This would not work. Reviewers would not feel comfortable being completely honest if they knew their reviews would be published publicly. In any case, there should be no need for the reviews to be public.
Reviewers should be prepared to stand by their reviews. If they're not prepated for it to be known that they said something in a review, then they should not say that thing. Anonymity and secrecy in reviewing doesn't help the field.
Why should the reviewers feel unconfortable? In the current system, reviews are made available to the paper writers, and that doesn't seem to inhibit the reviewers, despite the fact that one often has a good idea of who the reviewers are, especially in the more niche fields of science.
Even today reviewers run the risk of having their reviews made public, since nothing prevents the author from disseminating it. They are however normally anonymous.
