The peer review is supposed to be anonymous. So what happens in practice is you upload your paper to a central website from which a set of reviewers can downloa

The peer review is supposed to be anonymous. So what happens in practice is you upload your paper to a central website from which a set of reviewers can download a copy for review and give feedback with comments, anonymously. This is also why there is a login system: to ensure the comments are from actual peers which have their own papers up in the repository too. I'm not saying this is the best way to do it, just how it is now. Also, if you want someone to just read your paper, sure, email is fine. One does not exclude the other. The central system is there to ensure that should any doubt arise over the quality of the work, the feedback can be read and interpreted. Again, there might be many other more high-tech solutions out there to solve this problem.