1/ A lot has been said about @tana_inc in the last week.
I've been playing around with it for a few days, and the hype is not unfounded.
I scratched together my first impressions & comparisons, along with musings from older tweets on what this means for knowledge work in teams.
2/ First, a word on sticking with your current tool.
There's a lot to be gained from doubling down on one tool, but there's no harm in play 😉 I'm already seeing benefits cross-pollinating into Logseq - the tools complement each other well.
3/ Let's start with problems rather than tools, as software exists to solve user challenges.
PKM users require frictionless input, organisation & retrieval of information to gain mental clarity, generate original output or improve outcomes in their field
Tana is a collaborative information management tool.
Using other tools as mental models, you have the building blocks to build custom workflows like Notion, although like Logseq/Roam, it's an outliner built on a graph DB.
Tana leads as a graph DB in 2 areas:
6/ Firstly, 'Organising'.
The supertag feature is a major innovation. Instead of a tag which simply states 'this note/block is related to this', supertags allow you to add multiple levels of structure & resolution to that relationship with user-defined fields.
Your fields 'Project' & 'Attendees' automatically pop-up, allowing you to easily add the necessary details. This facilitates consistent structuring of information.
8/ Properties in Logseq allow for similar structure, but in Tana supertags are core to the implementation.
They're elegant, intuitive & the UI is gorgeous.
You can also display/manipulate multiple levels of information for multiple nodes simultaneously.
9/ Side-note: The ease of manipulation feels trivial, but is an important detail.
I've been jerry-rigging project mgmt workflows in Logseq, & whilst they do the job, the inability to change information across multiple blocks at once has been frustrating.
10/ The other dimension where Tana shines is collaboration.
Whilst you can use Tana for PKM, I'm thinking about it more for 'collective knowledge management'. It serves as a great team hub, combining knowledge repositories and task mgmt/workflow orchestration in one interface.
11/ Think of it as a blank canvas, allowing you to build whatever your team needs, e.g. bug tracking, OKRs, product backlogs, documentation, even CRMs.
Importantly, you can cleanly separate work and personal information, and seamlessly share information between your workspaces.
12/ Beyond the team environment, @tana_inc is excellent for collaborative research and knowledge synthesis.
The ability to natively structure ontologies in a multi-player 'tool for thought' is a step-change in building discourse graphs.
15/ There's more to be said on Tana and how it compares with other programs (esp. Logseq in my case), but I will make a more detailed video in a few weeks.
I think another interesting area to think about is related to the challenges in the 'world of work'.
16/ Collective knowledge management is hard.
"Collective systems require bureaucracy (the good kind): structure, accountabilities, and governance. They need significant upfront investment in building consensus and setting up team norms."
17/ And software is only half of the solution. The real challenge is in implementation.
"Software is like a lump of clay, waiting to be moulded according to your organisational requirements... most do not have a clear understanding of their requirements."
1/ There are two broad approaches to deal with 'information overwhelm':
1. Manage the flow (but still consume it all!) - use shortcuts, tools & tricks, and triaged prioritisation 2. Close the tap - stop the information coming in at the source.
I think #2 is more critical.
2/ I've realised the need to improve my filters & make peace with the fact I won't get to it all.
You don't need to listen to every podcast/read every newsletter (esp. if you're trying to appear smart!) Attention is too precious to not enjoy the process of engaging with content.
3/ My goal is to have 2 lists: 'read' & 'don't read'.
There are thousands of non-fiction books out there, and everyone is convinced that the current book on their nightstand is a 'must-read'.
Whilst I take note of these recommendations, there is only so much time in the world.
33 thoughts to commemorate another trip around the sun.
These are some things I've been chewing on over the last while, and I thought it would be a good marker to remember the year by.
The statements are completely unsubstantiated, so take with a pinch (handful) of salt.
As always, I feel the need to caveat.
Any effort to capture broad perspectives in pithy tweets is invariably going to miss some marks. Hopefully there's sufficient ambiguity and minimal preachiness.
Where I fall short, remember your handful of salt 🙏🏽
1. Universal statements on the best way to live your life are best avoided.
Everything is contextual. The current moment is not a snapshot in time, but rather the downstream consequence of innumerable circumstances, and everyone's path to this moment is different.
The lazy answer: all the nouns - people, projects, places. But there's far more to it!
I've been thinking about different components of 'corporate' work & putting together a rough list of links that might help structure workflows.
Many knowledge workers will shift to personal (or collective) knowledge graphs & need frameworks for organising their information.
There are existing frameworks (e.g. the RDF, see wikidata.org/wiki/Wikidata:…) which can be leveraged, but everyone will have their own preferences.
You'll know the feeling of wanting to do something your own way if you've ever opened someone else's Excel model or Notion DB.
It will take time to develop & refine approaches that feel intuitive to the individual, but still conform to protocols that allow for interoperability.
1/ I believe that page properties are going to become an ever-increasing part of @logseq workflows and systems in the future (possibly more so than block properties)
Adding properties to pages (and blocks) adds resolution to links by specifying the nature of the relationship.
2/ Have a look at what @zsviczian is doing in Obsidian with TheBrain plugin to make the graph view more powerful.
It's all based on meta-data, which is implemented in Logseq using properties. It's only a matter of time before we see similar functionality in Logseq (I hope!)
3/ However, most people's experience with page properties is this greyed out block.
Depending on when you joined Logseq, you will often find page titles in YAML format when you click into that block. To change this to the new format, delete the dashes and add another colon
These parts are constantly expressing themselves in the beautiful and chaotic dance of life. Over the years, it's natural for dominant parts to take centre stage.
Other parts of you might be suppressed or put away for a rainy day.
2/ Now combine this with our dualism. When we judge others (or ourselves) we often default to binary definitions of 'good' or 'bad'.
Judgements become all-consuming causal loops: 'she is a good person because of x'; 'he is a bad person because he did y'.
One strike, you're out!
3/ A family man of 30 years has a one-night-stand on a business trip...
Does that make him 'bad'? The act itself is bad if it was outside of the marital agreement.
However, I'm sure many can acknowledge the part of themselves that longs for similar forbidden, novel excitement?