There's an exception to that ( from http://www.acm.org/publications/policies/RightsResponsibilities#dissemination http://www.acm.org/publications/policies/Right

There's an exception to that ( from http://www.acm.org/publications/policies/RightsResponsibilities#dissemination http://www.acm.org/publications/policies/RightsResponsibilit... ): Publication is only a part of the broader goal of disseminating ideas and results. Authors can expect ACM to contribute to this wider goal, and in particular to encourage dissemination in multiple forums. ACM expects authors to acknowledge ACM's contribution and not to publish the same material in other venues, except as permitted by ACM copyright policy. Thus authors can expect ACM to: * Allow a submission to be posted on home pages and public repositories before and after review * Allow an authors' version of their own ACM-copyrighted work on their personal server or on servers belonging to their employers * Allow metadata information, e.g., bibliographic, abstract, and keywords, for their individual work to be openly available * Allow authors the right to reuse their figures in their own subsequent publications for which they have granted ACM copyright * Provide statistics for each journal, transaction, and newsletter on its average turn-around time and its current backlog of articles. And ACM expects authors to * Appropriately acknowledge the publisher's effort * Ensure that whenever the authors or their employers provide a link to a personal copy that there is a link to the ACM definitive version * Ensure that all versions copyrighted by ACM bear the ACM copyright.