Took me two weeks to write 100 lines of C.
Sure, I wrote a few thousand lines to test it and pages of conversation, but most of my time was spent understanding, not writing.
Reminder that *writing* code is a tiny minority of the time we spend on “programming.”
Sure, I wrote a few thousand lines to test it and pages of conversation, but most of my time was spent understanding, not writing.
Reminder that *writing* code is a tiny minority of the time we spend on “programming.”
Comments
More constraints = clearer what will happen when the program runs
At the start of my journey I was the opposite of that, as I improve I spend less time typing and testing and more time thinking and planning it seems.
The value I’ve gotten from tools like Cursor so far is that I can write more exploratory code than ever before, because the barrier to sketching up a new idea is so, so low.
trying to build a charting library, after going back and forth on the code, the remaining working pieces are so simple that in hindsight feels obvious, but took me a lot of time to get there.
if you're making something from scratch, it does not help at all, at least in my experience
1. Intelligence requires a goal
2. Intelligence requires sentience
3. It requires a perspective
4. It requires consciousness
5. Emotions
6. Life
Or ask it for context, like how should I structure this?
LLMs are experts on normative questions (how should) because that’s really what they are
I’ve also had good luck pasting in code and asking what it does or why it works