This is the first part of a series about how I write and reflect using Obsidian, an extensible digital note-taking interface with some surprising and unexpectedly useful features. This post serves as an overview and index for the rest of the series, which will explore in depth how I use Obsidian for note-taking, journaling, and creative writing. I’ve been using Obsidian since 2021-02-06, and it has completely revolutionized my writing process. New posts every Wednesday until the series is complete.
In Search of a Digital Note-Taking Analogue
What are the primary virtues of analog note-taking? Immediacy, for one. A notebook requires no loading time (aside from locating said notebook,) no power source (if there’s sufficient light,) and creating a new note is as fast as turning the page. Geography is another one. Our brains evolved to remember the relative locations of things, and physical books maintain that sense of space—the pages exist in an immutable order that becomes familiar over time. Then there’s portability. I used to keep a Field Notes notebook in my back pocket at all times, and my current Leuchtturm 1917 notebook fits as easily in my small café bag as it does in any of my backpacks.
But there are some disadvantages to analog note-taking that left me wanting more. While my Leuchtturm has page numbers that enable me to create an index of its, that index is only really useful for that single volume. I’m currently on my twelfth volume, so unless I’m going to maintain an independent index, I can no longer keep track of where all my writing lives. Another disadvantage is impermanence. On the one hand, we still have Leonardo DaVinci’s notebooks some five hundred years after his death, but on the other hand, I lost one of my Field Notes notebooks on a bike ride back from the climbing gym and I never saw it again. Thirdly, so much of what I read now is digital that it would be much simpler for me to copy and paste excerpts to add commentary than to painstakingly write down every piece of information I want to recall—and then index it.
Where this all came to a head was when I started to get serious about writing essays and long-form fiction. Keeping a notebook for each project made sense for a time, but I would often brainstorm about them in my morning pages or jot down a quick thought in my Field Notes, and I had no simple way to consolidate them when it came time to finally type up the draft. While I still wanted to use my notebook for principal drafting—and loose-leaf notepaper for my morning pages—I had to believe there was a better way. I wanted to eat my cake and have it, too.
Roam (If You Want To)
I don’t remember how, exactly, I found out about Roam Research, but I think it was Mark McElroy who really sold me on the concept. He wrote about wanting to find connections between different dreams in his dream journal, and I had recently finished digitally transcribing some fifteen years of my own journals into Day One . His article about using DEVONThink as a “thought processor” both resonated with my goals and validated my frustrations with my database of choice at the time, which was also DEVONThink.
Roam Research offered a fourteen-day free trial, so I borrowed a python script to export my fifteen years of journaling and get it uploaded into my new Roam database. Like Mark, I hoped it would help me discover unrealized connections between disparate journal entries, picking up on recurring thoughts and ideas, and give me a way to seamlessly call back to them whenever I noticed a connection on my own. And, for a time, it was good.
However, some things began to grate at me. The biggest issue was that everything lived on Roam’s servers, the system was entirely cloud-based, and that meant no offline access. I shuddered to think what would happen if I went on vacation somewhere without an internet connection (like my sister-in-law’s lake house in Washington) and found myself completely disconnected from my external thought processor. On a related note, the initial load time every time I opened my Roam graph was unforgivably slow. I read somewhere that even accessing a single page causes the browser to load the entire graph into its cache, and that meant sometimes I had to wait several minutes to even start typing. This wouldn’t have been as big of an issue if there were an easy way for other apps to dump data into Roam—I’ve been using Drafts as a springboard for digital notes that I can then file into other apps for years—but there simply wasn’t one. Everything had to happen in the browser.
It’s a big deal to ask me to trust a brand new company with a lot of my personal data, especially when it all lived on their servers and I couldn’t access any of it locally, but the last straw was when my free trial was coming to an end. Roam wanted me to pay $15 per month just to access my own data, and that was too expensive. I’ve backed out of app subscription fees before and lost features, but I’ve never had to part with my data itself. Before the trial was up, I exported all of my data to a local folder and installed Obsidian—for free.
Obsidian — Fast, Linked, Local, and Future-Proof
From the moment I started it up, Obsidian felt like it was designed to address the same frustrations I had been experiencing with Roam. Obsidian was fast, due in large part to the fact that it ran locally and operated entirely within my existing filesystem, using nothing but plain text files—well, Markdown files, but still. I could point Obsidian at a directory that lived on my hard drive or a folder in iCloud and it didn’t matter. Everything just worked.
The JSON I got from Roam proved a little difficult to parse, but I only had about two weeks of journal entries to copy back into Day One before I could use another handy python script to export my entire Day One journal as Obsidian-friendly markdown files. The idea to using plain text Markdown files was genius. So many file-formats are proprietary (I’m looking at you, Pages that they depend on regular maintenance of the app that created them to be accessible again. What if the file you want to open hasn’t been touched for two decades?
When I was a kid, I wrote my own parody of Sideways Stories from Wayside School in some word processor that was so old, I could hardly get the prose out of it when I stumbled upon the file a few years ago. Plain text is a format that’s almost as old as I am, readable by any modern computer, and it’s likely to persist well into the future.
Once there, it was super easy to start linking things together—it was like creating my own personal Wikipedia of places I’d been, people I’d met, and books I’d read. And those connections didn’t cost any extra storage space or processing power—again, we’re working with just plain text. The files are so small that my entire Vault—about 7,000 individual notes—only weighs about 300MB. I pay Obsidian for their Sync service (as an excuse to help fund its development,) but such a small package means I have an entire copy of my vault locally on each of my devices. If there’s a blackout right now, and I lose wifi and cell service at the same time, I can keep writing without even missing a beat. It’s great!
A Place for Everything
I relied on external programs like Ulysses and Scrivener for long-form writing, and Things was still my task manager of choice. Now, I rely on nearly forty community plugins and counting, and all my work (and workflow) lives entirely in this one app. I couldn’t have done it without the kindness of strangers in the official forum and the Obsidian Members Group Discord server.
Recently, I adopted the Johnny Decimal System for my folders and created a personal guide to My Obsidian Setup, so I think it’s time to share my setup and how I use Obsidian on a daily basis to do everything from manage my journal to write my novel and process the feedback on it I receive from my critique group. Over the next couple months, I’ll go in depth on the series of goals and challenges that led me to Obsidian in the first place, the features and plugins I use to address them, and how I put it all into practice in service of my specific tasks and projects. Below, you can find an overview of all the posts in the series, which will serve as an index once each post has been uploaded: