Résumé : Publication methods have been and are constantly evolving in Computer Science. My position, expressed in this paper, is that in this current landscape, there is room for a new publication model, combining OpenArchives for immediate access with editorial peer-reviewing. A key interest of this proposal is that peer-reviewing happens after the results have been released. I start by reviewing recent changes in publication methods in Computer Science and some of their consequences, drawing practical examples from Computer Graphics. I then review existing forces that are contributing to changes in publication methods, and how these forces push towards a new publication model. I describe this model, and review practical requirements to make it work.