If this was a function to encode a dict into something then I would see your point and agree. urlencode is specifically designed to work within the domain or URIs. In this domain, it is acceptable to have an empty value for a key in a query string. None is a representation of nothing, empty, null, the absence of a value. Therefore you would expect a function in the domain of URIs to construct a valid URI component when you specifically tell it to use None. Valid is up to you, either ignore the key-value pair completely, or use key[=&] to represent the empty value.

Take Requests as an example that gets it right:

    >>> import requests
    >>> requests.get('http://www.google.com/', params={'key': None}).url
    u'http://www.google.com/'
    >>> requests.get('http://www.google.com/', params={'key': ''}).url
    u'http://www.google.com/?key='