Re: CDNP naming

["Abbie Barbir" <abbieb@nortelnetworks.com>]

Eric Dean wrote:

> At least Content Distribution Internetworking is a bit different from
> Content Internetworking or Content Peering.  The difference being that a
> CDN would probably ensure that certain Distribution capabilities are
> supported by a "peered" or "internetworked" CDN before it actually
> delivered the content to surrogates within that CDN.

That is true, I do believe that there should be some form of a retaionship between
the distribution system and the direction system.
This also should translate to a uniform understanding among CDN about what is in
the surrogate (in terms of stale, policy, cache organization
etc..) . This information is needed such that a good proximity (includes the
ability of the surrogate to provide the data, as opposed to knowing another cache
in the network that could do) decision is made regarding the surrogate and the
client.

> It is for this reason that I may not in favor for exclusively proposing a
> DNS-based request mapping method unless everyone uses the exact same
> surrogates with exact capabilities.

Here, I have a similar obeservation. For this reason, I have proposed in a
previous e-mail to start talking about QOS  (may quality of delivery is a better
term) of peered CDN.

regards
abbie
________

>
>
> On Mon, 4 Dec 2000, Phil Rzewski wrote:
>
> > On a going-forward basis, I'd like to be a pest and say that I'm still not
> > 100% comfortable with the name... but perhaps no name exists that everyone
> > could like. In a separate e-mail, Eric Dean made a nice statement about why
> > Content Peering isn't a great term:
> >
> > "I do believe that Content Peering somewhat implies an exchange of content.
> > That may be one particular mode of an interconnect; however, there are also
> > other potential modes whereby references to content (URL) along with
> > certain characteristics about the content (Headers) are exchanged."
> >
> > As I read it, an ideal term would make room for exchange of both content
> > itself, as well as meta-data. Therefore, my response would be to go with
> > "Content Layer Internetworking", as it turns "Content" into an adjective
> > rather than a noun. Furthermore, by not naming any one specific thing being
> > internetworked (distribution, delivery, accounting) it makes room under the
> > umbrella for all these and more.
> >
> > But if Content Distribution Internetworking becomes the name, I just wanna
> > state for the record that I'll throw water on anyone that ever tries to
> > tell me any certain cross-network content-layer data exchange is "out of
> > scope" for the working group unless it somehow applies to distribution. :)
> >
> > --
> > Phil
> >
> >
> > At 10:26 AM 12/1/00 -0500, Mark Day wrote:
> > >I fear I may have provided bad guidance to Phil on this issue. I noticed
> > >just now that the agenda for our BOF has the name Content Distribution
> > >Internetworking already.  I had provided that as a possible alternative, and
> > >either the ADs or the secretariat must have chosen it in preference to our
> > >previous name.
> > >
> > >Since that name seems workable and accurate, I would suggest that we go with
> > >it rather than debating further.  And it can be part of the group lore as to
> > >why "Content Distribution Internetworking" has the acronym "cdnp".
> > >
> > >--Mark
> > >
> > >Mark Stuart Day
> > >Senior Scientist
> > >Cisco Systems
> > >+1 (781) 663-8310
> > >markday@cisco.com
> > >
> > > > -----Original Message-----
> > > > From: owner-cdn@ops.ietf.org [mailto:owner-cdn@ops.ietf.org]On Behalf Of
> > > > Fred Douglis
> > > > Sent: Friday, December 01, 2000 10:10 AM
> > > > To: Phil Rzewski
> > > > Cc: cdn@ops.ietf.org
> > > > Subject: Re: CDNP naming
> > > >
> > > >
> > > > Phil,
> > > >
> > > > I agree by and large with your summary.  And at one point, you suggested
> > > > "content internetworking" as an alternative.  In fact, I'm not
> > > > sure if that was
> > > > mentioned previously, but in one of the strawman calls, we had the same
> > > > discussion and I threw out "Content Distribution
> > > > Internetworking", intended as
> > > > tongue-in-cheek, but it seemed like people picked up on it.  While in the
> > > > drafts, it didn't end up taking hold, I'd already mentioned it to
> > > > the person
> > > > in charge of AT&T's CDN (from the business perspective) and I've
> > > > heard him
> > > > using that term in place of "peering" since then.
> > > >
> > > > I believe that one of the major stumbling blocks to the term
> > > > "peering" is the
> > > > relationship with settlements, and the question of whether it is
> > > > bilateral.
> > > > The CDNP BOF clearly covers cases where traditional "peering"
> > > > doesn't apply,
> > > > so I hope when it's chartered as a working group, the name better
> > > > reflects the
> > > > spectrum of possibilities.
> > > >
> > > > Fred
> > > >
> > > >
> >
> > --
> > Phil Rzewski - Senior Architect - Inktomi Corporation
> > 650-653-2487 (office) - 650-303-3790 (cell) - 650-653-1848 (fax)
> >
> >
>
> Eric Dean
> President, Crystal Ball Inc.
> W 703-322-8000
> F 703-322-8010
> M 703-597-6921
begin:vcard 
n:Barbir, Ph.D.;Abbie
x-mozilla-html:FALSE
org:Nortel Networks;613 763 5229
adr:;;;;;;
version:2.1
email;internet:abbieb@nortelnetworks.com
title:Senior Designer
x-mozilla-cpt:;0
fn:Abbie Barbir, Ph.D.
end:vcard