Mind The Gap, Please
2025-06-21
Matthew Bornhorst, Unsplash
It has somehow been nearly two years since I've written anything on this page. I haven't stopped having thoughts, tinkering with things, building new stuff, or writing ideas down. After some thinking, and a lot of learning about myself over this time, I can firmly attribute this to friction.
We, as humans, will just about always take the path of least resistance. This can manifest in very straightforward ways, such as skipping a morning run because we're tired, or even in less straightforward ways, like going on that morning run rather than tackling deep feelings. Now, taking the path of least resistance isn't necessarily always a bad thing. However, it can draw us away from actually meeting our goals, or from doing so in a sustainable way.
As such, a lot of my self-evaluation cycle (inspecting my own habits and making changes to get closer to the way I want to live) ends up involving reducing the friction around something, and making the path of least resistance a path that is created by me, actually moves me towards my goals, and is sustainable. To that end, this post is a summary of how I've massively reduced the friction involved in information synthesis as well as publishing.
Part I: Information Synthesis
I've already mentioned different note-taking and information management tools in past posts (1, 2), and it's continued to be a thorn in my side, with every good solution around having at least one key downside, such as pricing, lack of data sovereignty, strict adherence to a Zettelkasten approach, lack of good mobile support, UI/UX that I dislike, or something else.
This is a topic that I plan to cover in the future, but I found inspiration for the solution in my physical notebooks. I don't keep many physical notebooks due to my awful penmanship, but each one I do keep is differentiated in multiple ways. Each is used for a different subject or type of information and only used for that, is of a different form factor best suited for the information, and has a certain epistemic effort associated with it. See the below examples:
| Notebook | Content | Form Factor | Epistemic Effort |
|---|---|---|---|
| Pocket Notebook | Unstructured information from conversations, back-of-the-napkin calculations | Slim, passport-size dot grid notebook | Extremely low, just whip out the book and scribble in it |
| Spring 2024 Linear Algebra | Crystallized, heavily refined topic notes, complete with heavy formatting and diagram | Thick, hardcover dot grid notebook | Very high, this is meant to document my understanding of concepts such that I can "load" that information back into my brain, notes are written in dedicated day-long sessions |
| Summer Internship Daily Journal | Daily notes about accomplished tasks, meeting notes, and reminders | Medium-size lined journal | Medium, they're working notes but I should be able to get a good picture of what I wrote about later |
Mashing these all together into one notebook would never make sense. I can't just keep a giant tome of knowledge in my pocket for quick notes, and the organizational overhead of such a solution would be insanity. Furthermore, having notebooks be compartmentalized offers many benefits. Accidentally putting my pocket notebook in the wash and ruining it doesn't affect my ability to pass my math test. I can leave my internship journal in my cubicle so that I don't need to carry it around after work. To put it simply, separate notebooks for different use cases means each can be tailored for that specific case.
This is pretty intuitive to most. Even in grade school, it's common to have different books for math and art class, and to use a lined book for math and a sketchbook for drawing. But when it comes to digital notetaking, the focus is instead on having a one-size-fits-all solution, a single Roam/Notion/Obsidian Graph/Workspace/Vault that fits every piece of information of every type from every area of life into it.
So after spinning my wheels on finally dealing with my lack of a note-taking solution, I finally thought to apply the same logic as my physical notebooks. I started off by setting up a Notion workspace for university. This use-case takes advantage of Notion's strengths, such as the block editor allowing for more free-form structure and the web-based nature making it so that I can access my workspace from shared computers, while making the caveats a non-issue, as what I need for school falls squarely within free-tier, and data sovereignty doesn't matter as much for me since this doesn't hold very important information and what is stored in there for the most part is only relevant for a 3-month semester. I then started using Logseq for work notes. This use-case needs certain capabilities that Logseq has, such as full data sovereignty and easy installation, and benefits from what would otherwise be a caveat for me, the fact that Logseq is based on a Zettelkasten approach, with use of tags and search. The Zettelkasten approach would not at all be feasible for most areas of my life, but is perfect for work, allowing me to quickly jot down information with minimal setup or organization overhead.
Finally, for personal notes I'll be (partially) using Obsidian. I include the "partially" because while my Obsidian vault is meant to serve as a hub for information gathering and synthesis, it is not meant to be a one-size-fits-all solution, even within the areas of my life not covered by school and work. This vault, as with every other knowledge system I have, has an intent statement, and is restricted to specific use cases. As I find myself in need of more places to store knowledge, I plan on using the tools that make the most sense for that use-case, in an effort to keep each digital "notebook" focused, intentional, and maintainable.
Part II: Publishing
Over the past year or so, I've made substantial changes to the workflow for publishing to this site.
This site is a static site, generated by a bash script I've named bliss(see the colophon for more details). After an article was added, I'd run the script locally to generate the site files in a subdirectory of the GitHub repository, then commit and push everything, then point GitHub Pages to that subdirectory and serve it. This had a number of friction points:
- I needed to be on a machine that had the script's assorted dependencies installed, and just hope there were no breaking changes or platform differences with any of them
- It was a multi-step process, with steps that needed to be run in order
- It was a sit-down process Ultimately, this wasn't ideal, but it felt like the best workflow at the time, since GitHub Actions/Workflows felt strange and impossible.
Since that time, I've "seen the light" of CI through heavy use of Gitlab CI at work. It turned out that CI itself is not strange and impossible, just GitHub's implementation. I've been able to understand how to work with GitHub Workflows better as I've become more proficient with CI systems as a whole, but I'll maintain that it's a really obtuse and difficult-to-use system, and they could do way better. I finally bit the bullet and started migrating all of my personal sites and little projects over to Gitlab, meaning that it's now seamless to use Gitlab's excellent CI to deploy projects. This site is now finally deployed via CI, as it should have been all along, meaning that publishing is now as simple as copying my Markdown draft from Obsidian into a file, committing and pushing it to Gitlab, and sitting back.
Epilogue
With these new "friction-reductions", I hope to start writing (and especially publishing it) much more consistently. One of the best parts of my day is opening my RSS reader and seeing new unread blog posts, the ideas and musings of all sorts of people who take the time to share what's in their head for no motive other than the desire to spread information and the joy of writing it down. I too hope that I'm able to put things out there that inform, inspire, or hell, that are at least an interesting read.