It depends what level of abstraction you're working at. At the level of sockets, you're right that a mail server receives a connection from a sender when a newsletter is sent, whilst an RSS aggregator initiates connections to each feed's source when it polls, even if there are no new articles. At a lower level, the server's doing polling to check for new connections, but that's a fast local operation so it doesn't break the higher-level abstraction.
I wouldn't say there's much difference from a user point of view, i.e. "clicking". Polling for feeds with an RSS reader is comparable to polling for new messages with a mail client. Both can be made fast by fetching mail and/or news in the background, either using a client or using an OS service.
You are quite correct. But for vast majority of real world users RSS is quite misterious piece of technology. Emails are ok, everyone 'knows' what the email is? But 'feed'. Can I save it for offline use? Can I forward it to friends and family?
I wouldn't say there's much difference from a user point of view, i.e. "clicking". Polling for feeds with an RSS reader is comparable to polling for new messages with a mail client. Both can be made fast by fetching mail and/or news in the background, either using a client or using an OS service.