Oddbean new post about | logout

Notes by 616c1b2d | export

▲ ▼
Andrew Helwer | 1 years ago (raw) | root | parent | reply | flag 
 nostr:npub1tkze6tszqp7vlw48tv2537qttppxysym0jptfwak0z7r9wv79tss6ncd27 testing it by running it an...
▲ ▼
616c1b2d | 1 years ago (raw) | root | parent | reply | flag 
 @8bb49bc2 okay, so you end up with developer tests, or do you mean regression tests?
▲ ▼
616c1b2d | 1 years ago (raw) | full thread | reply | flag 
 It's interesting and surprising to me that it's difficult to show TDD actually has a positive effect.

How does the non TDD method work? Test just after in short iterations probably works well too, but it appears the methodology commonly compared against doesn't actually involve any automated developer testing. I wish the alternative "conventional" methodology was defined better - what is it?

Is it devs testing by hand and throwing it over the wall and hoping for the best?

#programming
Event not found
▲ ▼
616c1b2d | 1 years ago (raw) | root | parent | reply | flag 
 @5150cec1 I don't have anything in particular to recommend, but just learning a bit about CPU caches can help. Basically to exploit CPU caches you need to keep things you use together in memory instead of separate. And that an array scan can be insanely fast.