I don't think it's accurate to group in ACM and IEEE with Elsevier. Taking from something I said last week, the ACM and IEEE are professional organizations. Their main purpose is to represent the interests of the community of professionals in computing. If they stop charging for articles, they can still exist as an organization. Their members just need to elevate the issue to the point that the larger organization changes its policy.
Elsevier, on the other hand, will cease to exist if they stop charging. Their main purpose - their business model - is to charge for access to their journals. This business model is no longer necessary, and these companies will eventually die.
A world where knowledge is locked behind several $200/year paywalls is better than a world where knowledge is locked behind a greater number of $X,000/year paywalls. But it is not nearly as desirable as the kind of world we could have if everybody could link to, index, share, and discuss academic articles on the open web.
I agree with that. But you're only looking at them in terms of the price of the paywalls. That misses the point that the ACM and IEEE are organizations that the members can convince to do otherwise. That is not true with Elsevier.
Agreed entirely. Personally, I agree strongly with USENIX, which makes all their work publically available, and which requires only a simple grant of permission from authors, not a copyright assignment.
Elsevier, on the other hand, will cease to exist if they stop charging. Their main purpose - their business model - is to charge for access to their journals. This business model is no longer necessary, and these companies will eventually die.