The saveState feature seems interesting. I will definitely need to check it out. One thing I am wondering about is how this feature might interract with something Craig is doing with Struts Shale.
In Struts Shale there is an optional interface that you can have your backing beans implement. This interface includes an init method (can't remember the exact name of the method offhand) which will be guaranteed (by Shale) to be called right after the bean is created by the managed bean facility. (At least I think that is right, Craig can probably explain it better than I.) Anyways, it would be interesting to know if there is a potential conflict between these two features. I'm thinking that there wouldn't be much of a problem because if you were saving the state of the bean on the client, then you would just choose not to make that bean implment the special interface that provides the extra lifecycle methods. Just something to think about. I don't know enough about Shale or saveState to tell you one way or another. I do think that some of the uses for saveState might be handled by having the initialization functionality of Shale. sean
