Re: VCC cost models .... (was Re: Limits on SVCCs)

"Eric W. Gray" <gray@ctron.com> Wed, 01 May 1996 14:19 UTC

Received: from ietf.cnri.reston.va.us by IETF.CNRI.Reston.VA.US id aa16199; 1 May 96 10:19 EDT
Received: from guelah.nexen.com by IETF.CNRI.Reston.VA.US id aa16191; 1 May 96 10:19 EDT
Received: from maelstrom.nexen.com (maelstrom.nexen.com [204.249.97.5]) by guelah.nexen.com (8.7.3/8.7.3) with ESMTP id KAA04128; Wed, 1 May 1996 10:10:47 -0400 (EDT)
Received: (from root@localhost) by maelstrom.nexen.com (8.7.3/8.7.3) id KAA20309 for rolc-out; Wed, 1 May 1996 10:03:51 -0400 (EDT)
Received: from nexen.nexen.com (nexen.nexen.com [204.249.96.18]) by maelstrom.nexen.com (8.7.3/8.7.3) with ESMTP id KAA20300 for <rolc@nexen.com>; Wed, 1 May 1996 10:03:44 -0400 (EDT)
Received: from gatekeeper.ctron.com (ctron.com [134.141.197.25]) by nexen.nexen.com (8.7.3/8.7.3) with SMTP id KAA25297 for <rolc@nexen.com>; Wed, 1 May 1996 10:03:42 -0400 (EDT)
Received: (from news@localhost) by gatekeeper.ctron.com (8.6.12/8.6.9) id KAA01576; Wed, 1 May 1996 10:03:37 -0400
Received: from stealth.ctron.com(134.141.5.107) by gatekeeper via smap (V1.3mjr) id sma001532; Wed May 1 10:03:08 1996
Received: from olympus.ctron.com by stealth.ctron.com (4.1/SMI-4.1) id AA02519; Wed, 1 May 96 09:55:24 EDT
Received: from blarney (blarney.ctron.com [134.141.66.40]) by olympus.ctron.com (8.6.12/8.6.9) with ESMTP id KAA01367; Wed, 1 May 1996 10:03:03 -0400
Received: from blarney by blarney via SMTP (940816.SGI.8.6.9/930416.SGI) id KAA19037; Wed, 1 May 1996 10:03:12 -0400
Message-Id: <31876F1A.4691@ctron.com>
Date: Wed, 01 May 1996 10:03:06 -0400
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: "Eric W. Gray" <gray@ctron.com>
Organization: Cabletron Systems, Inc.
X-Mailer: Mozilla 2.0 (X11; I; IRIX 5.3 IP12)
Mime-Version: 1.0
To: schulter@zk3.dec.com
Cc: Keith McCloghrie <kzm@cisco.com>, Andrew Smith <fddi1-ncd@baynetworks.com>, dhc2@gte.com, gja@bellcore.com, rolc@nexen.com
Subject: Re: VCC cost models .... (was Re: Limits on SVCCs)
References: <9604302054.AA30356@dogfish.zk3.dec.com>
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Orig-Sender: owner-rolc@nexen.com
Precedence: bulk
X-Info: [Un]Subscribe to rolc-request@nexen.com, submissions to rolc@nexen.com
X-Info: Email archive at ftp://ietf.cnri.reston.va.us/ietf-mail-archive/rolc/
X-Info: Hypermail archive at http://cell-relay.indiana.edu/mail/archives/rolc/
X-Info: FTP archive at ftp://ftp.nexen.com/pub/rolc/

schulter@zk3.dec.com wrote:
> 
> > If the "real client" is an NBMA-attached host,
> 
> Hmmm.  Well, the "real client", as Grenville points out, is actually the user
> isn't it?  Regardless of what information the host/network is able to gather
> about the cost of a particular call, it should be up to the user to decide if
> each long-distance call should be placed (or any call that will incur any
> charge over and above the local call to the default router).  This decision
> would be based entirely on how much a user is willing to pay to access certain
> content or services.  Each user will have a different cost/benifit ratio for
> different content.  So I would think each user would want to know how much a
> call would cost, and be given the chance to decide if the cost is worth it for
> the target content or service.  The computer can't make this decision since
> this is really a value (quite literally) judgement.  There's also an issue of
> call volume even if the per-call cost is low (1000 HTML links per month
> resulting in long  distance calls at $0.20/call is rather significant). How
> do we expect users to manage this?
> 
> Would this mean something like a pop-up dialog window for each possible
> connection (like on every click on an HTML link)?  Very annoying.  I'm sure
> other possibly less annoying ways to handle this could be developed,
> but I think it would still be rather hard to manage.  And of course,
> cut-through of any type would have to be disabled when your 9 year-old was
> using your PC ;-)
> 
> > If the "real client" is, say, an Ethernet-attached host and there's a
> > router between the Ethernet and the NBMA network, then it's not clear
> > to me how the router can communicate "cost" information to the host,
> > but whatever method is used would seem to be independent of NHRP.
> 
> Well then, maybe it's the owner of the edge device that would make this
> decision.  I would guess that the owner of the directly connected device would
> be the one that would be billed.  My guess is that the owner would want to pass
> these costs back to the individual users, so now the "cut-through" decision
> is even more difficult to deal with (economically).  Do we now want routers
> and edge devices to deal with accounting? Or if not, then would the owner
> of the edge device want to incur any expenses that couldn't be properly
> passed on to the consumers?  If there is no clean way to bill this, I
> not sure if any owner of an edge device would ever allow the edge device
> to do cut-through.
> 
> Finally, all this is very different than the current Internet model.  Users
> don't have to worry about these things now, so why should they have to
> worry about them in the future?  My ISP rents bandwidth to his ISP for a flat
> rate and distributes the cost of that bandwidth to his users. I pay a flat
> rate every month to my ISP, a flat local call rate to NYNEX, and I don't have
> to worry about how much each click on a hypertext link is going to cost me.
> I can Internet to my heart's content knowing I will pay only the flat rate.
> It's pretty much the same here at work.  I imagine Digital pays a flat leased
> line rate too.  IMHO, this is exactly the model we should maintain for ATM.
> I don't think cut-through maintains this model and introduces too many costs
> and too much complexity for something very few people will be able to
> afford to use (until the day when I can get unlimited long distance
> calling from AT&T/MCI/SPRINT for $20/month).
> 
> Now cut-through on a private (local or campus) net is a different issue.
> 
>  --- pete
> 
> ------------------
> Peter Schulter                                  schulter@zk3.dec.com
> Digital UNIX Networking                         voice (603) 881-2920
> Digital Equipment Corp                          voice (DTN) 381-2920
> ZK3-03/U14                                      FAX   (603) 881-2257
> 110 Spit Brook Road                             FAX   (DTN) 381-2257
> Nashua, NH 03062

Peter,

	Often the "user" is not actually paying the bill.  Consequently, very often
a pop-up dialogue box would be neither appropriate nor appreciated.  In every case,
such a decision can be made by a properly "policy-instructed" computer (or other
device - such as a router) and - for that less-common user who wants to make a per-
instance choice - the policy might even include presenting the user with a pop-up.

	BTW, the "user" - sometimes referred to as the loose connection between the
keyboard and the seat - is frequently and uniquely unqualified to determine benefit
for a connection as well.  Just ask yourself how many of your users can distinguish
TCP and UDP...

--
Eric Gray