>Once upon a time there was no peer review in science. Instead, the editor made his own comments and edits, and then published it or not. This system was fast. You only had to know one person, the editor. Send him a mail (a letter, not email!) with your submission, and then wait for the reply. This is in fact how most of scientific history worked, and the great discoveries were published in this manner. Later in the 1960s or so, peer review won in for various reasons. Funnily enough, 1970 is just about when science started going to shit. No wonder. The future of scientific publication should be people uploading the content and metadata of their paper to a P2P network/index that anyone can use and contribute space or bandwidth to it. People would use a "paper reader" that renders the paper data in a consistent format of their choosing. Anyone (actual journal editor, scientist or layman) can be an editor by simply publishing a list of links to the global index as their curated collection, and if they do a good job choosing the papers people will follow them. The current journals can employ the same complex system of peer review to construct their own "premium" curation. And if this peer review really is that valuable people will surely pay to see it. They can't charge $30 for a PDF anymore if all the content is in a free public index, so they can justify their fees by paying the reviewers instead. The first part is almost there, between arxiv and libgen. The second part is sort-of happening through scientists publishing blogs, twitter and putting research on Github for e-fame. Hopefully it will continue and the current system of gatekeepery dies for good.