Refactoring DLTJ, Winter 2021 Part 1: Picking up Obsidian

Refactoring DLTJ, Winter 2021 Part 1: Picking up Obsidian

As 2021 comes to a close, I’ve been thinking about this blog and my own «personal knowledge management» tools.
It is time for some upgrades to both.
The next few posts will be about the changes I’m making over this winter break.
Right now I think the updating will look something like this:
Ramp up automation for adding reading sources to Obsidian (this post)
Refactor the process of building this static website on AWS
Recreate the ability for readers to get updates by email
Turn the old DLTJ «Thursday Threads» concept into a newsletter
I’ll go back and link the bullet points above when (if?) I create the corresponding blog posts.
I’ve been using
Obsidian
for about six months as a place to note and link ideas on stuff I’m reading and watching.
In case you haven’t run across it yet, Obsidian is a personal wiki of sorts.
It is software that sits atop a folder of Markdown files to provide indexing as well as inter-page linking and graph views of the folder’s contents.
Most people use it to build up their own personal knowledge management (PKM) database.
You can make notes for the sources you are reading, then build knowledge by linking sources together using keywords and adding commentary at the intersection of related ideas.
Before Obsidian, I was using the
Pinboard service
to store bookmarks of interesting sources and using the paid subscription search engine and my own memory to find stuff.
I’ve found that this setup works okay for retrieval—I can usually find things that I know I’ve read about before—but doesn’t do so well for making new connections or creating new knowledge.
The
Thursday Threads
series on this blog years ago was, in part, a way to find those connections and explore them a little bit in writing.
I’m expecting Obsidian to help improve this area.
The start of the knowledge curation process is creating pages in Obsidian for the important/useful things I’m reading—each of these is a «source».
I like the idea of having a bookmark service as the start of the queue of sources feeding into the PKM; It is a universal tool that is available from a wide variety of entry points.
In my desktop browser, I use the
Pinboard Bookmarklet
to add new sources.
On iOS, I use the
Pins app
on the share sheet to add things.
The Pins app works not only in Safari but also in other places like the New York Times and Twitter apps.
To get sources from Pinboard into my Obsidian PKM database, I wrote a
Python script
that uses the Pinboard API to copy bookmarks into an intermediate SQLite3 database, and then every morning creates a page in the Obsidian database for each new source.
Please note that this Python script is quite the mess; it started simple but has had functionality grafted into it a dozen times now, and it is in need of a serious rewrite.
For better or for worse, it is out there for others to inspect and get ideas from.
For the sources I add to my PKM, I’m also concern…