Question about object tags
Happy New Year!
I follow Sascha's advice to use object tags: https://zettelkasten.de/posts/object-tags-vs-topic-tags/
In the following note, I have used the object tags: "#good-technician #mediocre-boss #human-capital".
I often encounter this situation where I want to refer to a specific entity (e.g. a good technician), so I use adjectives to be specific about it. But I don't know if this is good practice. The other options I could use are:
- use two tags: "#good #technician".
- use only the core of the idea: "#technician".
What do you recommend me to do?
Because in this note, I just want to refer to good technicians (I don't want to talk about technicians in general). Therefore, I think that's the object to define with the tag. But I could be wrong :^)
# 202401060957 Turning a good technician into a mediocre boss destroys human capital #good-technician #mediocre-boss #human-capital In companies (public or private), one way to progress economically is to acquire responsibility for the work of others: to move from being a technician to being a boss. However, the skills that define a good technician are different from those of a good boss. This causes many good technicians (motivated by economic incentives) to become mediocre bosses: they do not have the skills needed to be good bosses. For the company, this situation destroys human capital value: we have turned a good technician into a mediocre boss.
Creative work doesn’t play by conventional rules · Author at eljardindegestalt.com
Howdy, Stranger!
Comments
I like the multi-level tags, progressively becoming more specific, rather than using simple tags. That way, you don’t end up with the same tag on too many zettels (which is not very helpful). Your list of tags also doubles as a useful index.
Maybe the level higher (more general) than "good technician destroys human capital" is "human-capital-destruction".
It can be a challenge to find the sweet spot of indexing and tagging. You want to find what you are looking for, but you also want it just a little bit "fuzzy" (higher level/more general/less granular) to get exposed to less-related ideas that have the potential to surprise you.
Searching for "good technician" brings you to this one note or two, searching for "human capital" takes you to 100 notes, searching for "human-capital-destruction" takes you to 10 notes which contain observations of other things that cause destruction of human capital such as pandemics, warfare, erosion of educational institutions.
@GeoEng51 Could you show us an example of the multi-level tags you use?
What I understood is using tags like: "#human-capital", "#human-capital-destruction", "#reversible-human-capital-destruction", etc
Thank you for pointing out the tag: "#human-capital-destruction".
One of the objects of the note is "human capital", but you might consider that I am not thinking of "human capital" in a broad sense. In this note, I am talking about its destruction, the object is then "human-capital-destruction".
The decision of which tag to use depends on what previous tags I have used (or how much I foresee myself discussing that idea). If I already have 100 notes on "#human-capital" (or if my goal is to get to that point), it makes more sense to use "#human-capital-destruction" (or "#good-technician" in the case of technicians).
Thank you for your time and ideas :-)
Creative work doesn’t play by conventional rules · Author at eljardindegestalt.com
Sure, good luck. Everyone will have unique tags depending on what they are thinking about or reading.
Keep in mind that organizing notes by topic, theme, or project can happen at any time along the way. As new projects or ideas come up, you can also organize by hub notes or project notes or new very-specific tags to generate a subset of a broader tag.
I do my Zettels on paper index cards, but have a searchable text file "card list" where I list the titles all of my "entry point" Zettels- over 400 now; often adding a keyword or hashtag next to the card title. I am also working on finding that sweet spot of just enough search returns to be helpful. Cheers!
For sure. One aspect of my Zettelkasten is capturing hints and tips for my engineering work, which includes work on dams, canals and reservoirs (I'm a geotechnical engineer).
So, in the early days, I had a simple tag: "#Dams", which was maybe used on 5 or 6 zettels.
As time went on, and the count for this tag got higher, this was split into:
"#Dams-Design"
"#Dams-Performance"
"#Dams-Maintenance"
Later, as the number of zettels with the first of the three tags above became excessive, I split that first tag into:
"#Dams-Design-Stability"
"#Dams-Design-Seepage"
"#Dams-Design-Consequences"
I haven't done it yet, but I can see the first of the above three tags becoming even more specific, such as:
"#Dams-Design-Stability-Seismic"
"#Dams-Design-Stability-Rapid_drawdown"
"#Dams-Design-Stability-Weak_foundation"
"#Dams-Design-Stability-Construction_defects"
etc.
I don't push for multi-level tags at the beginning; I create them when it becomes clear that I need them. If I only have 1 or 2 or 3 zettels using the tag "#Dams", then I hardly need ten different 3-level tags. One simple one will do. But as the number of zettels using a particular tag grows, (in my experience, more than about 5 to 10), I find it makes sense to split that tag into several tags with one more level of specificity.
If you follow this process, you will find your list of tags makes a wonderful index to all of your zettels.
Note that I am also a supporter of the idea of using Structure notes, but I prefer to create top-down structure using tags, which grow organically as described above. In my Zettelkasten, I have a few structure notes and a lot of (single and multi-level) tags. But sometimes it makes sense to have structure notes. One shouldn't be too "stiff" about his or her approach on this.
Hope that helps.
@FernandoNobel Check this thread, where I discussed with Cratermoon about it https://www.reddit.com/r/Zettelkasten/comments/10g5irl/comment/j5w3lp9/
David Delgado Vendrell
www.daviddelgado.cat
Thanks @GeoEng51 for your example. Now it is totally clear ;-)
So, the algorithm for using object tags is as follows:
I will now answer my self :^)
If you have few notes on an object, you can choose the level of specificity you prefer. This is an arbitrary decision that depends on your knowledge.
If you have many notes on an object, the level of specificity has to be high enough to ensure that the object tag is useful (avoid having 100 notes on the same tag).
So, in my particular case, both "#good-technician" and "#technician" are good object tags to start writing about that idea.
Creative work doesn’t play by conventional rules · Author at eljardindegestalt.com
I'd think about the concepts you are core to the idea. My tags would be:
Career and PeterPrinciple are implicit, Technician and Boss are explicitly mentioned.
Career and PeterPrinciple would refer to the mental models used in the background and are implied to be understood.
I wanted to write more, but my daughter decided that the night would be the best time to do a bout of high intensity intervals...
I am a Zettler
I'm also immediately thinking of
These tags are a level removed from the verbatim content, unlike
#manager #career, because in my field (programming/IT) it's a very typical (Peter Principle-ridden) path to discontent#good-technician, so I suspect they add more value in a search for management career paths. I wouldn't find anything in my notes about#good-technician. I would find something aboutgood technicianas a literal phrase search for some rough content, maybe, and in that case the tag wouldn't add much on top for me.Author at Zettelkasten.de • https://christiantietze.de/
Wow, I didn't know that Peter Principle existed. At least I am not the only one suffering it :^)
I like the idea of implicit objects, like #career and #peter-principle. I will look for them from now on. Thank you @Sascha !
I also agree with @ctietze that tags closer to verbatim content are more useful (as they are easier to remember to look them up).
So, my question is: Do you care how many times a tag is used? If you have 100 notes with the tag #technician, would you still use that tag for a new note related to technicians? Or would you go down the path of using multi-level tags as @GeoEng51 pointed out?
Creative work doesn’t play by conventional rules · Author at eljardindegestalt.com
Yes, because I want certainty which means consistency. If I had 100 notes, most likely I'd have a structure note and the tag would be used for filtering only and not to access notes.
So, no. I don't care for a certain reason: I almost never control the amount of notes per tag. If I do, I check if a structure note is needed.
I am a Zettler
This simplifies the use of object tags because then there is no problem with overusing a tag (100 notes with the same tag). In this situation, the tag will be useful for filtering the search, and the tag will indicate that a structure note may be needed to handle those notes.
I think you explained this "natural evolution" of tags in the workshop, but I forgot it, I'll go back to check the material :^) .
So, what do you think of the idea of generating a set of notes (with their respective object tags) collaboratively to serve as training for other people?
The idea is to hide the identified object tags so that the person who wants to train can guess them and then check the solution.
Do you think this is a good idea or too naïve?
Creative work doesn’t play by conventional rules · Author at eljardindegestalt.com
If it's too naive, I am too naive myself. This is exactly what I thought about to add to the upcoming member's area.
I am a Zettler
I have been struggling with tags for years, esp. trying to keep the number under control because it is very easy to create a new one just for that 1 important note, and surely I will use it again, but that often does not happen. So, I decided to make a list, and each time I wanted to create a new one I would check the list to see if there wasn't already one that could more or less be appropriate for that one note. But I still had a list of disparate tags that I was not happy with so I abandoned using tags altogether.
And then I discovered your reply above and I realized that that would be excellent for my workflow. I just finished restructuring my tags so I could arrange my list in groups of clusters. That in turn has encouraged me to start using tags again in a more sensible, logical manner.
Many thanks for your suggestion !!
P.S. I worked as a geotech geologist for a couple of years in an underground mine in Zambia a long time ago 🤝
@Ngungu
I'm happy to see that my tagging suggestion made sense to you. I thought, from the sparsity of feedback on the forum on that particular comment, that what I suggested was too simplistic or just not that appealing to others. It works well for me; my list of tags reveals the depth with which I have explored certain subjects and also acts as an excellent index to my ZK. The fact that any particular tag only connects to a few zettels keeps the index tight and thus useful.
I got the idea from previously using Bear, which allows nested tags. Many apps do not accommodate nested tags, but they can recognize multi-level tags, as I've described them.
I was an avid user of Logseq for a while, and still very much like that app, although I've moved back to another app I also like called NotePlan. Logseq had some other tagging features that were very powerful and opened up other avenues for tagging.
I've recently discovered that NotePlan allows nested tags; I'm researching how that works before converting my multi-level tags to nested tags. I may stay with the multi-level tag approach because of its simplicity and portability.
There are a few engineers on this forum; good to run into one who understands the geotechnical / geological engineering world!!
I find your suggestion elegant because of the refinement strategy: start with a broad term and break it down as needed.
Over the years I've also been experimenting with hyphens and slashs in tags. Hyphens are more compatible, but I do like how Obsidian displays nested tags. As a result I now have a mix of both. I still can't make up my mind. :-)
On the other hand I'm replacing more and more topical tags with wikilinks to topical notes (aka definition notes, encyclopedic notes, etc.) It works well with Obsidian's backlink features.
As far as topical tags go, I consider your suggestion the gold standard. :-)
I agree with this with one refinement. If you can control the formatting of the result list - I mean, that list of tag matches that gets too long - then insert a blank line between every six or seven lines. This makes it much easier to take in the results and you can postpone splitting the tags until you get even more cards. Here is an example; without the blank lines there would be too many hits to work with effectively:
Zettelkasten
The reason you don’t ‘get’ Zettelkasten
Zettelkasten Explainer
Zettelkasten and the Five Levels Of Thinking Support
The ZettelKasten at a Higher Level
The Zettelkasten at a Lower Level
Zettelkasten and Mind Maps as Implicit Topic Maps
A Zettelkasten Needs a Table Of Contents (toc)
A Zettelkasten is A Book And Its Index
References for zettelkasten design, use, and benefits
The Zettelkasten as a Topic Map
A ZK card should be a bundle
Zettelkasten-Topic Maps Duality
How Zettelkasten Work
The Zettelkasten Triad (Serendipity, Synergy, Syzygy)
Zettelkasten And Surprise
Dialogs: Zettelkasten and Others
Luhmann on Zettelkastens
Luhmann On Zettelkastens Distilled
Zettelkasten - Do The Hard Work
You might not have hit on the right level of abstraction yet. For example, the qualities it takes to win an election are not necessarily the same qualities that make for an effective leader once elected. Someone mentioned the Peter Principle - people rise to their level of incompetence. Also, many companies try to deal with this issue by have management and non-management tracks for people to move up. What is a more general concept that you have in mind?
The tag(s) you use also should be easy to search for, so you don't need to remember so much to get a successful match.
Only you can know what you are thinking of here, but often we can't quite articulate the concepts. That's fine, but it will affect how you tag. You should allow for that possibility. In this particular case, it sounds to me like you are thinking about "human capital" more than anything else. My mind started running like this:
Organizations Assets Human Skills Competence Enhancing Destroying Peter Principle Tracks Management Technical Physical FinancialSomewhere in there you could find a useful tag set. Your own breakdown will probably be different, of course. BTW, I found it very helpful to put these terms into an indented list like this. It helped me to clarify my own thinking as I worked on it. Creating a mind map can be another way to get clarity.
Once you've got the structure to your liking, you can flatten it out into a compound tag or tags. If you can keep the list around (maybe as a structure note) it will help you create more tags in this area as you need them.
Going back to @FernandoNobel's original post, we could have a closer look at the example. The title contains a claim about a consequence of promoting the wrong people to a leadership position:
Why is it so difficult to find useful tags? In my experience this difficulty is often a symptom of some thinking error in my note.
Instead of searching for better tags, I'd start by clearing up the note.
I like the example, because it's a well known issue in management. So what might be the problem with the note? I think it conflates too many issues. So this might be less an issue with tags, and more an issue with atomicity.
Here are some comments. The hashtags indicate topics that come into my mind:
Here we talk about a reason why employees seek promotion. They want to "progress economically".
#incentives #promotion #salary #benefits #job_satisfactionThat's the key argument. I would capture this claim on a separate zettel.
#qualification #skillset #technician #management #job_description #job_profile #competenceThe verb "causes" hides a flaw in the argument. Who promotes good technicians? It's not the technicians themselves, it's their employer. Why do employers promote incompetent people to leadership positions?
#career #management #employer #career_paths #management_decisionsThe Peter Principle has been mentioned before. The book of the same name is a good read. :-)
This adds yet another perspective. It talks about the (indirect) economic impact of bad management decisions for the whole company.
#management_decisions #human_capital #human_resources #profit #business_strategy #incompetence #organizational_cultureBut how exactly do such promotions destroy human capital? That's a big hole in the argument, that might require additional zettels like:
A company's management might look at these costs and decide to find other ways for good technicians to "progress economically". It might be in everybody's best economical interest to let good technicians do what they do well. :-)
I find in-text hashtags useful to find flaws in my argument. Adding hashtags to individual sentences or paragraphs is an intermediary step. Once the zettels become clearer, I tend to remove some of the hashtags or replace them with better fitting ones.
It's a bottom-up approach that works best with digital notes. I don't mind renaming tags across all my notes, when I change my mind.
That's different from top-down approaches and/or controlled vocabularies.
@GeoEng51 I used Logseq for a few months in 2023 but when I opened some notes in Obsidian I was horrified to see the mess. Every bullet point and note has a multi-multi-digital alphanumeric ID, and links to blocks/bullet points or notes are links to those IDs.
There is no way to make sense of all that in another markdown editor, except where it concerns text. What I am saying is that, to me Logseq's portability is close to, if not, at zero, i.e. you are effectively locked in.
I shall even go a step further. Obsidian is an ace markdown editor with a tremendous number of useful features, which, in combination with its now 1000s of plug-ins, makes it one of the best markdown editors out there.
I was using some of all that enthusiastically, notably embeds, canvas and certain plug-ins. But based on my Logseq experience I took a 2nd look at Obsidian and realized that all that fancy stuff in effect reduces portability too.
Sure, you can refrain from using embeds, but if the feature exists, the temptation to use it is great — that is what happened after I walked away from Logseq back to Obsidian: I was going to refrain, but the temptation got the better of me.
So, I decided to eliminate the temptation and use Typora instead: no more canvasses, no more embeds, no more automatic heading numbering with a plug-in, no more etc. Instead of embedding a blockB from Note2 below blockA in Note1, I place an ID (format: §HHMMSS) at blockB, then below blockA place a link to Note2 and in the alias put the ID.
This way humanly readable portability to any other markdown editor is ensured.
The only non-markdown I indulge in is inline HTML, such as
1<sup>st</sup>, or<span style="font-size:12px">text</span>, or coloring part of a text. Even in a non-HTML-supporting app this is still humanly readable.I am satisfied with this set-up because, having used many different markdown editors on the same set of notes, I came to realize how disparate they all are and therefore how important portability is.
@Ngungu I didn't have the same issue with Logseq because I used very little linking to individual bullet points (I'm assuming that's where your problems arose). I could read my Logseq files using either The Archive or Obsidian. I left Logseq because they were moving to a database model for their files rather that plain text. I never got excited about Obsidian - it was too complicated and bloated for my tastes.
I get your desire for simplicity in the text files, which I find is respected in both The Archive and NotePlan, and in the apps themselves. But if you're not in the Mac universe, neither of those are options.
In the end, we have to find what works for us and provides a minimum of friction, and that will vary from person to person!
@GeoEng51
I thought they were going to run that version in parallel with the standard version, leaving it to the user to choose.
No, I am not in the Mac universe, I am in the Linux universe and would not want to give that up for anything else.