At least for computer scientists, one alternative is to publish at conferences which are either run by Usenix, or where Usenix is a co-sponsor, since at least U

At least for computer scientists, one alternative is to publish at conferences which are either run by Usenix, or where Usenix is a co-sponsor, since at least Usenix requires that the papers be released on the Usenix website w/o any kind of ACM-style b.s. paywall. At this point, my personal policy is simply to refuse submit or assist with any conference where papers are held hostage behind a paywall, whether it is ACM or some other organization. Fortunately I'm not an academic, and so I don't have to get tenure, so it's relatively easy for me to have this policy. But if enough people do this, maybe ACM will either (a) get this message, or (b) disappear, like the dinosaurs....