I mention urllib because, as noted earlier in the discussion, code that
fetches resources over HTTP should really be caching (looking at the
Expires header, sending an ETag and an If-Modified-Since header, etc.).
 It's difficult for us to add these features to the urllib/urllib2 APIs,
so we should make users aware of other libraries that provide these
features,  so that users will use such a library instead of just writing
to the lower-level APIs provided with Python.