"Personal Knowledge Management Is Bullshit"

edited March 16

An article I started reading a while back and just revisited today while working through some notes in my zettelkasten. The link is to the annotated version in my Evernote account but also contains the link to the original.

I'd be interested to hear what others think about the argument(s) being made in the article as well as my reflections on them:

https://www.evernote.com/shard/s170/sh/d69cf793-1f14-48f4-bd48-43f41bd88678/DapavVTQh954eMRGKOVeEPHm7FxEqxBKvaKLfKWaSV1yuOmjREsMkSHvmQ

• @jameslongley: I skimmed your notes on this polemic. You spent much more time with it than I would have! I wonder why you felt the need to engage with it?

It's part of a recent genre of texts against personal knowledge management. There is an analogous genre of writings in this forum, for example, the discussion: "Zettelkasten as a coping mechanism (or: why I abdicate the zettelkasten method)" (April 2020).

Another text in this genre that I recently read is Doug Toft's blog post "Why I gave up on personal knowledge management — and what I do instead" (February 2023). Doug gave up a personal knowledge base in favor of "one Big Ass File", a topic that also has been discussed in this form, naturally: "Zettelkasten vs One Big Text File" (November 2020). (My response to the One Big Text File solution: if it works for you, great, but I'm not going back to my habits from the early 1990s.) At the end of his post, Doug said he is "fascinated by the growing backlash against PKM", and he listed other similar texts.

A common theme in many of these texts is the search for something simpler, and that's valid if you have simple requirements. But if your mission is to synthesize complex knowledge (which some call "knowledge synthesis" as we recently discussed in "Discourse graph and Zettelkasten") over a long period of time, especially in a way that meets the intellectual standards of science or Wissenschaft, then it would be wise to have a system of personal knowledge management with the requisite complexity. Realistically, what would be the alternative?

• I don't understand the argument in the original article. It just sounds like a lot of angry words. Perhaps the one thing I do agree with is I am not sure for the utility of the graph view in its current incarnation. Perhaps something more of a dynamic view - Ie: show me all notes tagged with #somethingamazing and view the connections.

As you mentioned in your summary notes - knowledge compounds, why lose it?

Also I think the author confuses exploration of ideas vs categorizing. How can you know the value of an idea or its place in a hierarchy if you do not understand the overall structure - because you are learning. Hence the messy work of incremental formalization and organization and lots of messy notes.

On a personal note, I would be quite lost and frustrated if I had to repeatedly gather the same knowledge because I forgot it. My 20 odd years of poorly organized but very searchable work notes saves me a great deal of time.

• edited March 17

As a reader, one cannot do much with an article like that. But it sure feels good to be heard The things written about graphs, the signalling effect of sharing them, and combinatorial explosion resonates with me and I captured this reference accordingly, but I'd actually have preferred a more academic backing of these claims. Now I have to do that work

For what it's worth, during the process, one gardening metaphor imposed itself: pruning. Relevant section:

> After only two weeks, my knowledge graph was utterly unintelligible and distressing.[#20230317pkm][]

The "combinatorial explosion"[#20230317pkm][] of connecting things diligently made the graph useless to traverse:

To make use of it, you need to keep the growing tree of knowledge well-pruned to stimulate a plentiful harvest. When users are encouraged to connect everything to anything and abdicate the role of the gardener who selects and plans, they'll be facing a jungle, an unintelligible mess of intertwined _stuff._

[#20230317pkm]: Justin Murphy: "Personal Knowledge Management is Bullshit" (2022-04-27) <https://www.otherlife.co/pkm/>

Post edited by ctietze on

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

• A

That most of the variance between people is genetic is an empirical claim founded on poor (=no) evidence.

Even if this is correct, it is part of ones maturing process and moral duty to be able to cover the full temperament spectrum. Meaning: An industrious person needs to learn to relax, a person with low industriousness needs to develop habits and grit.

The claim that you barely will be able to adopt skills and habits from the opposite of your temperament traits is equally poorly founded claim.

B

The article itself shows that one highly benefits from knowledge work training. In that case, being able to understand when you make empirical claims which need evidence.

C

I agree with the sentiment that a huge part of the social media (Twitter etc.) activity is driven by the signaling. However, the vast majority of the activity on Reddit, this forum and the Obsidian forum is driven by people who sincerely try to understand and improve.

The signaling is mostly done by content creators who give in to the temptation of the attention you get by those kind of signals. I cannot estimate the proportion because I don't follow other content creators dilligently since most of the content is either published for beginners or people who publish as they learn. But intuitively the proportion of people who are sincere is high.

The graph view for big archives is still poorly understood and people trying to figure it out. So, it is natural to either overshoot or undershoot in the way to communicate.

And: If you actually like what you are building and have an inclination to share your feelings online it is only natural to post images that are expressions of those feelings with no intention to actually further the collective thinking. The only ones who need to be careful with those feelings are the ones who have upsides for sharing this (professional content creators) since you create an environment for yourself that incentivises you to monetise your abilty to invoke certain emotions that are distracting.

D

It might be a little bit personal but this article is a nice example of what is lacking in many articles (or podcast): time spent with the matter. Almost anything takes years to learn and to develop a level of skill and understanding.

It took me, for example, years to even be able to publish something about organising knowledge. It took more years to actually scraping a draft together. My years were spent with obsessively working on and with my Zettelkasten. Obsessively means that I worked on the edge of my capabilities. The first book I processed diligently was [The Perfect Health Diet] by the Jaminets. I examined every footnote and cross-referenced every reference (more than 600 at that time). For years, I worked up to 100 hours per week. From these 100 hours approximately 60 were spent with my Zettelkasten.

To me, this is necessary to feel justified confidence in what I am stating.

These kind of articles are written within hours after minutes or inspiration and insight.

E

Or simply: The same argument could be made for any invention.

1. The plow wouldn't be invented. The best farmer's didn't spent time to invent stupid tools but brute forced through the soil.
2. Note-Taking would be invented. The best thinkers didn't spent time to invent stupid scribbles but brute forced through the matter by just memorising.

Innovation takes time, shitty prototypes and hitting some dead ends during its process.

I am a Zettler

• @Sascha said:

The article itself shows that one highly benefits from knowledge work training. In that case, being able to understand when you make empirical claims which need evidence.

I was thinking this also, but I was too nice to say it.

• Very good points from everyone. The only thing the author of that article managed to convince me of, is that he doesn’t know what he’s talking about.

Would he suggest that people who aren’t naturally orderly shouldn’t organise their notes in any way? Or just toss them immediately rather than keeping them? Or not to take notes at all?

And I found it sad that anyone expects hierarchy and insight to automatically appear from using a software. No wonder he was disappointed. The act of linking knowledge is useful only when deliberate.

I would also like to stand up in defence of the knowledge graph visualisation. I do like to have the graph, but I couldn’t put my finger on why for a long time. I agree that the full view of the graph has mostly only aesthetic value, but I find that zooming in locally provides a quick overview of which notes I have on a given topic, how they relate and which other topics they branch to. It’s mostly a way to remind myself of the context for a note without having to navigate around within the text. Crucial? No, but nice to have.

Anyway, PKM is not for everyone, and that’s ok, but that article wasn’t particularly well backed up by facts.

• @Sascha said:
My years were spent with obsessively working on and with my Zettelkasten.

This resonates with me. Just the act of working with my Zettelkasten (highlighting in the text I'm reading, writing notes and associating those notes to other existing notes) forces me to really interact with something that I'm learning, instead of just glossing over material and forgetting it all in three months time. Ironically, putting it in a note helps me to remember stuff much better, up to a point where I might not even need the note anymore. Having the note afterwards and being able to cool stuff with the relations around a note is a bonus. I already have the pay-off from just writing the note.

@Nori said:
I find that zooming in locally provides a quick overview of which notes I have on a given topic, how they relate and which other topics they branch to.

I think I mentioned this in another thread on visualisation, but I like the graph as well in a localised context. Having a graph with all incoming and outgoing notes for the note I'm looking at helps with remembering the context of the note. And it will show if those notes are related to each other as well, giving hints as to how notes are clustered around the note I'm looking at.

• I find that these sorts of articles against the variety of practices have one thing in common: the writer fails to state a solid and realistic reason for why they got into it in the first place. They either have no reason "why" or, perhaps, just as often have all-the-reasons "why", which may be worse. Much of this is bound up in the sort of signaling and consumption which @Sascha outlines in point C (above).

Perhaps of interest, there are a large number of Hypothes.is annotations on that original article written by a variety of sense-makers with whom I am familiar. See: https://via.hypothes.is/https://www.otherlife.co/pkm/ Of note, many come from various note making traditions including: commonplace books, bloggers, writers, wiki creators, zettelkasten, digital gardening, writers, thinkers, etc., so they give a broader and relatively diverse perspective. If I were pressed to say what most of them have in common philosophically, I'd say it was ownership of their thought.

Perhaps it's just a point of anecdotal evidence, but I've been noticing that who write about or use the phrase "personal knowledge management" are ones who come at the space without an actual practice or point of view on what they're doing and why—they are either (trying to be) influencers or influencees.

Fortunately it is entirely possible to "fake it until you make it" here, but it helps to have an idea of what you're trying to make.

No piece of information is superior to any other. Power lies in having them all on file and then finding the connections. There are always connections; you have only to want to find them. —Umberto Eco

• @chrisaldrich said:

Perhaps of interest, there are a large number of Hypothes.is annotations on that original article written by a variety of sense-makers with whom I am familiar. See: https://via.hypothes.is/https://www.otherlife.co/pkm/

@jameslongley, I think you just found your scene right here at Hypothes.is. I found the format of these Hypothes.is notes to be much more readable than the notes on the same topic in Evernote.

This note wins the prize for succinctness: "citation fuckin' needed, my man"!

• @chrisaldrich said:
I find that these sorts of articles against the variety of practices have one thing in common: the writer fails to state a solid and realistic reason for why they got into it in the first place. They either have no reason "why" or, perhaps, just as often have all-the-reasons "why", which may be worse. Much of this is bound up in the sort of signaling and consumption which @Sascha outlines in point C (above).

I think that is very underappreciated. People go into information management tools with a vague notion of what they are organizing the information for. So they build a system with no feedback mechanism and don't produce anything of note, then get confused why that isn't happening.

• @Nori said:
The only thing the author of that article managed to convince me of, is that he doesn’t know what he’s talking about.

Yes, but he did it so well.

The snarky title of the article suggests he is capitalizing on contrarianism at least as much as research or insight.

My personal use of knowledge management is pedestrian compared to the heavy hitters in this forum. Even so, my life would crumble if not for curated information.

• edited March 18

The author's remarks on writing can fairly be called fatuous:

The point of writing—and what the greatest authors have always done—is to cut through the knowledge graph with a bold and forceful line.

Almost anyone you know and respect as a writer is known and respected because they're able to muster brute, linear willfulness.

Linear as opposed to an interconnected, non-linear graph. I never cared for the buzzword "personal knowledge management," which sounds like calling a garbage collector a sanitation engineer.

The author of this piece runs a media company and a blog for ronin scholars, among other ventures.

Post edited by ZettelDistraction on

.Erdős #2. GitHub

• @Andy said:
@Sascha said:

The article itself shows that one highly benefits from knowledge work training. In that case, being able to understand when you make empirical claims which need evidence.

I was thinking this also, but I was too nice to say it.

You wouldn't imply that @Sascha wasn't nice. :-) "a sociologically adequate account of kindness needs to recognise the ways in which it is socially embedded and differentiated"
(Brownlie, J., & Anderson, S. (2017). Thinking Sociologically About Kindness: Puncturing the Blasé in the Ordinary City. Sociology, 51(6), 1222–1238. https://doi.org/10.1177/0038038516661266; p.1222)

• @Andy said:
@chrisaldrich said:

Perhaps of interest, there are a large number of Hypothes.is annotations on that original article written by a variety of sense-makers with whom I am familiar. See: https://via.hypothes.is/https://www.otherlife.co/pkm/

@jameslongley, I think you just found your scene right here at Hypothes.is. I found the format of these Hypothes.is notes to be much more readable than the notes on the same topic in Evernote.

This note wins the prize for succinctness: "citation fuckin' needed, my man"!

That's classic 😆

• edited March 25

I extracted a note for the "garbage in, garbage out" concept, and found I had used it in a note related to Zettelkasten, which fits the cited article, but then contains a warning that also fits the culture of internet discussion

# 202101110913 Garbage in, garbage out in a Zettelkasten
#zettelkasten #orthodoxy

When you feed garbage to your Zettelkasten, you can only retrieve garbage later.

This concept is known as "Garbage In, Garbage Out" (GIGO).[[202303250922]]

That could be an explanation why for some people the Zettelkasten Method doesn't work. The expectations might be that adding any information and associating freely and without constraint will suddenly create a beautifully structured paper, while it actually only produces (what was put in:) a wildly associated mess.

**Heads up:** this might turn into orthodoxy and zeal: "Oh it didn't work for you because you were using it wrong, it's your fault!" -- Ultimately, yes, it's  the user's responsibility, because the tool and method cannot do anything on their own, but it's also a problem of didactics (proper expectations, and how to use a Zettelkasten).


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