Re: [wsrp-interfaces] JSF-style systems and WSRP

[Michael Freedman <michael.freedman@oracle.com>]

It will be good to have him join.  By the way, to get him thinking and 
other experts who may be unsure of the problem, the question to ask is:
    Are there situations in JSF in which state is created or altered in 
the lifecycle phases up to and including Invoke Application, stored at 
request scope, and then referenced in the render phase?  If so, how 
common is this?

My examples of request scoped backing beans and handling validation 
errors were two situations our JSF expert identified which causes us to 
answer the first question yes.  Because of request scoped backing beans 
his answer to the second question was common enough.
     -Mike-

Subbu Allamaraju wrote:

> Thanks for reminding. I was preparing for today.
>
> I discussed this with our JSF lead, and he does not see the issues you 
> raised regarding producers supporting JSF.  Perhaps, this is a 
> question of use cases and not necessarily implementation. To make sure 
> we are not miscommunicating, I will ask him to join the call next week.
>
> Subbu
>
> Michael Freedman wrote:
>
>> FYI .. the call is next Wednesday, not today.
>>    -Mike-
>>
>> Michael Freedman wrote:
>>
>>> The primary discussion topic for tomorrows concall is the issues 
>>> created by portlet containers that are impelmented over wsrp which 
>>> have a concept of a request lifecycle that includes both the action 
>>> phase and the render phase.  JSF is one example.  When last we 
>>> discussed this I had asked folks who had access to in-house JSF 
>>> experts to discuss this internally to get a better 
>>> understanding/appreciation if a problem exists and if so what type 
>>> of problem (and how we might address it).  Did anyone get a chance 
>>> to do this?  If so can you please post your findings/conclusions 
>>> before the meeting so we have a basis for discussion?
>>>    -Mike-
>>
>>
>>
>>
>