[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [wsrm] Rev "d" of Rel52, 57
>I guess as long as the sender does not have to use the group termination
>timeouts I am happy.
<JD> they are optional... They may make sense for some application, and not for others.
There just provide more ways to limit groups, because not all apps will be able to terminate their
group with an "end" message.
I see them as helping implementations to better manage groups.
I don't think we can retire a group just based on ExpiryTime of its messages,
because it is an orthogonal notion: you may have groups with very sparse messages,
where all previous messages expire before the next one is sent. That should be allowed.
>However, we need to clarify that the receiver must be prepared to deal
>with them it it receives them.
<JD> this is implied in proposal for Rel52d.
>Also the receiver must be prepared to have them not be present for a group.