RE: Objectives and [EFM] Banana networks
Jonathan,
I would be very interested to find out how many of these have other,
competition, service providers that have direct access to their customers
through their infrastructures, and what the services are for those
competitive services.
Thank you,
Roy Bynum
At 05:48 PM 12/23/2002 -0800, Jonathan Thatcher wrote:
>Roy,
>
>We have a fair number of customers providing these services now. When the
>holiday is over, I will look to see which of these, if any, are going
>public with what they are doing. Most are not and will not.
>
>There are quite a few others who have gone out public ally with plans.
>But, these are plans, not implementations.
>
>Of course, these are not on 802.3ah technology. But, the difference will
>be as small as changing out a pluggable transceiver with another pluggable
>transceiver. In some cases, it may not even require that as we expect the
>current transceivers, ultimately, to be compliant to the standard when it
>is complete.
>
>As you are aware, much of OAM can be implemented in software.
>
>jonathan
>
>| -----Original Message-----
>| From: Roy Bynum [mailto:rabynum@mindspring.com]
>| Sent: Monday, December 23, 2002 2:59 PM
>| To: Jonathan Thatcher
>| Subject: RE: Objectives and [EFM] Banana networks
>|
>|
>| Jonathan,
>|
>| You wrote: "Who give a #@$%& what Cisco uses as definitions?"
>|
>| I think that you should care what definitions Cisco uses. If
>| there is not
>| a specific meaning of a technical term to the exclusions of
>| other meanings,
>| then any use of any terms is an attempt manipulate and to
>| create a false
>| perception in the minds of the audience. The whole purpose
>| of "standards"
>| is to create specific meaning to specific technology. Working in
>| "standards" as you are, I should think that you would care very much.
>|
>| I used the Cisco site because it was easy to find on the web. I have
>| several different GOATs (Glossary of Acronyms and Terms) in
>| hard copy that
>| I could have scanned in and sent at attachments, all saying
>| basically the
>| same thing. I thought the audience to this thread would
>| appreciate the web
>| site more than the scanned images.
>|
>| you said: "Then you state: This means that 802.3ah will only
>| support one
>| $ubscription service provider, the owner of the physical
>| facilities, what
>| ever diversity of services they may offer.
>|
>| Nope. Simply not true. This is happening all over the world,
>| literally."
>|
>| I hope that I am wrong, but I have not run across any and I
>| made a very
>| hard search concerning that issue. The real problem is
>| finding a service
>| provider that has implemented 802.3ah, since it has not yet
>| been finalized.
>|
>| At present, there is only one noted success of a service
>| provider that is
>| using primarily 802.3 as the L2 transport, a company in
>| Sweden that has
>| representatives at 802.3ah. It would be interesting to find
>| out if they
>| have unbundled their access and core network to provide
>| customer access to
>| other service providers and if they have what type of services those
>| are. Other than some sort of "packet" or "content" services,
>| I am sure
>| there are none. In the US, competition has pretty much
>| eliminated the
>| "legacy free service providers" that were using 802.3 as
>| their primary L2
>| because they, for the most part, did not originate their own services.
>|
>| Thank you,
>| Roy Bynum
>|
>|
>| At 12:33 PM 12/23/2002 -0800, Jonathan Thatcher wrote:
>| >Roy,
>| >
>| >Why, oh why?
>| >
>| >Roy, you make some excellent points. You always have. But,
>| then you add
>| >comments that fly in the face of common sense, and in doing
>| so, negate all
>| >the good things that you bring to the table.
>| >
>| >Who give a #@$%& what Cisco uses as definitions?
>| >
>| >You state: ...802.3ah does not provide for multiple
>| independent signals,
>| >"channels", the physical facilities can not be "unbundled".
>| >
>| >Okay, if you mean that you can't unbundle something that
>| wasn't bundled in
>| >the first place.
>| >
>| >Then you state: This means that 802.3ah will only support
>| one $ubscription
>| >service provider, the owner of the physical facilities, what ever
>| >diversity of services they may offer.
>| >
>| >Nope. Simply not true. This is happening all over the world,
>| literally.
>| >
>| >Regarding: Any newly funded facility that has the political will of
>| >supporting multiple service providers with equal access and
>| competition
>| >would NOT deploy EFM technology.
>| >
>| >They already have. They will continue to do so.
>| >
>| >No response required.
>| >
>| >jonathan
>| >
>| >
>| >
>| >| -----Original Message-----
>| >| From: Roy Bynum [mailto:rabynum@mindspring.com]
>| >| Sent: Saturday, December 21, 2002 5:40 PM
>| >| To: Geoff Thompson
>| >| Cc: Hugh Barrass; Sanjeev Mahalawat; ariel.maislos@passave.com;
>| >| 'Mccammon, Kent G.'; Thomas.Murphy@infineon.com;
>| Vipul_Bhatt@ieee.org;
>| >| wdiab@cisco.com; cribeiro@mail.inet.com.br; stds-802-3-efm@ieee.org
>| >| Subject: Objectives and [EFM] Banana networks
>| >|
>| >|
>| >|
>| >| Geoff,
>| >|
>| >| In the tread "Banana networks" you said: " I disagree here. I
>| >| believe that
>| >| Ethernet is "broadband in the protocol sense." The difference
>| >| is that it is
>| >| "flexibly broadband". Ethernet's main value as a trunking
>| >| mechanism is its
>| >| ability to make its total bandwidth available to all comers
>| >| based on their
>| >| instantaneous demand (up to the total available bandwidth, of
>| >| course)."
>| >|
>| >| By that reasoning, 10Base5 would also be considered as
>| >| "broadband" and
>| >| should actually be named 10Broad5. Also by this reasoning,
>| >| that all of the
>| >| existing Ethernet standards already support broadband
>| >| $ubscribtion services
>| >| and there is nothing distinctive about 802.3ah in meeting the
>| >| objectives
>| >| that would fit the criteria of "Distinct Identity" in that
>| >| there already
>| >| are several 802.3 standards that meet the first objective.
>| >|
>| >| Of course, in the actual wording of the initial objective, "Support
>| >| $ubscriber Access Network Topologies", the word "network" is
>| >| used, not the
>| >| word "services". This limits the definitions to the
>| >| physical facilities
>| >| and how they would be used.
>| >|
>| >| I refer you to
>| >| http://www.cisco.com/univercd/cc/td/doc/cisintwk/ita/b12.htm
>| >| , a glossary on a vendor's web site, under the term
>| >| "broadband" there are
>| >| four definitions:
>| >| "Describes facilities or services that operate at the DS3
>| >| rate and above.
>| >| For example, a Broadband DCS makes cross-connections at the
>| >| DS3, STS-1, and
>| >| STS-Nc levels. Similarly, Broadband ISDN provides about
>| 150 Mb/s per
>| >| channel of usable bandwidth.", under one heading of
>| "broadband", and
>| >| with another heading of "broadband":
>| >| "1. Transmission system that multiplexes multiple independent
>| >| signals onto
>| >| one cable."
>| >| "2. Telecommunications terminology: Any channel having a
>| >| bandwidth greater
>| >| than a voice-grade channel (4 kHz)." (This is actually
>| >| incorrect, since a
>| >| voice channel is referred to as "narrowband" and a T1 to DS3
>| >| is referred to
>| >| as "wideband".)
>| >| "3. LAN terminology: A coaxial cable on which analog
>| >| signaling is used. An
>| >| RF system with a constant data rate at or above 1.5 Mbps.
>| Also called
>| >| wideband. Contrast with baseband." (Other glossaries make the
>| >| distinction
>| >| that broadband signalling consists of multiple RF frequencies.)
>| >|
>| >| Under the term "baseband" is:
>| >| "Characteristic of a network technology where only one
>| >| carrier frequency is
>| >| used. Ethernet is an example of a baseband network. Also
>| >| called narrowband.
>| >| Contrast with broadband."
>| >|
>| >| Using these definitions, Ethernet, including 802.3ah qualifies as
>| >| "broadband" relative to the bandwidth, but does not qualify
>| >| as "broadband"
>| >| in any other sense.
>| >|
>| >| Since 802.3ah does not provide for multiple independent signals,
>| >| "channels", the physical facilities can not be "unbundled".
>| >| This means
>| >| that 802.3ah will only support one $ubscription service
>| >| provider, the owner
>| >| of the physical facilites, what ever diversity of services they may
>| >| offer. The ability for customers to link to different or
>| >| multiple service
>| >| providers over the same facility is not supported. As such,
>| >| 802.3ah is
>| >| specific to only the incumbent providers, or one specific
>| >| provider, not the
>| >| industry as a whole. Any newly funded facility that has the
>| >| political
>| >| will of supporting multiple service providers with equal access and
>| >| competition would NOT deploy EFM technology. In this, I
>| >| would question
>| >| whether it meets the criteria of "numerous users" in that is
>| >| only meets
>| >| the requirements of a limited number of users. (Remember that
>| >| the "user"
>| >| here is a service provider not the customers of that service
>| >| provider.)
>| >|
>| >| Thank you,
>| >| Roy Bynum
>| >|
>| >|
>| >|
>| >|
>| >|
>| >|
>|
>|