Here are my views:
1) Server sessions should be enabled by actual specific action of a developer or deployer
2) Client sessions should be enabled by the following model:
a) If the programmer repeatedly uses the same instance of the same stub then sessions will be *possible*.
b) Only if the server uses HTTP cookies or WS-A reference params the session will be *actual*.
i.e. for a session to take place, the consumer has to program in a certain way AND the provider has to enable sessions.
Paul
On 12/31/05, Dennis Sosnoski <[EMAIL PROTECTED]> wrote:
Sanjiva Weerawarana wrote:
>The question Dims asked is about what the default should be for clients.
>I disagree with the apparently popular choice of no sessions because if
>a service has multiple operations then in most cases the operations have
>some relationships between them. The question really amounts to asking
>how often do people have session scoped services vs. application scoped
>services. If they are application scoped then basically the cookie stuff
>makes no difference: either the service is totally stateless and it
>ignores all context or its truly stateful and remembers something from
>every request.
>
>IMO the natural behavior should be to maintain sessions by default.
>That's what even Apache SOAP did back many years ago.
>
>
I'll vote +1 on "no sessions by default" for now, just to keep things
simple.
That said, does WSDL 2.0 include a way to indicate service statefulness?
If so, that would be the ideal way of controlling the client defaults.
If not... well, is it too late to get a new feature in, Sanjiva? ;-)
- Dennis
--
Paul Fremantle
VP/Technology, WSO2 and OASIS WS-RX TC Co-chair
http://bloglines.com/blog/paulfremantle
[EMAIL PROTECTED]
"Oxygenating the Web Service Platform", www.wso2.com
