urllib2 is closer to curl. It’s closer to a real client—and for many of us, makes much more sense than doing 4 or 5 steps to request an url and read the response.

httplib is a lower level library. As it says in the docs: “It is normally not used directly — the module urllib uses it to handle URLs that use HTTP and HTTPS.”

Because Python folks love abstraction, and so do I. It’s a lot easier to deal with, and just like @andrew said it’s a lot closer to a real client like curl, which is mostly second nature for me because I mostly use Unix and it’s hard to avoid when you’re programming with PHP.

Okay, I guess I’m mostly just complaining about how messily the abstraction is implemented in urllib2. Timeouts aren’t always handled correctly (although they seem to be with httplib), and so many third-party libraries that I’ve seen utilizing it don’t properly catch the exceptions it raises (possibly because they’re buried away in the middle of the document instead of having a heading and section of their own; like most other docs).