Thanks, Dan.
Yeah, I did remember it was somewhere in the reqs, but just
couldn't put my finger at exactly where during and right
after the meeting.
Joe Hui
Digital Island
===============================================================
> -----Original Message-----
> From: Dan Li [mailto:lidan@cisco.com]
> Sent: Tuesday, August 14, 2001 12:33 PM
> To: Joseph Hui; Mark Nottingham; webi@lists.equinix.com
> Subject: RE: RUP scope - invalidation or generalized metadata
>
>
> The payload extensibility is addressed in
> draft-ietf-webi-rup-reqs-01.txt
> section 6.2:
>
> "6. The protocol must define an extensible format for RUP
> messages which is
> capable of carrying a variety of payloads. Possible payloads
> include (1)
> cache invalidation, (2) content location update, (3) content prefetch
> hints, (4) removal and addition of resources to a resource group, (5)
> adjustments to cache consistency parameters, etc. While the
> above payloads
> may share the same RUP mechanism, it's not a requirement for
> the initial
> protocol to address all of them simultaneously. "
>
> The only thing is that it sort of limited the payloads to the 5 types
> instead of saying any metadata as future possibilities.
>
> Dan
>
> At 08:16 AM 8/8/2001 -0700, Joseph Hui wrote:
> >I'm for the payload concept for the same reason I was for having
> >"hooks and extensions" (for accommodating needs that are either
> >prorietary in nature or unforseeable now) in RUP by design.
> >
> >Joe Hui
> >Digital Island
> >========================================
> >
> >-----Original Message-----
> >From: Mark Nottingham
> >Sent: Wed 08/08/2001 04:22
> >To: webi@lists.equinix.com
> >Cc:
> >Subject: RUP scope - invalidation or generalized metadata
> >
> >
> >
> > It appeared that we've been moving towards
> consensus that RUP's
> > initial focus should be on an invalidation-specific payload
> >(based on
> > mailing list activity).
> >
> > However, there was significant feedback at the
> London meeting
> >that
> > while other payloads should not be defined, they should be
> > accommodated.
> >
> > In other words, it was felt that the design of RUP should be
> > separated into 'distribution' and 'payload'
> components, with one
> > payload (invalidation) defined. This would allow
> other efforts
> >(CDI,
> > for example) to define their own payloads (e.g.,
> metadata) and
> >reuse
> > the RUP distribution mechanisms.
> >
> > This message is intended to instigate discussion;
> our current
> >sense
> > of the list is that an invalidation-specific
> protocol is called
> >for.
> > If you agree with the latter views, as expressed at
> the meeting,
> > please participate.
> >
> > Cheers,
> >
> > --
> > Mark Nottingham, Research Scientist
> > Akamai Technologies (San Mateo, CA USA)
> >
>
>
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:22:59 MST