cloud computing is relatively cheap, could't the universities themselves or labs host their publications? I mean, the contents of the paper should be enough by

cloud computing is relatively cheap, could't the universities themselves or labs host their publications? I mean, the contents of the paper should be enough by itself to prompt momentum, otherwise is it worth publishing the paper at all?