But then it's a trivial observation; of course you should do something.

i think it's do something small, manageable, something you can finish soon, to get the feeling you're being productive, while the stuff in the slow-cooker works its magic.

I think it’s different. It’s not just doing something but doing something with an end in mind.

I’m not saying one is better than the other. But having an end in mind will yield more productivity.

I write a lot of dev ops type stuff so I ship a lot of smaller things as a rule.

But, I have adopted a ship every day rule for myself, which can often keeps me from yak shaving, which is actually a direct combat for my ADHD deep dive tendency. I love what I do and I love to make perfect code, but sometimes that's less than helpful.

One of the ways that ship every day is extremely helpful for me is in helping me with realistic time boxing.

If I'm working on a bigger module or something really complex, I'll note that in the morning and spend 15 minutes adding a new graph to the dash, or knock out a quick bug fix that's been annoying me.

While there's lots of circle jerking about dopamine rush of ADHD, for me, if I had my way, I'd go so deep and ship perfect code once a month or twice a year.

But, we also know there is no such thing as perfect code, but there is a sweet spot that gets quality code into the wild with the least possible amount of time working speculative edge cases.

Looking at my stack and deciding if I'm going to ship my main project today or take a 15-30 minute warm up bug/push thing, has been extremely useful for me and I end up with better small single function commits, and pushing working code with a slot for the edge case. I still do the edge case before we go to prod, but my projects move a touch faster, I don't rabbit hole and my bus factor is quite a bit lower.

In other words, "make progress every day".

Learning something useless for fun counts as doing something!