Following Fred's advice, this's copied only to wrec mailing list.
At 04:20 PM 12/20/00 -0500, Fred Douglis wrote:
>The next question is how WCIP relates to RUP, which is maybe what you were
>really asking. I'll let one of the WCIP authors answer that.
Here is what the discussion summary recorded (see
http://www.content-signaling.org/)
>2. Architecture
>
>Mark Nottingham described a conceptual architecture for RUP ("Resource
>Update Protocol") in his presentation at the official meeting.
>
>
>RUP components
> ---------------------
> metadata/state -- invalidation, delta/update, ...
> ---------------------
> resource selection
> ---------------------
> update message "NOTIFICATION?"
> ---------------------
> channel
> ---------------------
> subscription mechanism | reliability
> ---------------------
> announcment -- HTTP header, well-known location, ...
> ---------------------
>
>In later informal discussions, two main issues were raised:
>
>a) To what extent should the group's work focus on a "general
> framework" v. a specific protocol.
>
> In particular, there was some concern that focusing on a general
> framework could slow progress on the invalidation protocol (see
> discussion above), and that some of the specific protocols that Mark
> floated as solutions for some of these layers were not mature.
>
>b) How does this conceptual framework match the specific wcip
> proposal?
>
> The abstract design and specific instantiation were designed by
> largely different groups of people and with different immediate
> goals. A systematic comparison of the two designs may shed light on
> both.
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:21:29 MST