Comments

• Andy

  May 14 edited May 14

  This summary is pretty good. I'd say you understand how to do it.

  @zettelkasten said:

  • Use links, not folders
  • Folders assume fixed hierarchies; links reflect thinking paths

  In general this is true. However, I agree with @tjex's arguments for a minimal set of folders. But I emphasize again that any folders would be a minimal set, and you should think through the rationale for that set as well as @tjex did.

  • Periodically re-read old notes
  • Update with new insight; link forward rather than deleting

  How much you need to do this depends on your purposes for using your Zettelkasten. Some notes you may not need to re-read at all. I wouldn't re-read them just for the sake of re-reading. Have a purpose when you re-read.

  • Obsidian, Logseq, plaintext editors with Git backup
  • Versioning tools (e.g. Git, Time Machine)

  Some people around here like Logseq. Others, including me, find it to be too extremely opinionated in its markup constraints. Its features come with the high price of these constraints. Obsidian, which elsewhere you said you use, is more forgiving in this regard.

  • Keep a "last touched" date

  In general, I find that the creation date of any information in the note system is much more important than the last-touched date; I mostly ignore the latter.

  Sometimes I make batch changes to my notes using find-and-replace, which bumps the "date modified" metadata in the filesystem, so that a note's modification date is often meaningless to me.

  The importance of the creation date is reflected in the fact that many people, including me, use the creation date as a UID (unique identifier). Unlike many other people, I generally treat information that I add to my note system as immutable: I don't change it, only append to it, and appending creates a new creation-date UID. I'm not recommending what I do; I'm just explaining why "last touched" date is unimportant for me. In any case, it's important to have some system of accurately tracking time metadata.

  Post edited by Andy on May 14
• andang76

  May 14 edited May 14

  Backup is very important, but I see you have already put it into the list :-)

  An issue than can threat the long lasting of the system is if it doesn't scale well. Meaning what, if the system become hard and frictioning to manage, to fed with new content or to consult, as it grows.

  1) Scalability is not guaranteed or affected by a single issue. Some of the issues are already in the list.

  According to my "Scalability note" :-):

  • bottom up development promotes scalability
  • frictionless notetaking practices promotes scalability
  • leaving the note system completely flat, without a minimum of structure, degrades scalability
  • a bad thinked tagging system doesn't scale
  • a well thinked linking system scales very well
  • forgetting machine nature of zettelkasten improve scalability
  • If I make the system antifragile it will scale very easily

  2) I have a more generic "future-proof" dedicated note:

  In the context of the tools, I've some notes about future proof:

  • even if the tool is plaintext based, be careful when using the software's unique features for things of high value. For example, Obsidian is text based but features like dataview and backlinking are not universal. Use these features only for secondary aspects or short term use.

  In the context of content structure and content organization:

  • folders are a pretty universal approach of organizing
  • tags are supported by many different tools
  • beware of mechanisms that can degrade link stability, adopt the right strategy to maintain it. Note title naming, in particular

  In the context of the development of content and ideas:

  • develop your content trying not to tie it too much to short-term projects
  • pursue concept orientation, make decontextualized content and recontextualize it according to your point of view
  • don't trust that a link to an external source will exist forever, so save the the source itself into your system or make your content independent from it, adding enough context
  • don't create and destroy your system everytime pursuing perfection, let it grows, curating it, yes like a garden

  Post edited by andang76 on May 14
• Will

  May 15

  Let's not forget to include The Archive in this discussion. The more critical the thinking tool, the less it should change. That’s why I stick with The Archive. It does exactly what a Zettelkasten needs: plain text, timestamps, a link-based structure instead of folders, super fast search, backlinking, and no feature bloat—just the essentials, without distractions.

  Meanwhile, tools like Obsidian and Logseq appear in fancy attire, promising the world with every new plugin. But here’s the catch: behind the gloss is the same lock-in and even more cognitive noise.

  The Archive? It’s boring, and that’s the point. Like a library card, it’s stable, quiet, and built to last. No surprise updates. No visual chaos. Just a calm space for thinking that’ll still be working decades from now.

  Minimalism is the feature. And when you’re building something meant to outlive trends, that matters more than glowing graphs or social clout.

  Will Simpson
  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
• zettelkasten

  May 15

  @Will said:
  Let's not forget to include The Archive in this discussion. The more critical the thinking tool, the less it should change. That’s why I stick with The Archive. It does exactly what a Zettelkasten needs: plain text, timestamps, a link-based structure instead of folders, super fast search, backlinking, and no feature bloat—just the essentials, without distractions.

  Meanwhile, tools like Obsidian and Logseq appear in fancy attire, promising the world with every new plugin. But here’s the catch: behind the gloss is the same lock-in and even more cognitive noise.

  The Archive? It’s boring, and that’s the point. Like a library card, it’s stable, quiet, and built to last. No surprise updates. No visual chaos. Just a calm space for thinking that’ll still be working decades from now.

  Minimalism is the feature. And when you’re building something meant to outlive trends, that matters more than glowing graphs or social clout.

  Thank you for your comment. I checking also Emacs solution but have not found any "easy" route in that path
• ctietze

  May 16

  @zettelkasten said:
  I checking also Emacs solution but have not found any "easy" route in that path

  That comes with the territory, I'm afraid (Try Spacemacs for a smoother experience.)

  Author at Zettelkasten.de • https://christiantietze.de/
• zettelsan

  4:55AM

  @zettelkasten said:
  Thank you for your comment. I checking also Emacs solution but have not found any "easy" route in that path

  With Emacs, unless it's already your tool, it's important that you don't overwhelm yourself with all the things that people tell you that you can do with it. The learning curve is definitely its biggest downside, but you can use Emacs in a minimalistic way if you choose to do so. I wish there is a good resource for non-programmer to learn from for those who wants to integrate Emacs to their PKM system.

  ---

  I'm also interested in getting familiar with how The Archive does things mainly because it's the baseline system many people in this forum use. Unfortunately I'm not in the Apple camp, so I don't have hardware to run it.

  That platform consideration is actually relevant to Emacs as well. I think that Emacs runs best in a Unix/Linux-like environment, so it may take a bit more effort to run it outside of that environment.