A great thank your for writing a response, Jacob! I already felt that most users on this forum aren't interested and don't care about the feature request, even though it's a feature not freshly conceived by me, but a feature which is already available and proven in Spring Web Flow. I'll read your post later this evening and then think about your use of nested conversations (the master-detail editing). Perhaps I'll understand then why the output-mapper function still misses in Seam. PS: Your workaround for transferring state to the parent conversation state, with a non-Seam POJO placed in parent conversation scope, is smart. One of the simple workarounds I tested was to "fake" a write access to the parent conversation at the end of a nested conversation. This can be done, given that the underlying store for the Conversation context is the Session scope (entries for a certain conversation are stored with a special generated key that includes the conversation id). If we know the format of this key and generate this key ourselves, we can write to the parent conversation scope. It should only be kept in mind that by doing this we completely bypass Seam's support for Entity passivation (when an entity instance or a collection of entity instances is regularly saved in conversation scope, ie. not with the described fake write, Seam wraps the Entity instance into a special wrapper class such as org.jboss.seam.contexts.EntityBean or org.jboss.seam.contexts.EntityBeanList and then saves this wrapper class instance). View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4085545#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...