote). If one uses multi-level tags (e.g., "dams-failures-overtopping being an example of a 3-level tag), one can also develop a very nice index just by creating a list of tags, (@Will has a Keyboard Maestro macro for that).
There are no guarantees in life. Just start and make adjustments as you learn more.
This is not necessary with modern file systems and apps.
Using a time stamp will help future proof your notes.
Use markdown links when referencing an external file or website.
Yes, you should use tags. The way tags are searched for is different between the two apps.
Yes, this is a mess right now. I'm still trying to figure out a good strategy for tags in the various apps. I use The Archive as my zettelkasting tool, but I’ve started using a static site generator that uses the tag field of the YAML but refuses tags prepended with a #. This is likely a feature or reserved character in YAML. I use, and recommend using YAML for every note. It is part of my note templates. This is a sample of the YAML I typically use. Followed by the YAML, which I use for notes destined for 11ty, my static site generator.
--- title: # Confident of Uncertainty UUID: ›[[202412231629]] date: 2024-12-23 archive_tags: #blog-post #thinking-tool #zettelkasting tags: - note - zettelkasting - journaling - uncertainty description: Clear seeing builds confidence in uncertainty. ---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
@Will YAML indeed treats
#in a special way: as a comment!https://github.com/yaml/yaml-grammar/blob/75083c2aaabd0d7453d72de895666b21021826c1/yaml-spec-1.2.txt#L458-L460
You should get a string with
Which at least is not nothing
Author at Zettelkasten.de • https://christiantietze.de/
This community is awesome.
Thank you everyone for every comment regarding my questions.
My post will be long
> @andang76 said:
As a clarification, I have an Obsidian Vault, and inside I have a folder with my Zettelkasten. In my Obsidian Vault, I have Daily Notes and other notes not related to my Zettelkasten. I like Obsidian and the customization I can apply to it, even using it in a very simple and raw way. But I like being able to open my Zettelkasten and interact with it inside The Archive, which is a simpler and more direct app for Zettelkasten. Also, I really like the search function and the list of notes in The Archive.
Obsidian is also a good tool to visualize data with the graph views and Dataview plugin.
This seems to be a very good and wisdom advise. Thank you!
Right! On macOS I can easily bulk file rename. That is very truth I hadn't even thought about it.
I preffer this approach too!
I don’t get the software agnostic argument against Obsidian, for example, with the argument of doing a search on The Archive and not on Obsidian. Since this is just text, on the backend, when you click on a Wikilink, it is doing a search and redirecting to the note (this is why in Obsidian you need to have unique titles). Also, you can search for the content inside
[[]]and get the note in the search. It’s just text.I don’t understand why Obsidian is not considered software agnostic when you use it as it is (without plugins) since it is just a markdown editor.
This is why I choose to use
[[yyyyMMddhhmm Title of the note]]and an alias with[[yyyyMMddhhmm]]to be able to use it in Obsidian and The Archive. Also, using my note in two tools with different approaches challenges me to make my note more feature-proof, I think.Instead of tags to connect zettels to the names of individuals I preffer to use links, since for me a parson/author also can have a note where I can put things about that peson inside. Also it helps to use as a search for backlinks. If I open in The Archive or Obsidian a note from an Author, I can find all the notes where the author was mentioned using as link. But I also like to use tags for status!
I Will
Yes, normally I do that, it is the makdown way. My question was more about internal links.
Some comments on this:
This is not an UUID (universal unique identifier) this is just an unique identifiers.
An example of uuid is more like:
8D8AC610-566D-4EF0-9C22-186B2A5ED793You can read more about UUID here on wikipedia
But why do you use the › before the link to the note? I’m very curious about that
I have the same questions regasrding tags. I also use 11ty but I did’t start post notes online. But I still having hard times to figure out if I should use tags on the body (that i dont like) or in the YAML that I prefer but thei are not tags with the
#.But this commetn from @ctietze is something very interesting and probably the best solution for me since it also works on Obsidan.
I can have the best of 2:
Thank you @ctietze
Sorry for my misunderstanding. I use links formatted like
[[yyyyMMddhhmm]] for my internal links. I adopted this standard early on and have sometimes wondered if it was the right decision. Starting over, I'd have the link be the file name and I'd put the UID in the YAML and not the file name. I'd spend more time developing a Keyboard Maestro macro that monitored note title changes and then changed the filename and all the links.
Yes, this is not an IETF UUID. 202412230736 is a notes ID unique to that note. When it is encapsulated in [[]], it becomes a link to any note with that ID anywhere in the note. It is no longer specific. Prepending the › character allows me to create Keyboard Maestro macros restricting a search to that particular note. I use this in a number of macros to make sure I'm acting on the target note.
I feel your pain. I've resolved to put tags in the YAML. With Keyboard Maestro's help, I currently use both formats where appropriate.
Here is a sample
--- title: Curiosity as the Key to Communication UUID: ›[[202412200523]] cdate: 12-20-2024 05:23 AM date: 2024-12-20 archive_tags: #blog-post #mindset #conversation-skills #fundamental-attribution-error #correspondence-bias tags: - blog-post - conversation-skills - mindset - fundamental attribution error - correspondence bias description: Genuinely engage in conversation with others by showing curiosity. ---This does not work when 11ty processes a note using this formatted tag scenario. I get one long concatenated tag. (#science #article #medical)
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
We could maybe discuss YAML frontmatter parsing for 11ty in another thread if you want. This was just meant as a way to get the data into the YAML preprocessor at all -- I hope you can parse these if needed in a preprocessor step? Otherwise, you need
which would give you an array of 3 strings with the hashtags inside, but it's also maybe more typing than you would want.
The Archive also Does What I Mean™️ with these.
Author at Zettelkasten.de • https://christiantietze.de/
I don't know how software handles tags, but I would assume you don't need a pound sign for tag names in a YAML header. So this would be fine I guess:
The pound sign is used to make tags human readable, when you use #tags embedded into your texts.
my first Zettel uid: 202008120915
Others have previously argued more convincingly for the software agnostic approach than I have. But perhaps sharing my experience with Obsidian will show why I think using Obsidian is not software agnostic. Basically, I don't think Obsidian — even without plugins — is "just a markdown editor."
I think the biggest thing for me was realizing how link-as-search is more software agnostic (i.e., "more preferable" in my opinion) than direct links. See this discussion where I link to additional posts on the topic of link-as-search as software agnostic. In the ZettelkastenDe YouTube livestreams, Sascha and Christian demo note-taking in the Archive and show/explain how the link-as-search feature in the Archive is advantageous.
I think that direct linking alone proves Obsidian is not software agnostic. But beyond that, I generally think Obsidian forces you into a certain way of working that makes you rely on idiosyncratic features. Out the box, Obsidian comes with a bunch of eye-catching features enabled that I find distract me from the work of concentrated thinking. A good example of this is the command strip and the ribbon. It's a hassle to disable all these features. And I found I was enticed by all the features, wanting to enable, customize, and maximize everything.
I about 2 years ago, I knew I needed a new approach to note taking — what worked in school wasn't enough anymore. At the time, I knew very little about ZKs and the ZKM. What I found first was Obsidian and so-called "Second Brains". My first approaches were primarily modeled on Nick Milo and Tiago Forte. And I was hooked on Obsidian.
But at a certain point I found my process was cluttered and complicated by unnecessary workarounds and "hacks", and I needed to simplify things. Over time, I learned that the ZKM, as developed by Sascha and Christian (from Niklas Luhmann's prodigious techniques), was the simple, software-agnostic approach I needed for real, focused knowledge work. Of course, this is my own experience and opinion, which I'm happy to discuss further if it's at all helpful.
@dylanjr I understand where you're coming from with Obsidian. I went through a similar rabbit hole of "oh that looks cool, let's do that" a thousand times and ended up with an unwieldy system that I had to start from scratch with intentionality.
I am however strongly disputing the fact that Obsidian forces you to do anything. At its core, it deals with Markdown files with an agnostic enough approach (wiki links became standard) that you can move your data pretty much anywhere…
… unless you start messing with Dataview and all those plugins, of course. But Obsidian is completely open-ended, and whatever features you decide to add on top of it – possibly way too much; been here, done that – is on you.
I'm not blaming Obsidian for having made a mess of it, I blame myself because I could not resist and didn’t have a clear view of what I wanted to get out of it. To me, it's a little like Scrivener or OmniFocus – totally open ended tools that first require of you to think exactly what you want to get out of them if you want to ever be able to wrangle them.
"A writer should write what he has to say and not speak it." - Ernest Hemingway
PKM: Obsidian + DEVONthink, tasks: OmniFocus, production: Scrivener / Ableton Live.
I have an open relationship with my Zettelkasten. Ok half-open, since it doesn't know what I'm doing with other note-taking systems. As far as my Zettelkasten is concerned, I'm committed.
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.
@ZettelDistraction I appreciate your morally ambiguous humor
Yes, this exactly illustrates one of my points.
The point I made regarding link-as-search and software agnosticism is not my own:
"One big reason for Zettlr is the link-as-search functionality. This is perhaps the most important feature implementation to make an app software-agnostic." — Sascha
The assertion is that link-as-search functionality is the most important feature implementation to make an app software-agnostic. I do not dispute this assertion — do you? The implications are clear. Obsidian does not offer link-as-search functionality, therefore Obsidian is not software-agnostic.
To speak of totally open-ended tools, I've been diving into Emacs lately, which in comparison makes Obsidian seem completely "closed-off".
@dylanjr Actually, I do. Software is not agnostic (Obsidian cannot be software-agnostic, it's Obsidian). Only files are. Are your files agnostic enough to be used with anything? That's the real question; how apps implement features does not ultimately matter. If one app does link as search and the other uses a database, as long as the underlying data is open, available and interoperable, what practical difference does that make? (I say: none)
And it seems to me that wikilinks in the files Obsidian produces will work with many other software because they will be recognized for link-as-search (as long as you don't do anything fancy like heading or block IDs). And even if they don't – I can always search for stuff in another app ingesting those files using my link titles, block IDs or (if I thought this in advance) UIDs in notes.
For instance, I have done this importing my old Obsidian system into Bear. Bear does not do block links, nevertheless I could search for whatever I needed using block IDs, because the data was just there in plain text – something like
^l54fj6. Not as easy as just clicking the link, but it was not that big of a hassle: I had lost no meaning. At the moment, I have UIDs for all notes in YAML – links will survive with any Markdown editor going forward.Now, if you use Dataview, Mapview, embedded queries, very fancy CSS and whatnot, for sure, you will definitely entrench yourself in Obsidian. That's a very different thing, and one has to decide whether to cross that border or not. (And if you don't, it's very relevant to question whether you actually need Obsidian at all.)
"A writer should write what he has to say and not speak it." - Ernest Hemingway
PKM: Obsidian + DEVONthink, tasks: OmniFocus, production: Scrivener / Ableton Live.
I’m very happy with this discussion started by @dylanjr and @KillerWhale.
I completely agree with @KillerWhale. Obsidian is an application, so it is not agnostic. In fact, no application is truly agnostic. I see The Archive as very close to this, but I don’t think it can be called agnostic just because it has a “link-as-search” functionality. I believe it is more due to the fact that it represents a markdown file exactly as it is.
This is exactly what I was saying initially. Even within Obsidian, I can use the search to look for wikilinks instead of relying on clicking.
Agnosticism lies in the file itself. If the file is not manipulated by the application at its source, then the file is agnostic.
A good argument is the one shared by Obsidian’s CEO, Steph Ango, also known as Kepano, in his article File over app.
If the focus is on the file (ensuring it is accessible and remains in its original format), it becomes agnostic—it does not depend on any specific application to “exist”
If, in Obsidian, I create a link [[202501281151]] and open it in The Archive, I can search for 202501281151 and always find it because the file will contain exactly what I wrote.
This statement is crucial. We shouldn’t blame the software, but rather our lack of clarity on how we want to use our notes.
Ultimately, it all comes down to personal preference and feeling more comfortable and in control when using one app over another.
That’s why I use both The Archive and Obsidian. I know they each have different strengths, but I also know I can use both because my focus is on keeping my files as agnostic as possible from the software I use to read them.
Without opening another debate, I recently started using NeoVim, and this principle became crystal clear to me. If a terminal can read a note, then, in principle, my note is agnostic.