I know my incredimail will bounce the email, then block them. maybe the other side will think when it's returned that my email is no longer in use.
Yep, that's one of the ways...
Unfortunately, that works only with the "legitimate" (bona-fide companies that like spamming its customers) and "semi-legitimate" (legitimate companies who use/purchase harvested addresses) spammers. I.e. those who use mailing list distribution tools/service (e.g. services provided by Constant Contact) that track bounces. Those actually have the special "bounce-to" address that encodes the recipient's address.
While the "bounce" method usually works with them, the irony is that with the same companies, it is usually possible to unsubscribe (even though it is a waste of time and energy).
The most notorious spammers (where you have no chances unsubscribing) do not use any special bounce-to addresses, and use fake "From" addresses: Usually those addresses either do not exist, or they are randomly taken from the harvested list of addresses. As a result, those bounces are a nuisance to the owners of those addresses or to the sysadmins of the mail servers, especially when the spammers are using compromised computers/accounts to send their spam. (Just for the record: I am not saying that those bounces shouldn't happen.)
In my opinion, server-based integrated anti-spam tools are more effective, especially when they are using adaptive filters that learn from what you mark as Spam. (I believe that Gmail is actually good at that.)
What makes that effective is that the incoming message can be blocked from being accepted by the recipient's mail-server, and that is (1) more effective (it blocks it for multiple users on that server at once, hence minimizing the waste of resources) and (2) more obvious for the sender that the address is not accepting their messages.
