More and more universities are paying for 'gold' (open) access publication. They're beginning to realise that you get more citations if people can actually read

More and more universities are paying for 'gold' (open) access publication. They're beginning to realise that you get more citations if people can actually read your papers without needing to go through the paywall. One of the advantages of publication is peer review. When you cite someone's work, if it's in a reputable journal then you can assume that someone reviewed it and would vouch for the quality of the research. You have to take it with a pinch of salt (reviewers are never totally unbiased and they make mistakes), but in fields where research is easy to verify like computer science, a publication in somewhere like Pattern Recognition holds a lot of weight. If you just cite a random paper that someone's self published on arXiv, you have no real idea whether what's being presented is true. Again, this isn't so much of a problem for CS papers where you can implement and test the work yourself, but for fields like physics where you're relying on other people's data, you want to be sure that someone smarter than you checked the numbers.