You can break out of the network effects with some effort. One of the top journals in machine learning nowadays is a community-run open-access journal that does

You can break out of the network effects with some effort. One of the top journals in machine learning nowadays is a community-run open-access journal that does not charge authors, JMLR (https://www.jmlr.org/ https://www.jmlr.org/). There's a bit about its model here: https://blogs.harvard.edu/pamphlet/2012/03/06/an-efficient-journal/ https://blogs.harvard.edu/pamphlet/2012/03/06/an-efficient-j... (previously discussed: https://news.ycombinator.com/item?id=15280736 https://news.ycombinator.com/item?id=15280736)