[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: RE: [wsrp-wsia] [wsrp] initEnvironment and groupID - my final thoughts before we vote
Andre, the part of the spec I am refering to is: "The PortletSession and HttpSession objects share the same attribute set. Basically the PortletSession and HttpSession are the same object or at a minimum the PortletSession wraps the HttpSession. If a portlet session is created then a servlet session is also created." Best regards Carsten Leue ------- Dr. Carsten Leue Dept.8288, IBM Laboratory Böblingen , Germany Tel.: +49-7031-16-4603, Fax: +49-7031-16-4401 Andre Kramer <andre.kramer@eu. citrix.com> To wsrp-wsia@lists.oasis-open.org 10/10/2002 04:02 cc PM Subject RE: [wsrp-wsia] [wsrp] initEnvironment and groupID - my final tho ughts before we vote I don't believe that the JSR168 does actually require a HTTP session per application. Separate application and private session scopes are mandated but one HTTP session can be partitioned up locally. All the user's Portlets end up on one server though. But I would also disagree that JSR 168 mandates that application sharing MUST take place via the HTTP session. Developers see a Session object but it should be possible to implement the API spec without cookies and even Web App Sessions (e.g. using SAML to pass user context information). [At least this is my understanding as an JSR 168 expert group member. I view WSRP as being more than just for JSR 168 remoting and therefore will support both features, even if not strictly required as load balancing could be limited to be per-user.] -----Original Message----- From: Carsten Leue [mailto:CLEUE@de.ibm.com] Sent: 10 October 2002 13:58 To: wsrp-wsia@lists.oasis-open.org Subject: [wsrp-wsia] [wsrp] initEnvironment and groupID - my final thoughts before we vote Hi everybody. I want to send around some of my thoughts on initEnvironment and groupID before we vote on that issue today: - I understand that we can live without a groupID if we restrict ourselves to what is currently possible in JSR 168, i.e. sharing information between portlets on a user/application level. But we will not be able to share state on application instance levels. - there is the possibility to implement this scenario in a cluster with parallel access WITHOUTgroupID, initEnvironment and cookies (see Andre's description some weeks ago) - JSR 168 mandates that application sharing MUST take place via the HTTP session. Only for this reason we need to define cookie handling in the WSRP protocol and introduce initEnvironment. This leads to a protocol specific aspect of WSRP. - the use of initEnvironment also mandates that the consumer must have the information what entities form an application (to call initEnvironment for each app). To call initEnvironment just per user and producer would not be an option as JSR 168 mandates that each application has its own distinct HTTP session. - the only way to rely on initEnvironment without having metadata that describes applications would be to limit a producer to having one single application. I tried to state facts neutrally. Can everybody agree that the statements are correct or did I oversee important facts? If so I can vote happlily today. We should vote on two issues: - groupID yes/no - app description metadata yes/no Best regards Carsten Leue ------- Dr. Carsten Leue Dept.8288, IBM Laboratory Böblingen , Germany Tel.: +49-7031-16-4603, Fax: +49-7031-16-4401 ---------------------------------------------------------------- To subscribe or unsubscribe from this elist use the subscription manager: <http://lists.oasis-open.org/ob/adm.pl> ---------------------------------------------------------------- To subscribe or unsubscribe from this elist use the subscription manager: <http://lists.oasis-open.org/ob/adm.pl>
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC