ACM is a 501(c)(3) non-profit. It's not like they're profiteering (cough Elsevier). They have to cover costs somehow.

Isn't that what membership fees are for? If they came out and said "our entire publication catalog is free, but to cover the costs of hosting we are raising membership prices from 100$ to 105$ a year", I don't think any of their members would mind.

Also, if you don't have hosting capacities, just seed the yearly archives out as torrent files. I'm sure there are some nerds out there who will voluntarily build a frontend for that.

If everything was open access, then the memberships would have much less of a value proposition.

I've been a member since February 1971 (I was still in college), and signed up for the ACM Digital Library when it launched. So you can thank people like me for the existence of the ACM and the availability of all those papers that you can now read for free.

Thank you!

Yes, positive reinforcement is generally the most successful way to get people to do what you want.