Why isn't Microsoft trying out Haskell? Reading "Coders at Work" I was surprised to learn that research/development of Haskell's GHC is financed almost entirely by Microsoft, yet they aren't using it even for internal purposes as far as I know. And for their own functional language (F#) they picked ML to build on.
I don't remember where I saw this, but I recall reading that they actually considered it. Ultimately, however, the pure functional style just didn't jive with .Net very well, so F# was born to allow functional style to live along side of the rest of .Net comfortably. The F# lite syntax also takes some cues from Haskell.
Hi there. This happened last night after I was tweaking the number of FastCGI processes. I changed it from four backends to one. Then Lighttpd believed the backend was overloaded:
2010-03-01 07:16:21: (mod_fastcgi.c.2926) backend is overloaded; we'll disable it for 2 seconds and send the request to another backend instead: reconnects: 0 load: 131
2010-03-01 07:16:24: (mod_fastcgi.c.3569) all handlers for /haskell-json.fcgi on haskell-json.fcgi are down.
I doubt there were 131 requests to the FastCGI process at the same time but I'll have to seek more information about it. http://redmine.lighttpd.net/issues/1825
For now I've set it back to four backends as that never had this problem.
11 comments
[ 3.4 ms ] story [ 51.7 ms ] threadBy the way, if you like this, feel free to subscribe to http://haskell.reddit.com/
[1] http://www.reddit.com/r/haskell/comments/b7dil/try_haskell_n...
I've been having fun with it. I skipped to the exercises on page 20, and started reading and doing the exercises from there.
I downloaded a full platform install from: http://hackage.haskell.org/platform/
I wonder why.
2010-03-01 07:16:21: (mod_fastcgi.c.2926) backend is overloaded; we'll disable it for 2 seconds and send the request to another backend instead: reconnects: 0 load: 131 2010-03-01 07:16:24: (mod_fastcgi.c.3569) all handlers for /haskell-json.fcgi on haskell-json.fcgi are down.
I doubt there were 131 requests to the FastCGI process at the same time but I'll have to seek more information about it. http://redmine.lighttpd.net/issues/1825
For now I've set it back to four backends as that never had this problem.