comments draft 18

[Stefan Hepper <sthepper@hursley.ibm.com>]

Hi Rich,
here my comments on the updated draft:

3.10:
- "This specification defines a state distribution mechanism for 
navigational state. This mechanism exposes a portion of the Portlet's 
navigational state to the Consumer such that coordinated updates across 
multiple Portlets can be effected"
I think this should be more specific, like this:
"This specification defines a state distribution mechanism for 
navigational state. This mechanism exposes a portion of the Portlet's 
navigational state to the Consumer such that coordinated updates of the 
shared navigational state across multiple Portlets can be effected"

- why do we still have the wsrp:modifiedCoordinationContext event?

6.1.2:
- now that we removed the properties from session params, shouldn't the 
session id be required?

6.1.13:
- "The public portion of navigational state could be set by the Consumer 
even if not supplied by the Portlet."
seems like I missed that discussion, why would I get nav state that I 
don't set? What is the contract here? That I can get always any nav 
state that I've declared as public? Is the consumer required to always 
send me all nav state that I've declared?

6.11.2:
- if we leave this event in the spec we should further explain it. What 
is a coordination context? I couldn't find a definition for this in the 
spec.

Appendix D:
- why are the session properties removed completely? Isn't that also a 
coordination mechanism between consumer and producer to coordinate on a 
user session concept?


Stefan