4 ms·

At your request I'm answering this question... I never used the word "fanboy" in my post. That said, there were two people who really thought the nextTick solu

by bascule 15y ago

At your request I'm answering this question... I never used the word "fanboy" in my post. That said, there were two people who really thought the nextTick solution was a legitimately good idea. Contrary to the idea that they were just countertrolling, the Node-fib guy legitimately thought he had done a good implementation of a Fibonacci server in Node because ApacheBench told him so, never mind what he was actually benchmarking was TERRIBLE Read the "Coming from Hacker News?" section of the Node-fib README and tell me if you think this guy is a troll or if he truly believes Node can be good at these sorts of problems: https://github.com/glenjamin/node-fib https://github.com/glenjamin/node-fib

IsaacSchlueter 15y ago
He's not a troll. This is simply a clash of cultures. In the node community, it is considered impolite to discuss performance without reproducible benchmarks and specific numbers. Mr. Dziuba, a newcomer to our shores, did exactly this. While many of us were offended, Glen decided to give him the benefit of the doubt, and provide the example program that Ted had neglected to share. The results are facts, not opinions. I'm not sure how you could think they're trollish.