Using Obsidian for issue monitoring and analysis

Wanted to share a way to use Obsidian that isn’t about daily notes and task management (which I think there are many other options available). I’m a political analyst by training. So I’m always trying to figure out how to keep track of all that is happening across a wide range of issues and my thoughts about those events.

The problem statement
Most real world issues are reported in a mishmash of sources, lengths, facts and opinions, and are linked to other issues. You can’t always tell what’s going to be important or how it’s going to be important. So it’s hard to put an event into a specific box.

If you include what you think about an event in that note, it’s also hard to keep track of your analysis. But your analysis could be based on multiple events. Or your analysis of a single event could change. How do you keep track of all of that?

The solution

  1. Record events as a single note with a compulsory #event tag.
  2. Record your thought/reaction to that #event note in a separate #idea note.
  3. New #idea notes should also be created when you refine an existing #idea.
  4. Other tags etc can be added to help you figure out MOCs or wider issues, but the idea is not to get too bogged down in creating a maximal taxonomy from the start.

Not sure if it’s allowed, but I wrote this up in more detail here.

Would be great to hear your thoughts!

1 Like

Much appreciated!

Creating a new note whenever you refine an existing idea seems like a recipe for clutter.

However, I understand your desire to be able to recall the process by which you arrived at new conclusions. I recently worked on a difficult data analysis project where I found it helpful to keep a log, so I could pick back up my train of thought and recall what I’d tried and where I ran into difficulties. It looked something like this:

Project X

2022-07-24
Aggregating by quarter is not a good approach but noticed the following patterns…
2022-07-25
Five possible reasons why this isn’t working.
2022-07-28
Eureka! Linear regression isn’t working as expected because the relationship is nonlinear. R-squared and p-values are misleading because the data is auto-correlated.
I revised [[How to do multiple linear regression]] with a reminder to plot the data before running statistical analysis.
Here’s a better approach for a non-linear relationship…

I see no reason to keep a note with outdated or incomplete information about multiple linear regression, but I do get a backlink from the revised note to my project log, describing the pitfalls and learning process. This could come in handy later.

For your purposes, you might add a change log at the bottom of each idea note. Here’s an example.

Atomic Notes

Blah blah blah

Evolution of my thinking

2020
Read Sonke Ahrens and Andy Matuschak and thought atomic notes were a great idea.
2021
Had to take notes about something other than personal knowledge management. They were long, I wasn’t sure how best to split them up, and I didn’t have time.
2022
Stopped worrying about it. I can always link to headings.

If your analysis is specific to an event, I can’t see any harm in putting it with the same note. An inline tag will allow you to find all the notes that include some original political analysis, and jump directly from search results to that part of the note.

2022-08-10 Earthquake in the Obsidian Forum

Link to news stories
A few details from the news story

My take

#analysis
One way or another, my political opponents are to blame.

If there’s another earthquake, and you want to make your claim more general and link it to both events, you can copy and paste your analysis into a separate note at that time.

1 Like

Thanks for your in-depth reply! Had a looong think about it and tried it out in my vault too. I think that for most issues, the change log approach is indeed going to be neater. But for some situations, I suspect it still won’t work.

The main one I am thinking of is when your assessment morphs into something completely different from the original. The change log approach assumes that it will be clear to everyone when you should just add a new line to the old assessment or when you should create a whole new assessment. To avoid that issue (which comes up for me a lot), it seems easier to just create related assessments.

So using your example, the 20220815 analysis could arguably be in a new assessment. But much time would be spent arguing one way or the other and would be hard to stick to an objective standard over time.

Earthquake in Obsidian forum due to political opponents

20220815 - Hey the lava seems to be green, suggesting copper contaminants that my opponents have
20220814 - Oh now I think my political allies are to blame
20220813 - I think my political opponents are to blame because blah blah blah

#analysis

Thanks again for your thoughts, really appreciate them!

Sharing here some new thoughts I had about Obsidian for Team KM (heresy?!)