Hacker News new | past | comments | ask | show | jobs | submit login

While I agree regarding correctness and concurrency, I wouldn't emphasize performance, even for Java and .NET, whose GCs are more developed, tuned and tested in the production environment than Haskell's one.

Edit: And for Haskell, achieving needed performance can be quite tricky with regard to its laziness.




Edit: And for Haskell, achieving needed performance can be quite tricky with regard to its laziness.

But Snap is driven by enumerators/iteratees (via the enumerator package), which make it much easier to do fast near-constant space IO.

I have built a web application that uses Snap to query large (parse) treebanks, and it was easy to make it performant.


There was some discussion on HN about the Snap Web framework benchmarks relative to some other frameworks: http://news.ycombinator.com/item?id=1380405

No .Net in there, though.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: