The first rule says to use URIs as "names for things". There are many URI schemes available, for example
The second rule says
- to use the
http URI scheme (¹), and
- to provide some information when accessing this URI (²).
A HTTP URI acts as identifier/name and locator/address, i.e., it’s a URL. If you’d use a tag URI like
tag:example.org,2014-01-01:foobar or a mailto URI like
mailto:email@example.com, no one could look it up (e.g., by entering it in a browser’s address bar) and learn something about it, as it’s just an identifier, not a locator.
So in other words, the second rule makes sure that the data is part of the Web.
https, while not explicitly mentioned, is probably fine, too
(²) and the third rule says to provide this information also using standards like RDF and SPARQL