freenode/#shirakumo - IRC Chatlog
Search
12:44:18
selwyn
i have a kind of dual attitude in that while i believe that it is important to prove stuff, one thing i don't like is having to implement an algorithm in code (unless it's a pedagogical exercise)
12:45:11
selwyn
that's when i say 'this algo has been around for years, why hasn't someone solved the problem of implementing it in the best/sanest way so i can just use it'
14:03:26
|3b|
ACTION thinks both doing proofs and implementing algos are important, if you want to be able to do more of them in the future
14:04:42
|3b|
but there usually isn't a need for a "better" proof of something, unlike implementations of algorithms, where there are lots of ways to measure "better"
14:06:56
|3b|
but also, time is limited, so it is reasonable to specialize in one or the other, or something completely different, particularly if you seem to be better at (or even just more interested in) one of them
14:20:31
Shinmera
I love the design aspects of programming -- meaning everything except the proof that it's working.