a long intro

Note keeping and knowledge management has for a long while been important to me. It’s part of why I started blogging in college (somewhere between 2000 and 2002). Those blogs represent some of my earliest attempts to document my thoughts (in joomla of all things back then). As the years rolled on I continued to make public notes via blogging and different systems. I also added new tools to collect, store and process knowledge as well. Theological notes moved mostly into a program called Accordance. Eaglefiler was one of my earliest attempts to collect disparate bits and bobs of files together back in 2005 or 6 - whenever it first came out. I tried (and loved) Together for awhile and then around 2014-15 "upgraded" to Devonthink Pro. I actually still have a library archive of documents in Devonthink Pro (but haven’t felt the need to upgrade to 4 yet).

That’s all in the realm of "things I thought" and "things I stored". Another aspect of knowledge management for me personally is "things I did". Even though that is a bit more ephemeral, there is all-to-often overlap and dependency between the categories. I struggle to separate them internally because one leads to another and then back again. So - projects and tasks and their associated work is another important part of the puzzle for me personally.

My earliest attempt that I remember where tools like Clear and TeuxDeux (this was pre-2010 but is actually still around!). There was an iphone app called Cheddar I liked a lot as well before it shut down. I then moved on to more dedicated systems: Omnifocus, Things, 2Do on the tasks side and Omniplan and inshort on the project side. A frustration and limiting factor was how disconnected everything was though.

There was an app I briefly used and loved that sort of bridged some of the gap: DashPlus. It was modeled after the Dash/Plus system by Patrick Rhone and allowed action states to be easily symbolically represented (done/undone/waiting/delegated/data point/carried forward/idea/thought) in a single place. I wish the app was still around as I got a lot of use out of it for a season.

You might notice an issue in all of the above: app dependencies. If an app shuts down (as a number of the ones mentioned did), you are left searching and not always with data to import elsewhere. Even if one doesn’t shut down you are often left with proprietary formats that make linking systems, processes and knowledge difficult.

Enter plaintext. Plaintext is definitely not *new. Org-mode in Emacs has been around since 2003. *Other plaintext systems have been out their as well but I think accessibility has been a challenge. Org-mode as example requires deep familiarity with a unix text editor. I wasn’t super familiar with the idea of plain-text to rule it all until Obsidian. Obsidian in 2020 opened a whole new world of possibility with notes and information interconnected on a graph. With the addition of plugins like dataview and tasks, all your information could exist together. I created my graph with a spot for "cold storage" (notes and info I didn’t regularly need to access) and spaces for "hot storage" (notes I worked out of day to day). And I had text dashboards linking it all together. It was great.

Obsidian wasn’t without issues though. I had issues with scale for one. When I left the platform in late 2023, starting up Obsidian on my desktop would take 3+ minutes with about 1000 text files in my system and too many plugins (I think it’s better now in this regard though). I also was getting bogged down in the lack of opinion. Obsidian is so open that you can do anything and there aren’t really any guard rails. That means I spent just as much (if not more) time building systems and processes as I did making use of the knowledge within. Importantly though, it showed me that my data (what I thought, what I stored and what I did) could all live together in harmony in a way that benefited me and the work I do.

I tried a number of options post-Obsidian. Craft was one of the first and is still my favorite app for document creation (as in, when I need a PDF and need it to look nice). It lacks though in what I’d call "deep learning" capabilities (namely the ability to curate and correlate knowledge). I dabbled with notion as well but only really for one particular project. It didn’t really "stick" for long. Capacities was another one I mostly liked but it didn’t really stick in the long run - it felt more like something somewhere between notion and craft. The longest post-Obsidian run has been SilverBullet. SilverBullet is an online app only (which was the biggest issue) that’s something like a markdown based wiki, that stores all the pages locally as markdown files. It’s very nice and works very well and feels something like a lighter version of Obsidian (which is nice). Mobile input was a struggle (I did some scripting work to automate the transfer of notes from an unrelated mobile app into SilverBullet) and ultimately kept me looking at options. I tried self-hosting Anytype which was straightforward but a bit resource intensive. That brings me to today. Well, not just today. The past month and a half, really because I decided to give Tana a try.

tana

I first saw Tana a few years ago and it was "waiting list only" and didn’t seem super accessible. At the time, there wasn’t mobile either and that’s a big part of my workflow so I didn’t give it much thought other than a signup because I’m a sucker for waiting (not really but I am a sucker for productivity).

I got an email invite some time ago and brushed it off. I saw another email when it officially launched. I brushed that off too. But - I saw an email newsletter and was frustrated enough with my system that I thought "why not". I went to the site and almost closed it because there is so much AI marketing. I don’t really want to base my stuff on AI so that made me skeptical but I signed up anyways.

💡

Thankfully, despite all the marketing, AI is something you don’t have to engage in but can tack on to nodes if you want; it’s not something as "in your face" as the marketing makes it. I know this will be a sticky point for some of you reading.

Tana is different from many/all of the other apps I’ve used in that all of the bits of information (a sentence or paragraph or property or ...) are distinct nodes to be worked with, described, networked and engaged with. They can be further defined with supertags (which are #tagged nodes with specific templates and properties. So, if I’m writing a blog (for example) and there is a paragraph I want to hold on to for further engagement, I can add a #thought supertag or #note supertag (two I’ve designed to serve specific knowledge needs). Supertags are highly dynamic - I might be working and realize some particular bit of information is worthy of a new supertag and all I have to do is #tag it. Once it’s tagged, I can filter and search based on those tags, see what amounts to a database of the tag, and make quick dashboards to engage it as I need.

I realize all of that is probably a bit confusing but it all works as big outline, basically. Think:

• #tag1

• #tag2

• #tag3

• Page 1 #page

• Content 1

• Content 2 #idea

• content 3 #thought

• References to Page 1

• Links to Page 1

• Page 2 #page

I can click into a focused view at any point on any node. Tag settings define the view for each node tagged as such. Dynamic queries can be defined to provide card or kanban or calendar or table views of the nodes. Focused views include references and links to the specific parent node (see above).

I realize it might sound confusing but it really clicks with the way my brain works. It’s opinionated (node/outline focus) in a way I wanted something to be and it still provides an openness to build, iterate and improve that I find important. Most of all, to that latter point, it’s open enough to grow with me. I don’t feel a pressure to have everything figured out from the get go.

closing thoughts

As I age, I want a source of context and truth in life. Context has always been tricky for me - I don’t hold on to much in favor of looking forward to what’s coming - so having a place I can trust to provide the needed references and links back is important. It’s note repository linked with quote library linked with task manager linked with lightweight CRM linked with media library and so on. I can run my projects and keep the information I need in front me.

What’s more, one of Tana’s strong opinions is working out of a "Today" space that changes day to day. I have a growing and improving template for keeping track of my day. It works and roots what I think, what I store and what I do squarely in time. It really is the best system I’ve used in quite some time. It’s the first one I’ve been excited enough about to spend some words on my blog about. I readily recognize that it might not be for everyone. But it is working for me now (and thankfully, if it ever stops being a good fit or the AI stuff gets too much in the way everything can be exported as markdown).