The users are not at fault here. Even a SSH or VPN will leave them vulnerable to attacks. Companies (Facebook, Twitter, etc.) have to increase their own security, because they are the only ones that can fix this problem.
This difference -- from random anonymous stranger whose only invested in software, to physical infrastructure with paid staff -- is also one reason bank phishing attacks happen via websites and not actual storefronts made to look like real banks.
If the only threat to Twitter and Facebook users was ISP-gnomes, the websites could put off fixing the issue for another decade.