A few scientific journals have started implementing a "publish, then review" model of publishing. It is explained well in this article: Article Implementing a "publish, then review" model of publishing
Are there any reasons why this model should not become the main one in the future?