dev@javaserverfaces.java.net

Re: [REVIEW/632] Server-side state saving sensitive to model changes

From: Michael Youngstrom <youngm_at_gmail.com>
Date: Thu, 6 Sep 2007 16:00:28 -0600

My only problem with making serializing not default is that I'd hate
for the RI to be smoked by some other JSF impls out there in random
benchmarks because they're improperly handling server side state. :)

However, for apps in my industry segment stability and consistency are
more important than performance so I agree that server state should be
serialized by default.

Mike

On 9/6/07, Ryan Lubke <Ryan.Lubke_at_sun.com> wrote:
> Ryan Lubke wrote:
> > Change bundle attached to the issue [1]
> >
> > The big question: Should serialization of server state be enabled by
> > default or not?
> My two cents:
>
> From a technical standpoint, having parity with client-side state
> saving is a good thing.
> We don't want two different behaviors depending on which type a client
> application uses.
>
> That said, this is the first report I've seen on this issue, but that
> doesn't mean people haven't
> hit it and worked around it. So, I'm fine with changing the default to
> false and documenting
> it's usage so if someone hits this in the future, then can flip this switch.
>
>
> >
> >
> > [1] https://javaserverfaces.dev.java.net/issues/show_bug.cgi?id=632
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
> > For additional commands, e-mail: dev-help_at_javaserverfaces.dev.java.net
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe_at_javaserverfaces.dev.java.net
> For additional commands, e-mail: dev-help_at_javaserverfaces.dev.java.net
>
>