Spending cognitive energy

07 Nov, 2023

Moral

Don’t forget to consider Cognitive Energy, a finite resource, when Estimating Cost.

Estimating Cost

I’m in the process of doing a bit of a career pivot, so have been working on an independent project to show potential employers that I can learn and make things independently, since I’m lacking in formal experience in the direction I’m trying to go.

The constant struggle with independent projects is choosing what to work on.

With no external constraints, you have unlimited freedom! Which can be exhilarating, motivating, and simply more enjoyable. On the flip side, you have unlimited freedom! Which can be paralyzing, exhausting, and an occasionally dangerous place to be if you’re a Chronic Overthinker.

One of the hardest parts of choosing what to work on is Estimating Cost of the things you’re choosing between. ¹

I think a lot of people focus mainly on Time and Money when Estimating Cost, but imo, that’s a very limited view that can lead to poor estimations. (And probably why so many products have such bad user experiences, but that’s a Story For Another Time.)

Some other underrated parts of Estimating Cost include: how much joy will the thing bring you and/or others emotional energy expenditure changes to social capital balance (in flow and out flow)

But today I want to talk about:

Spending Cognitive Energy

Example: choosing an approach, manual or automated

Here's an example of how I considered Cognitive Energy when Estimating Cost:

For the aforementioned independent project, I wanted to add a bunch of award data for movies into a database. The first step here was sourcing the data, and this excellent website is a great place to start.

A lot of stuff on Wikipedia is formatted in a way that plays (relatively) nicely with copy and pasting into a spreadsheet, which I can output as a tsv file, which I can (relatively) easily ingest into the database.

But you know what might be better than this workflow?

Writing a program to scrape data directly from Wikipedia!

Pros include:

• Much faster to collect much more data = saving Time (well… over the long run. probably.)
• Adding another Side Quest independent project to my experience

Cons include:

• Much more Cognitive Energy required

Ultimately, I stuck with the approach of copy and pasting, then manually cleaning up the data into a format that would play nicely with my database tables.

But Sam, you may ask, aren’t you wasting Time by doing all this manual work?

Well, yes, probably. I do intend to scrape A LOT more data off wikipedia for this project. ²

But Estimating Cost is tricksy.

Sure, automating this task will likely save me Time. But, only in the long run. Plus, the future in uncertain, and I don’t have enough information at this Decision Node to know the true Time costs.

But more importantly, if I automate this task, doing so will unequivocally consume way more Cognitive Energy.

My current goal is to finish a SLC of this project as soon as possible. I don’t need that much data to accomplish that. Even if I end up writing some program to scrape data off Wikipedia later on, if I write that program now, it will almost certainly delay when I accomplish my current goal, compared to simply doing the manual way.

My Cognitive Energy is a finite resource, after all. To replenish, requires Time.

So in this case, I decided to conserve Cognitive Energy.

Because what’s the point of saving Time if I’m too mentally drained to make good use of it?

Moral (again)

Don’t forget to consider Cognitive Energy, a finite resource, when Estimating Cost.