The reason I don’t bother writing code this days is because my use cases have been solved, and if they weren’t, I’d tweak the most suitable candidate. One of my principles is to keep my workload small. More often than not, things starts with a small script or plugin, and then grow according to my needs. Why replicate what others have already done?
replyIf others have built a whole-ass house and I just occasionally need the kitchen table I had to just deal with the hassle of having a whole house and just used the table. And even the table was the wrong shape, but I could deal with it. Asking the Others to make the table modifiable would've been a massive effort of PRs and mailing lists I didn't want to get into.
replyNow I can build a bespoke table in an evening or two and it fits my stuff just perfectly.
I could do it before too, but it would've taken too long for me to bother, so I dealt with the whole house along with the table.
There’s a lot of small programs out there. Especially if you go to the BSDs where small programs are the norm.
replyBecause its fun. And because your experience using a tool is fundamentally different if you made it yourself, compared to if its something someone else made for you.
replyI don't think I can explain the difference, but it feels really different. Even if you used claude.
It never felt fun for me to write software fully with LLMs. It feels disorientating, it produces a lot of code that you have no familiarity with and no authorship. It feels like you’re a teenager again, copy pasting code from internet or journal and hoping it will work.
reply