Zettelkasten Forum


Rethinking my use of UIDs

2»

Comments

  • @Sascha said:
    What are the old school elements that you are suspicious of possible modernisation?

    It's not suspicion (although I'm generally at chimp-with-abacus stage), more intimidation at the learning curve. I know there are a lot of resources out there to learn how to use KM etc. I just haven't found the time yet to dive in.

    Started ZK 4.2018. "The path is at your feet, see? Now carry on."

  • I continue to struggle with understanding the need for a UID as a reference. If it is occasionally necessary to change a title - I have yet to do that, but I don't discount the potential future need - then it would be a fairly straightforward process to use a search and replace operation to update both the title and any links. Three options that I know of that can search and replace across multiple files: BBedit, Keyboard Maestro (using the "For Each Item in a Collection" action) and, if the ZK is in Drafts, you can use a Drafts action without relying on third-party software. None of these are particularly difficult and all are fast.

  • @bjbarry said:
    I continue to struggle with understanding the need for a UID as a reference. If it is occasionally necessary to change a title […] then it would be a fairly straightforward process to use a search and replace operation to update both the title and any links. […] None of these are particularly difficult and all are fast.

    One problem with search and replace of textual information is that you need to be careful that you do not accidentally change normal – non-link – text in your notes that happens to consist of the same words.

  • I suppose that is possible, but it seems unlikely to me. However, if is a concern you could add a short trailing sequence that would make it even less likely. Example:

    "This is a title xx"

    The trailing xx does not interfere with the readability of the title and virtually guarantees that the issue you raised would not occur.

    If TA provided a sort capability based on date created and it provided a multi-file search and replace capability, then the argument for use of a UID would probably be a thing of the past. I guess my opinion on UIDs is shaped by the fact that Drafts provides both of those capabilities.

    For info, Sublime Text is another application that allows search and replace operations across multiple files.

  • @Phil said:

    @Sascha said:
    What are the old school elements that you are suspicious of possible modernisation?

    It's not suspicion (although I'm generally at chimp-with-abacus stage), more intimidation at the learning curve. I know there are a lot of resources out there to learn how to use KM etc. I just haven't found the time yet to dive in.

    Ah, I think we are the same. But I made my peace with my chimpness. :smile: Many times I thought that something could be improved but then I always remember that Luhmann did all on paper and yet was productive. So, after all, I don't mind. Even basic stuff works fine per hand with no automatisation. I produced my IDs per hand for a long time and were no more or less productive because of it. :smile:

    I am a Zettler

  • I've been a happy user of UIDs for a long long time, I've been using date-time based IDs for my digital photos since at least 2002. The reason for this is that it is a very robust way to ensure sortability (important when browsing photos). So it just came naturally to me when I started my Zettelkasten a few years ago.

    @bjbarry said:

    If TA provided a sort capability based on date created and it provided a multi-file search and replace capability, then the argument for use of a UID would probably be a thing of the past.

    Relying on the date-created information in the the file's metadata is actually very risky. Date-created is a file system feature that may be different between operating systems, and file systems. Operations such as copying files (which depending on the devices/cloud storage you copy to/from may alter the date created), emailing files, restoring files manually from crashed hard disks, etc may alter the date-created field – all this happened to me at one point or another in the past.

    I would go as far and say that using date-created from the file system is at odds with the plain text approach.

    I guess my opinion on UIDs is shaped by the fact that Drafts provides both of those capabilities.

    The thing is that Drafts does not work with plain-text files. Each draft has metadata fields attached to it which are stored in Drafts' database alongside the plain-text content of a draft. The date-created information is lost as soon as you export a draft to a file (Thats why my Drafts-actions for saving a draft to Dropbox automatically add the date-created to the file name ;-).

  • (I think I really cannot understand the problem many people on the forum have with the date-time-based UID. For me it's really a reliable friend, not a foe.

    One thing that @ctieze could do, would be to add a flag in settings to automatically hide everything up to the first space character in the note list. So that the UID does not distract or take away precious space.)

  • @rhubarb In The Archive, sure, but this is a conceptual qualm people have with Ids in general :)

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

  • @rhubarb I think its twofold. First I think a lot of people don't really understand the significance of TIDs and why they are used. Second, people aren't use to them or don't necessarily think about information in a temporal manner, so the TIDs serve as distraction to them (my case, which is why I'd like an option to keep them for functionality but hide them so they don't distract).

  • The suggestion that UIDs could be used, but hidden is, IMO, a very good one. Having one "thing" act as a token for another is not that unusual. Icons in the Dock are an obvious example. If I click on a Bear icon I'm not actually interested in a Grizzly Bear appearing on my computer.

    But, going further: I think that having a rock-solid way of ensuring that links are never broken is a very desirable feature. I would love to have that assurance. But, it is not the only desirable feature and sometimes you have to choose which desirable features are more important to you.

    For me, I want:

    (1) A system that is really good at ensuring that links are never broken, but not necessarily perfect. I don't change titles often and, if I have the ability to do a search and replace across multiple notes, then I am happy.

    (2) A system that is easy to use -- with sufficient automation that I'm not spending time figuring out how to do things (like creating links) that the software should do for me. In this regard, Drafts is significantly superior to TA, and getting even better rapidly.

    (3) A system that makes whatever investment I have made in creating a ZK pay off when I'm using my iPad or iPhone. All the major productivity applications I use: Ulysses, 1Password, Contacts, Fantastical, Drafts, Pages, Excel, Word, Photos, etc. all work on my iPhone. If a ZK has value to me then why would I not want it on my iPhone? I've tried using iA Writer. It works with TA generated files, but there is a significant downside. A title in the form: "UID This is a title" does not work on an iPhone - the horizontal space is insufficient.

    So, I don't think it is correct to say that people struggling with why they should use UIDs are not concerned with temporal information.(I make use of sorting by Creation Date in Drafts. Rather, creating a rock solid linking system (as opposed to a pretty damn good system) is a tradeoff that doesn't work for me.

    It is likely that this discussion hinges on how you use information. If your job or interests keeps you in front of a computer, then a Mac-only system with a big monitor (lots of horizontal space) may work for you. But, that is not a universal condition.

  • Sooooooooo (and the danger is here that my rather British sense of humour doesn't translate well, but here goes) - I've been playing around with Obsidian and I like it but....
    ...
    ...
    ... I don't like the fact it doesn't have the UID at the start of the note names 😵🙄 How ridiculous is that after all I've written on here?

    Why don't I like the absence of them? Because I can't easily see the order I created the notes in. Doh! Exactly what you've been saying all along. Obsidian does have the option to order by last modified (which I really like) but suddenly I can see the benefit in ordering by creation date.

    That said now I'm looking at TA, I think it's the font of the UID in the note list. I get why it has to be fixed width font, otherwise the text of the names wouldn't line up, but I find it's that which I find distracting and not the UID itself.

    But, yeah, sorry guys. I'll shut up moaning now and trust you more... 😉😉

  • @JKF We welcome you with open arms.

    I am a Zettler

  • @JKF said:

    ... I don't like the fact it doesn't have the UID at the start of the note names 😵🙄 How ridiculous is that after all I've written on here?

    I believe there is an option for that. Enable settings -> plugins -> zettelkasten prefixer. I haven't used it so no idea if it does the right thing.

  • GBCGBC
    edited June 2020

    Thank you: there is, although I've been using a keyboard maestro shortcut just to pop the timestamp into the filename as I create it.

    I'm still trying to get my head around what I prefer.

    I now fully recognise the value of (need of?) the ability to sort by creation date and / or see the creation date. The latter could be handled by including the timestamp in the body of the text, and I suppose a search for part of a date would reveal proximate notes. I would like the option to sort by creation date, for the reasons often cited on here, and I also do like the Obsidian option to sort by modification date. This covers the two user scenarios of wanting to find notes created in a similar time frame to each other, to track back on thought processes that were happening at the time, and wanting to find the notes last worked on (if you are the type of user who modifies notes) so you can see your recent thinking. So these, I guess, are the underlying requirements. Do they translate into having the UID in the title name by necessity? I guess that's a development question, which of course is what much of this thread has been about.

    Here's Obsidian with no UID in title - I do think this is cleaner looking, and focusses the eye on the title. But this, in Obsidian, allows only for sorting A-Z or by last modified date.

    Here's Obsidian with UID in the title. This now gives the benefits often discussed on here - the ability to sort by creation date, so in Obsidian I now have the option of sorting by creation date or by last modified date.

    Do I prefer this? No, I prefer the cleaner looking one. But I get the importance of the creation date ordering. But why do I suddenly not feel so visually distracted by the UID? For sure, it's the font. Here's the same view in TA:

    I personally find the monospaced font of the UID in the file name very visual distracting. I find monospacing very hard to read (and this is probably where I was going when I was talking earlier about the difference between developers and 'users' - developers tend to be extremely used to working with fixed width fonts but users tend to find them difficult to read). You'll note in the editing pane of TA I've chosen a proportional font for this reason, despite the advantages of (and warning signs recommending) a fixed width one. In addition, the UID visually dominates and, to my eye at least, overwhelms the note names in the note list because it looms large. It's harder to look past that and scan the note titles.

    So where did I end up after all that? I think the cleaner view of no UID is much, much better for visual scanning of titles, and much more 'friendly' to the non-tech user, but the ability to sort by creation date is important too. So if sorting by creation date requires a title UID, then I could put up with that. But it's extremely useful to have a user interface that helps the eye to look past them.

  • @JKF wrote:
    I'm still trying to get my head around what I prefer.

    Sometimes, it is not to adapt the system to you but to develop yourself in a way that you can adapt to a powerful system.

    1. Example: The learning curve to use Emacs was painful for me. I jumpstarting by a very long first session (from 3:30am to 19pm on a sunday). I am using Emacs productively not because I adapted it to my needs but because I accepted that I needed to change myself.
    2. Example: I interviewed over 20 professors and researchers when I was a student. When I asked about Luhmann and his Zettelkasten all of them said something along the lines: Oh, this is not my thing. Seems to much work. I have other preferences etc. Even the so called intellectual elites dismissed the best system that was available back then.

    Sometimes, the system is the bottleneck. Sometimes, oneself is.

    I am a Zettler

  • GBCGBC
    edited June 2020

    @Sascha said:
    to develop yourself in a way that you can adapt to a powerful system.
    Sometimes, the system is the bottleneck. Sometimes, oneself is.

    I agree... but only partially, if you're talking about the adapting to the UI / monospace font.

    Sometimes one's bottlenecks can't be simply developed away. UX/UI is a massive thing right now and it recognises that people have different innate abilities, neural wiring, needs, etc that can't just be self-developed away, and that in any case it isn't necessarily the best use of the user's effort. For me, developing skills in taking notes in a new way, and developing skills to link / connect / find notes sounds like productive self development, but learning to read a font and use a UI that really is distracting to my eyes - well, I'm not sure about the value of that in terms of time spent.

    Why?

    Because I've been exposed to fixed width fonts for nearly 30 years and they are still not something I read easily (for prose, at least: I can read code perfectly well, the visual flow is different) so I'm not sure there's huge value in me spending time to 'develop' myself in that area.

    So this brings me right back to a point I made a few days ago - it's quite a developer / techie answer to say 'the user can change themselves because this system is great': what will happen in practice is that potential users will say 'this really is not for me, I'll look elsewhere', and you'll end up with a smaller bunch of (extremely enthusiastic and capable) customers, all with a similar skillset / mindset.

    You might, of course, be perfectly okay with that: this might be exactly your product strategy but, if you're not already, you need to be aware and accept that some other people won't or can't adapt themselves.

    There's a danger these conversations can sound negative or critical. That's not my intention: I actually really like both TA (not withstanding my frustrations I've noted elsewhere on the forum) and the efforts / input / passion of you guys. What I suspect is that I'll use something like Obsidian until the process of taking / finding notes and making connections is embedded and I start craving, or needing, the additional functions of TA. I just wanted to highlight the assumption that users can change might not always be correct.

  • @bvs said:

    @JKF said:

    ... I don't like the fact it doesn't have the UID at the start of the note names 😵🙄 How ridiculous is that after all I've written on here?

    I believe there is an option for that. Enable settings -> plugins -> zettelkasten prefixer. I haven't used it so no idea if it does the right thing.

    The New Note button at the top of the File Explorer creates notes without a UID in the folder designated in settings -> Core -> File -> New note default location.

    The Create new Zettelkasten note button in the sidebar will start a new note with a date timestamp UID at the beginning of the title-filename in the folder designated in settings -> Plugins -> Zettelkasten prefixer. If a folder isn't set here it defaults to the Core -> File -> New note default location.

  • @GBC wrote:
    There's a danger these conversations can sound negative or critical. That's not my intention: I actually really like both TA (not withstanding my frustrations I've noted elsewhere on the forum) and the efforts / input / passion of you guys. What I suspect is that I'll use something like Obsidian until the process of taking / finding notes and making connections is embedded and I start craving, or needing, the additional functions of TA. I just wanted to highlight the assumption that users can change might not always be correct.

    Of course. No doubt about that. :smile:

    The question what the acceptable threshold of training age is. In my field (health and fitness), you cannot expect to build significant amount of muscle in the scope of days or weeks. Nobody expects that. But some people expect significant muscle growth in a quite suprisingly short amount of time.

    Translate that to knowledge work (in general, not using a Zettelkasten): I expect someone to become proficient in this domain if he spends half a year up to a full year if he practices it a multiple days per week for hours. There is very much implicit material to learn to feel that you improving in what you are actually doing. You might think that something in you approach is wrong and you even can name it. But many problem subside if you become better at identifying arguments, school of thoughts, improve muscle memory etc. When I teached some students they thought they needed to understand a particular software but they had much more problems under the surface and the issue with the software was just symptomatic.

    This has little to do with The Archive or the Zettelkasten Method. I am just doubtful that many people put enough training sessions under their belt. Most of knowledge work comes down to skill and not to knowing about it.

    A rule of thumb: After two to three hours about learning about the particular software, especially with something simple like TA or Obsidian, you should at least put 50 hours of practice under your belt.

    I am a Zettler

  • @Sascha I think - if I understand you correctly - you are talking about putting time into learning / practising the method, and I am talking about putting time into working with a user interface that I find distracting. Is that a fair observation?

  • @GBC said:
    @Sascha I think - if I understand you correctly - you are talking about putting time into learning / practising the method, and I am talking about putting time into working with a user interface that I find distracting. Is that a fair observation?

    No. In short, I mean: Take your time becoming comfortable with your solution. There are preferences and if you really cannot bring yourself to read prose in monospace fonts than it is like that.

    Partly, I am talking as if we are on a podium and a big audience is listening. You are saying that you need to tinker with the UI. I just gave an alternative view on the problem because quite a few people think that a solution needs to feel right immediately or even in a matter of weeks. In reality, one need to work for months (and not only once a week but hours and multiple sessions per week) to get comfortable. And even then, there is much hidden under the surface. For example, learning to work with knowledge in general.

    I am a Zettler

  • @Sascha oh okay. I think perhaps you are giving a general reply then, rather than one specific to what I’m talking about, since you’ll see from elsewhere on this forum that I’m putting a lot of effort into understanding how the system can work for me.

    I believe, though, that much of this thread has been more about about the UI implications of UIDs than a resistance to the system, hence my focus on that here.

  • @GBC I agree with your assertion that the "concept" is different from the "UI." I think in terms of three different meta-levels that need to be considered separately, although they strongly interact.

    1. The Theory - as discussed by Ahrens and in many threads in this forum.
    2. Personal Workflow - do you mainly interact with books or journal articles in the presence of your Mac? Or, do you conduct oral interviews, make observations in the field? Some combination? Do you make preliminary paper notes, dictate into your iPhone, both? Do you wish to make use of the ZK while in the field?
    3. A Digital or Analog System - TA being one obvious choice. Ideally, your system would reduce or eliminate all points of friction. Your comments re UID fit into this category and may be related to your specific workflow. That would be true for me because a title in the form UID+Title is unusable on an iPhone and only marginally usable on an 11-inch iPad.

    There are a large number of personal situations to consider. For example, I am not nearly as concerned as some for the long-term viability of any ZK I create both because the topics in which I am interested become obsolete relatively quickly - and because I'm 73-years old. As I recently told my dentist, I don't need a 40-year solution.

    However, it is important to me that I be able to use iOS devices and that forced me to find an alternative to use of a UID. That factor motivated my shift to Drafts because: (a) It has iOS versions, (b) It records and allows you to sort notes based on date created, and (c) If I need to change a title (and potentially corrupt associated links) I can use a Drafts Action to do a search and replace across the entire Workspace that contains a ZK.

    I agree with @Sascha that it takes a long time to become comfortable with any solution you choose. Personally, I created a first ZK over four months, and then decided I needed to start over from scratch.

  • @bjbarry said:

    >

    Thank you - helpful insights. I too need an ios solution - for now for reading I'm using notebooksapp (lovely experience although it tastes #tag and # header (with space) the same, and 1writer to a an alternative read / write tool.

    I agree with @Sascha that it takes a long time to become comfortable with any solution you choose. Personally, I created a first ZK over four months, and then decided I needed to start over from scratch.

    For what it's worth, I also agree with @Sascha that a tool takes time to master but I think I'm misunderstanding something in the discourse since that's not related to my point. For me, this thread passed through a point of various people saying we dislike the IUD in titles due to the visual distraction, to me (there's obviously other stopping place but these are the bits I remember) saying that I've tried obsidian and weirdly feel I need them / want them, to me then relaying that, for me at least, the font / UI of the note list in TA is what causes the distraction. So I've gone back to obsidian which is cleaner and easier to look at. That's pretty much it, so it's confusing that the conversation keeps shifting to the need to make a system work. It's a bit like me saying that the driver's seat in a car is too uncomfortable for me and the conversation keep returning just to the fact that it takes time to learn to drive.

    But perhaps I've missed the point somewhere...

  • That is to say, I know elsewhere (other threads) I’m talking about my difficulties getting this system to work, so maybe that’s confused things, but here I’m talking about the UI only. The skills to make it work for me are something I’m pretty committed to, but I’ll never find mono fonts easy to read.

  • @GBC I don't want to hijack this discussion with replies: could you open a new discussion with a request to change the note list, with some context? Then we can all address the problem properly :)

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

  • @GBC said:
    @Sascha oh okay. I think perhaps you are giving a general reply then, rather than one specific to what I’m talking about, since you’ll see from elsewhere on this forum that I’m putting a lot of effort into understanding how the system can work for me.

    Yeah, this is a tendency of mine in the forum. I am not just a user, so I often don't write specific to the topic or problem only, but sometimes use an opportunity to add to the topic. But:

    I believe, though, that much of this thread has been more about about the UI implications of UIDs than a resistance to the system, hence my focus on that here.

    I am thinking about your problem from the perspective of perceptual psychology for a while know. I can empathize with you because I just don't like Emacs. It is powerful, it does its job very well, but still: I am drawn to Taskpaper because of its cleaner look.

    I can't remember who long it took me personally to adjust to the block of numbers in front. But I observed that it is a common part of the learning curve to accept this additional object in your view and beginn to ignore it.

    Perhaps, you can modify a font in a way that numbers get narrower, smaller keep monospaced while the letters keep more pleasing to the eye?

    I am a Zettler

Sign In or Register to comment.