This is a pretty common problem across the spectrum of scientific publishing. The best way to solve this is to dissolve groups like the ACM and make all scientific published facts, notes, journals, etc the legal property of those who founded the research.
In the mean time, the best way to stop this nonsense is to refuse to cite papers from publishers who won't give authors the right to publish their papers outside of their systems and who charge people to access their systems.
A bit more on the pro-activity front, I'd suggest that every time they send you a plea to join, you send back in their postage paid envelope a nicely typed letter detailing just how cold it will need to be in hell before you would every join or re-join...
Or explain the problem exactly and what changes would be required for you to join. If enough such letters came back to them, it might actually have an effect.
In the mean time, the best way to stop this nonsense is to refuse to cite papers from publishers who won't give authors the right to publish their papers outside of their systems and who charge people to access their systems.