4 ms·
I still think the "Naïve" implementation is the best one given the actual specification of the problem, and not some made-up specification and assumptions that
by platz 12y ago
I still think the "Naïve" implementation is the best one given the actual specification of the problem, and not some made-up specification and assumptions that solve a different problem than the one actually being asked (which may be more interesting, but that is beside the point).
- placeybordeaux 12y agoI thought that was the point of the article.
- simonw 12y agoThat was the conclusion of the article, if you scroll down to the bottom. "The final implementation represents a veritable explosion of complexity".
- doorhammer 12y agoIt was funny to me, because I immediately liked the second version most, but implemented as as function that takes the range of numbers I want to use. The only reason I liked that version more is because it "solves" common "problems" I run into when I'm screwing around with fizzbuzz in new languages. Funny to me because I'm adding a layer of abstraction to handle specific issues I've encountered in the problem space for a question where the problem space is totally synthetic But yeah, as it's asked, I think the first is the best answer to Fizzbuzz