Zettelkasten Forum


Two images tell you a lot about graphic views and Zettelkastens that are actually used

2»

Comments

  • @matti said:
    @Sascha That is super weird.

    @matti said:
    You do not believe that "What I believe still holds is that well designed graph views are powerful if the note collection is richly linked and organized because it will display connections in a way that a flat hierarchical view cannot."

    versus

    @Sascha said:
    This is not the original claim. The original claim is:

    What I believe still holds is that well designed graph views are powerful if the note collection is richly linked and organized because it will display connections in a way that a flat hierarchical view cannot.

    Hö? Are we seeing the same thing? Didn't we both just copy and paste the same sentence?

    No. If you go back to all of your quote:

    In what way do you disregard those oposing premises (the alternate view can be clarifying/relevant/have more power than the original)?

    This is already diluting the original claim.

    Regardless of that: So you reject that claim because it is expressed as hope? Even though the sentence structure expresses a causal relationship? And your premises are not incomplete because a counter example exists to the original claim? Is that the meaning of "not incomplete"? All of this feels like a spirit of the law/letter of the law thing to me. And I wish you'd be more a spirit of the law guy, I guess. ;)

    It's not the hill to die on for either of us, I'm sure (I hope, I guess), but I would've wished for a little bit more "yes and…", instead of "computer says no", especially without explaining it properly: "I don't want to engage with claims expressed as hopes. Can you elaborate and/or give some evidence?"

    There is neither evidence nor a valid argument to make the case for the graph view's ability to provide any substantial benefit. Even more, there is just one complex ZK to test the effects of the graph view.

    I gave a demonstration on what happens if you try to view a complex ZK graphically. The valid lines of thinking went in the direction of using the graph view for limited part of the ZK. Nothing more is to be found.

    So, what are you expecting me to do if there are neither arguments nor (more important) evidence present I can engage with? Or: If you have any valid argument or (more important!) evidence, please share them with us.

    I am a Zettler

  • edited November 2020

    @Sascha said:
    This is already diluting the original claim.

    Aha. That is your point. Alright then. I read into the original claim that there is potential along the lines of a filtered down graph - which is based on a graph of the whole notes system nonetheless. I gave an example: Discovering nth-order-links. There are others in this thread. If they are or are not exclusive to a graph view is perhaps hard to prove - nonetheless I think it's valid that a graph of your notes offers some value. This is maybe a generous reading of the claim, but it's a possible one - which means that the original claim is surely not invalid. I at least take it that you can also see that filtered graphs are not completely useless at least to some people and/or in some situations. And you say so yourself:

    The valid lines of thinking went in the direction of using the graph view for limited part of the ZK. Nothing more is to be found.

    Well this little bit of "nothing more" is exactly where people including me see the potential.

  • I at least take it that you can also see that filtered graphs are not completely useless at least to some people and/or in some situations.

    No, I don't.

    As a user with skin in the game, I filter with no mercy. I got no time for ineffective tools. If you go with an attitude of "something offers some value" or "options never hurt" you'll run exactly in the direction of feature bloat. (Which is already looming with quite some tools out there..)

    I am a Zettler

  • @Sascha said:
    you'll run exactly in the direction of feature bloat

    Not necessarily. I have also some skin in the game AND an open mind. Maybe that's the difference? Options - in the form of plugins for example - are doing the opposite of hurting.

  • @matti said:

    @Sascha said:
    you'll run exactly in the direction of feature bloat

    Not necessarily. I have also some skin in the game AND an open mind. Maybe that's the difference? Options - in the form of plugins for example - are doing the opposite of hurting.

    Options are the exact thing that made word processors an increasingly bad writing environment. Distraction free writing is exactly productive because of the lack of options.

    How were those pictures made? Because I don't give any tool a chance? Or, perhaps more likely, because I experimented with graphical interfaces (TheBrain was one of them by the way) for years and am still testing any tool I encounter for my sake and my clients sake?

    The only demonstration in favor of a graphical interface in this thread

    1. is not demonstrated with a complex ZK.
    2. didn't use an atomic appraoch.
    3. stated its benefit as obscuration of the relationship and content. (To quiz herself)

    There is no rational case for graphical views for complex ZK.

    I am a Zettler

  • @Sascha said:
    Options are the exact thing that made word processors an increasingly bad writing environment. Distraction free writing is exactly productive because of the lack of options.

    Well badly designed options are never great. VS-Code has a shit-ton of plugins. As has emacs. As has vim. I can see the possibility for an IDE for notes. You can't. Your example is Word, which is in itself limiting to say the least.

    How were those pictures made? Because I don't give any tool a chance?

    You can test all you like, your premises and rhetoric are way too strong to allow for anything else than disappointment. There is a difference here in the approach of observing the new. I get the impression that you have enormously high standards which makes it impossible for you to go from a plateau through a valley up unto a higher plateau, so to speak.

    There is no rational case for graphical views for complex ZK.

    There is. Take your complex notes system and filter it down appropriately. What are you talking about?

  • edited November 2020

    @matti said:

    @Sascha said:
    Options are the exact thing that made word processors an increasingly bad writing environment. Distraction free writing is exactly productive because of the lack of options.

    Well badly designed options are never great. VS-Code has a shit-ton of plugins. As has emacs. As has vim. I can see the possibility for an IDE for notes. You can't. Your example is Word, which is in itself limiting to say the least.

    I used IDEs for the ZKM and teached the use of them for the ZKM for years. Please, don't make any assumptions. And: For the ordinary user Vim or Emacs is not within reasonable reach. Both a great examples for my case.

    How were those pictures made? Because I don't give any tool a chance?

    You can test all you like, your premises and rhetoric are way too strong to allow for anything else than disappointment. There is a difference here in the approach of observing the new. I get the impression that you have enormously high standards which makes it impossible for you to go from a plateau through a valley up unto a higher plateau, so to speak.

    My opinion is the result of a decade of testing, tinkering, using and teaching the method. You are speculating about issues I tested half a decade ago exensively with great hope for a powerful tool.

    There is no rational case for graphical views for complex ZK.

    There is. Take your complex notes system and filter it down appropriately. What are you talking about?

    I did it and tried to make it happen for years. You see a filtered graph with only two nodes depth in this very post which has not even all the internote links shown. Even two nodes deep it gets messy.

    Sadly, the only demonstration of complex note system is my own, yet. So please, you have every opportunity to show a case of practical knowledge work that is done with a graph view.

    Hopeful speculations are not rational.

    I am a Zettler

  • edited November 2020

    @Sascha said:
    I used IDEs for the ZKM and teached the use of them for the ZKM for years. Please, don't make any assumptions.

    I know. I have read the forums. Where is the assumption here?

    Both a great examples for my case.

    Aha. How?

    My opinion is the result of a decade of testing, tinkering, using and teaching the method.

    Whereas my opinion has just developed yesterday? A result of your experience is a blindness to the potential of graph views. Amount of experience doesn't necessarily equal observational correctness or good/just opinions.

    You see a filtered graph with only two nodes depth in this very post which has not even all the internote links shown. Even two nodes deep it gets messy.

    Sure. Then filter more. Your scale doesn't fit the requirements, yet. "I cannot navigate the metropolitan area of berlin, with this map. Maps are useless! I have already zoomed in to show the whole country of Germany. There is nothing more I can do. Why are people still using maps. Does nobody see that you can't navigate using maps?!"

    Sadly, the only demonstration of complex note system is my own, yet.

    There is no need for that. Also, ask yourself why the other demonstrations actually got disqualified - by you no less. A properly filtered down complex graph will become more simple and will look somewhat like the examples in this thread.

    Hopeful speculations are not rational.

    Rhetoric is no substitution for substance.

  • edited November 2020

    @matti said:

    @Sascha said:
    I used IDEs for the ZKM and teached the use of them for the ZKM for years. Please, don't make any assumptions.

    I know. I have read the forums. Where is the assumption here?


    I can see the possibility for an IDE for notes. You can't.

    Both a great examples for my case.

    Aha. How?

    Those are fringe apps for the geeks that are avoided by the normal people. The vanilla installation is strange to normal people. It is a hassle to tinker with the universe of plugins. Actually, it is cliché that those apps overwhelm people with options and no orientations.

    My opinion is the result of a decade of testing, tinkering, using and teaching the method.

    Whereas my opinion has just developed yesterday? A result of your experience is a blindness to the potential of graph views. Amount of experience doesn't necessarily equal observational correctness or good/just opinions.

    So, tell us about your experience with the Zettelkasten Method. Did you teach it and learned about the issues of beginners, their specific learning? Did you write books or articles as a result of your Zettelkasten? How did you tackle the problem of complexity (successfully)?

    You see a filtered graph with only two nodes depth in this very post which has not even all the internote links shown. Even two nodes deep it gets messy.

    Sure. Then filter more. Your scale doesn't fit the requirements, yet. "I cannot navigate the metropolitan area of berlin, with this map. Maps are useless! I have already zoomed in to show the whole country of Germany. There is nothing more I can do. Why are people still using maps. Does nobody see that you can't navigate using maps?!"

    Then demonstrate a correctly filtered graph and its effectivness.

    Sadly, the only demonstration of complex note system is my own, yet.

    There is no need for that. Also, ask yourself why the other demonstrations actually got disqualified - by you no less. A properly filtered down complex graph will become more simple and will look somewhat like the examples in this thread.

    Do you have anything to show or demonstrate? Something in the real world?

    Post edited by Sascha on

    I am a Zettler

  • edited November 2020

    @Sascha Alright, alright. I have not written books, articles or taught the method™️ to anyone - I held a pretty well received presentation in a research seminar ("Forschungskolloquium") however. But regardless: I have and am writing notes since forever - at least since the beginning of my twenties - so around 15 years (if we count paper notes add another 10 years, I was a big journaler back then)?

    I have written on this forum in some places fairly recently that one important thing when tackling complexity is mindfulness. Understanding the limitations (EDIT: and the costs!) of solutions in general. Being fine with uneven quality of notes, adding value to notes, etc. Expecting less is a great way to gain more from a complex note system - in my humble experience.

    Then demonstrate a correctly filtered graph and its effectivness.

    I would love to (not really, but read on). But after Evernote 10, my notes system is in disrepair. I am slowly moving my notes over (to DevonThink and Obsidian), but this will take time. And as I said before this argument is not a classy enough hill to die on for me.

    I just played around with what I have already outside of Evernote - some 1400 notes and Obsidian is lacking surely some features when it comes to the graph view. But imagine if you could filter out notes by dates (creation, updated), tags, words, sources, etc. You'll end up with a much smaller graph to view. I tried to emulate this by just searching for difference ("differenz" in German) in my notes:

    Now imagine further you show and hide neighbours as different contexts become relevant, different notes are selected. But even already through this simplistic filtering I found some orphan notes that ought to be connected and I had some ideas about opportunistic filtering like this - maybe there should be a plugin to show randomly filtered down graphs? Some people like getting random notes displayed, so why not random graphical views? That's the result of engaging with my notes on that level for 10 seconds.

    In any case, I see already now and in the future a lot of potential to connect with my encoded knowledge here.

    First point last:

    It is a hassle to tinker with the universe of plugins. Actually, it is cliché that those apps overwhelm people with options and no orientations.

    All true and still some people are using them fairly successfully. I personally am a fan of VS Code. Yeah, it took a while to set up and so on but now I feel like I have everything I need. And if I feel like it's not like that anymore at some point, probably somebody else has provided it in form of a plugin. And if push comes to shove, I can write that plugin myself. Overwhelming/powerful are probably somewhat two sides of the same coin. Try to see both.

    And that something is "fringe"… come on. I guess you might care because some of your clients are unable to reap the benefits of more complicated tools? But this actually means nothing in terms of potential of the tool itself. It's just that not all potential can be realized by all people. I'll never be a great real time strategy player for example. That shouldn't mean that an eSports friend of mine would need to think that the RTS genre as a whole has failed to deliver. No. Such is life. And that's all.

  • edited November 2020

    You may read Dietrich Dörner's "Logik des Misslingens". Even very simple graphs with very few items and just two types of connections are hard to read. Even experts have a very hard time to make sense of seemingly simple graphs.

    And that something is "fringe"… come on. I guess you might care because some of your clients are unable to reap the benefits of more complicated tools?

    To this day, I never witnessed any connection between the complicateness of the tool and the productivity of the user.

    https://rosstraining.com/blog/2009/04/another-vote-for-simplicity/

    If you have any real world use cases that can show fallacies in my line of arguments I am happy to entertain them.

    Post edited by Sascha on

    I am a Zettler

  • Deep breathe, everyone!

  • edited November 2020

    One last thing (not engaging with the topic anymore - even though there would be things to say - but I accidentally made an error and it eats me up from the inside):

    I have and am writing notes since forever - at least since the beginning of my twenties - so around 15 years (if we count paper notes add another 10 years, I was a big journaler back then)?

    Add 5 years. Not 10. I wouldn't count what I was doing when I was 10 to my note taking experience. I am 34 now. I have been taking notes and journaling more or less regularly since when I was 15.

    Phew. Now I can look into the mirror again.

Sign In or Register to comment.