While I am spewing out idea to help save the world from search engine abusers, I should also add that I am surprised that social networks have not yet been leveraged more to reduce spam. For example while I may not want my entire social network (especially 2nd and 3rd degree contacts) in my address book, I still might want them to be not marked as spam (or at least be considered less likely to be spam). Furthermore it would be kind of convenient to see where and to what degree I am connected to people right in my email client. Again this would help me weed out false positives or increase the chances of me reading the email more quickly.

That would work only if the sender identified with an email message were verifiable. Most spam uses a forged identity.

Even if you receive an email that appears to be sent by someone who is a first-order trusted person in your address book, or a second or third-order contact in your extended network, that's no guarantee that it's not spam.

I even sometimes receive spam that claims to be sent by my own email address.

So what you suggest could be helpful, but for it to work, first there needs to be a way to distinguish true sender identity from the easily-falsified email headers.

Right of course, but it would help. It would even help for legit mails.

Now the next step would be if social networks would start pushing GPG and things like that ..

@Bill: There is already - DomainKeys/DKMI, SenderID and SPF. People just have to implement it.