End of sophomore fall: a semester in review

大二上 is done. Here's a roughly honest account.

Coursework

Better than 大一. Partly because the material is getting more interesting, partly because I've gotten better at studying.

The highlight was a systems course where we implemented parts of a kernel. I'd read about virtual memory, page tables, context switching — but there's a specific kind of understanding you get from writing the thing yourself that reading about it doesn't give you. That course gave me that.

Algorithms was fine. I can derive most of the important stuff from first principles now, which feels like the right way to know it.

VALKI

Still running. I've been less deep in the codebase this semester — busier with coursework — but I've been doing more on the product thinking side. Figuring out what the next version of something should be is a different kind of problem than building it, and I've been spending more time on that.

The thinking I'm glad I did

I spent a lot of the latter part of the semester thinking about what I want to build next. Not in a vague "someday" way but in a specific, "what would I actually start in February" way.

I have a clearer answer than I had in September. I'm not going to write about it yet — I'd rather show it than describe it — but the thinking was useful.

What I want for spring semester

Start the thing. See if the idea holds up when it's real and not just in my head.

Write more here. I kept intending to this semester and finding reasons not to. That pattern should stop.