[Chicago] Erlang to Python

Garrett Smith g at rre.tt
Sun Aug 7 23:28:36 CEST 2011


On Fri, Aug 5, 2011 at 11:19 PM, Brian Herman <bherma3 at uic.edu> wrote:
> http://code.mixpanel.com/2011/08/05/how-and-why-we-switched-from-erlang-to-python/

I'm looking forward to the next post: "Why We Stopped Asking Interns
to Rewrite Performance Critical Systems".

>From what I can see, the only reason for the "switch" is that no one
there knew Erlang.

But that's hardly a very fun assessment. So, at the intern's expense...

"This server needs to scale, which for Python means using asynchronous I/O"

Of course this is utter nonsense. But oh, the fine puzzles that derive from it!

"string processing is very frequently the limiting factor in networked
systems because you have to serialize data every time you want to
transfer it"

That, and waiting for *really* slow things like disks and networks.
But having blazing fast string processing is a fun problem to solve.
For interns.

Not that anyone should care because "string processing" is typically
not a bottleneck, but Erlang has a C-based JSON library that's just as
fast as the other C-based JSON libraries.

"Finally, we use a few stateful, global data structures..."

I read this a couple time before concluding it was actually a troll
script from one of the talking bear videos.

"I’ve learned a lot about how to scale a real service in the couple of
weeks I’ve been here."

Oh intern, would that I could return to my own youth -- rife with
sweet, blissful ignorance.

Garrett


More information about the Chicago mailing list