Btw, ACM has started adding some drm banner (rightslink) to the papers. This seems like a recent addition. Why do these bodies (ACM/IEEE) exist ? I get the journal prestige angle and the loop that is hard to break, but for CS, most of it is conference driven. Any sub-field has about 50 people who call the shots. Why don't these 50 people decide not to be associated with ACM/IEEE ? What's in it for them ? Is it just apathy ? For example, the PC of a top conference (say ISCA) could just fork it for next year onwards (say to SCA) and remove the affiliation with ACM/IEEE. Since the sub-fields are quite small, there isn't a huge risk of loss of prestige or dilution. You follow where the PC goes. In general, the community decides that there are 3-4 top conferences in the area. Once you migrate those 3-4, you have taken back control. csrankings.org does a reasonable job of capturing the top conferences in every area. If you are on the PC of any of these conferences, I would like to know why this hasn't happened yet. Fortunately, some of the top ones (usenix ones, vldb, nips, etc.) are open access, and we should have all of them migrate.
