I find this topic super interesting because I recently created some side-projects using Claude (not in their entirety, but major parts of them) and felt strangely disconnected to them afterwards, even though they accomplished what I was looking for. That made me think about how AI usage for me should look like in my free time.
Maybe what the author is talking about, and what I felt to a degree, is the act of deciding and formulating decisions. Looking at the compiler vs. AI example for machine code, one aspect that immediately stands out to me is that a compiler should deterministically produce consistent outcomes for precise instructions. Claude, on the other hand, will take your instructions, which are likely less precise than code, and "reason" about them. It will produce output that varies each time you would ask it and it will bake in behavior you didn't explicitly ask for. AI makes decisions for you. And maybe that's what makes AI so useful in the first place.
One way around this may be to be just as strict with your instructions. Giving AI an exact spec to work with, like large Markdown specs or detailed tests to run against. But then again, if you take the time to formulate those, is it much of a jump to write the code yourself?
What's clear is that AI is awesome for helping with issues you can't resolve, or explaining concepts in a personalized way. But letting it make decisions for you prioritizes outcome over creation, and for side-projects you build out of the love for creation, that may just not make sense, just like as a consumer I feel disinterested in "AI art". AI is for products, not for art, and that dumb little idea to work on in the afternoon might just be more art than product.
I was in a similar boat some months ago but realized my inability to come up with good ideas was really just me getting frustrated by previous side-projects I didn't finish or got bored by.
What helped me get unstuck and get my creativity back up was setting myself constraints, like whatever I work on today, I'll ship it today, or let's try to make an intentionally useless bash script in 20 minutes.
In my spare time I'm working to finally complete creating my own blog/site.
It's built using Nuxt because I've never really played with Vue before and it seemingly comes with all I need for a static, markdown-powered blog. I guess what's been stopping me was me bothering too much about "When is it good enough to be online?" and "What should the first post be?". But I'm trying to get rid of the perfectionism by just putting it out there and just posting something. I think I'll reflect on this in the first post.
To your second point: For me, the major difference between goals and constraints would be that I can clearly achieve a good goal, but a constraint is something that will never be fulfilled. A good goal is to run and complete a 10k marathon, it's easy to tell when you're done, or if you failed, potentially even measuring how far off you were. But a constraint would accompany you until you choose to disregard it. You can respect a constraint, but you can never complete it, only in the context of a finite project.
To me, a lot of this post sounds like goals vs habits, caring more about what you do today than what you may achieve sometime in the future, only that the habits are constraints here, so not doing something. In short, "leave everyone better than you found them" is something you can adhere to constantly (like a habit), but for it to be a good goal you would have to know when you're done finding people I guess.
Ultimately, what I read from this post is that constraints are used to provide identity, to help you guide yourself everyday. And maybe that's what you need more than goals if a lack of identity (in your work) is what's troubling you.
I thought about doing something similar some time ago, because I never quite found the perfect note taking app for myself. There's a million ways how to do notes, and it feels like there's just as many different notes apps.
Eventually, I've settled with Obsidian because of its simplicity and extensibility. You can leave it with basic features and truly own your notes in a simple format (you can also put them into any cloud, as long as that cloud reaches your filesystem). It doesn't do everything just like I'd want to, but I've thought about just building another notes app that reads and writes to the same path your Obsidian notes are in, instead of trying to cover every possible editing feature like most big notes apps. Then I'd use different apps for different needs, with one place to store data.
Since you're focusing on privacy, have you considered using Obsidian? Is there anything particular you want to do differently?
Are you looking for something like `uv sync --upgrade`? This one should be re-assessing your dependencies (excluding version pinned ones of course) and regenerate the lockfile if I remember correctly.
Maybe what the author is talking about, and what I felt to a degree, is the act of deciding and formulating decisions. Looking at the compiler vs. AI example for machine code, one aspect that immediately stands out to me is that a compiler should deterministically produce consistent outcomes for precise instructions. Claude, on the other hand, will take your instructions, which are likely less precise than code, and "reason" about them. It will produce output that varies each time you would ask it and it will bake in behavior you didn't explicitly ask for. AI makes decisions for you. And maybe that's what makes AI so useful in the first place.
One way around this may be to be just as strict with your instructions. Giving AI an exact spec to work with, like large Markdown specs or detailed tests to run against. But then again, if you take the time to formulate those, is it much of a jump to write the code yourself?
What's clear is that AI is awesome for helping with issues you can't resolve, or explaining concepts in a personalized way. But letting it make decisions for you prioritizes outcome over creation, and for side-projects you build out of the love for creation, that may just not make sense, just like as a consumer I feel disinterested in "AI art". AI is for products, not for art, and that dumb little idea to work on in the afternoon might just be more art than product.