RE: [802.1] TGi use of OUI 00-00-00
Thanks Geoff,
I trust you will proceed to fix 802.2 and the 802 Overview and Architecture
document.
Dave H.
At 05:50 PM 10/5/2003, Geoff Thompson wrote:
>Mike-
>
>I have been late getting into this discussion because I haven't sat down
>to read all of the messages carefully.
>
>I will jump in now because some of the things that have been said have
>been completely inappropriate and I strongly feel that they needs
>correction. You have said some of them.
>
>I am not longer the Chair of the RAC but I believe that my following
>statements are immutable facts and not opinions that wax and wain with
>various chairs.
>
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>1) If an IEEE Standard or draft includes a value that it labels as an
>"OUI" then the ONLY entity that has the right to determine a new value to
>be included in that draft or standard is the IEEE Registration Authority.
>If the value is an already assigned value, then the right to determine its
>use in any application belongs to the entity to whom the ownership of the
>value has been assigned. The above statement is complete and absolute.
>
>2) If, by some screw-up, #1 has been violated then the screw-up should be
>fixed. Any proposed repair that proposes to continue an incursion into the
>use of an assigned OUI by an entity other than the entity that owns that
>OUI without the express written permission of the current appropriate
>designated agent of the owner of the OUI is not OK!
>
>3) If there is something in the universe that appears to the uninitiated
>to be an instance of a violation of #1 that instance provides no
>justification whatsoever for any violation of #1.
>
>4) If a corporate owner of an OUI chooses to make use of that OUI
>available to the standards community or the public, then such use should
>be specifically limited to whatever use the owner grants. The grant should
>be in written form. To preserve the terms and conditions of any such
>grant, it is a very good idea for that grant to be registered with the
>IEEE Registration Authority. However, whether or not an OUI owner chooses
>to do so is their own business.
>
>5) It is a VERY BAD IDEA for standards organizations to put OUI values
>into a draft before:
> a) their use of an established value has been approved by the owner
> b) their use of a proposed (supposedly unassigned) value has been
> assigned
> (The RAC has a policy against assigning desired values.)
> c) the content of the draft has stabilized (Submission to Sponsor
> Ballot is considered to be the appropriate point.)
>
>The entire point of having values come from a registration authority is to
>have values come from one and only one entity who assigns them
>unambiguously in a consistent manner and that the users respect those
>assignments. Any discussion or action to the contrary is not constructive
>to the very most very basic goals of having a registration authority in
>the first place.
>
>Sincerely,
>
>Geoff Thompson
>1st Vice Chair, 802
>Former Chair, IEEE RAC
>
>At 09:53 AM 10/2/2003 +0100, Mike Moreton wrote:
>
>>Tony,
>>
>>We can define a field any way we like. In this case it contains either
>>0:0:0 or an OUI. It can't contain an OUI of 0:0:0.
>>
>>OK, so Xerox can't use one of their many OUI values in this particular
>>context. So what? There's no Act of Parliament that says they have to
>>be able to.
>>
>>Are they complaining? No.
>>
>>Can we see any potential problems? No.
>>
>>Do we have lots of real problems to solve? Yes.
>>
>>Will changing this value reduce interoperability problems, or increase
>>them? Definitely increase.
>>
>>If people want to fix this problem for the future, then great. But
>>leave well alone for TGi.
>>
>>
>>Mike Moreton
>>Synad Technologies Ltd.
>>
>>
>>-----Original Message-----
>>From: Tony Jeffree [mailto:tony@jeffree.co.uk]
>>Sent: 02 October 2003 09:16
>>To: Mike Moreton
>>Cc: Johnston, Dj; David Halasz; stds-802-11@ieee.org; IEEE 802.1
>>Subject: RE: [802.1] TGi use of OUI 00-00-00
>>
>>Mike -
>>
>>Some observations:
>>
>>- Xerox clearly don't have the right to use the sequence 00-00-00
>>"...anywhere in the world", any more than (for example) Datapoint Corp
>>has
>>the right to use the squence 00-00-15
>>"...anywhere in the world". However, they do have a legitimate right to
>>use
>>those sequences anywhere in the world that an OUI value is called for,
>>as
>>they respectively own the OUI values 00-00-00 and 00-00-15.
>>
>>- We (the standards community) don't have the right to arbitrarily
>>assign
>>new meaning to an OUI value that has been legitimately allocated,
>>without
>>regard to the needs/wishes of the owner of that OUI value.
>>
>>- We don't have the right to reach any conclusions about what will or
>>will
>>not do Xerox any harm. Only Xerox has that right.
>>
>>- As DJ observes, the fact that 00-00-00 has been misused in previous
>>cases
>>does not create a valid precedent for us to further misuse the value.
>>
>>- Given the above, I believe there are (at least) the following
>>legitimate
>>solutions to this problem:
>>
>>1) Use some other OUI value to achieve the end that you have in mind
>>here -
>>possibly one specifically allocated by the RAC with the semantics "Null
>>OUI
>>value".
>>2) Use some other encoding to carry the semantics "this field does not
>>contain an OUI value". For example, given that OUIs will presumably not
>>be
>>allocated that would have the I/G bit set when used to generate MAC
>>addresses (this is the LS bit of first octet ), maybe this could be used
>>to
>>achieve the desired goal (comments please?).
>>
>>Regards,
>>Tony
>>
>>
>>At 08:15 02/10/2003 +0100, Mike Moreton wrote:
>>
>> >Dj,
>> >
>> >I can't accept that Xerox are the only company with the right to use
>>the
>> >sequence 00-00-00 for any purpose in any piece of software or hardware
>> >anywhere in the world.
>> >
>> >They may have been allocated that sequence when used as an OUI in a MAC
>> >address, but that doesn't mean they have any legal claim on the
>>sequence
>> >when used for a different purpose.
>> >
>> >The proposed use is a method for uniquely identifying proprietary
>> >security algorithms. Given Xerox have 11 other OUIs they could use for
>> >this purpose, I don't really think we need to worry about them running
>> >out any time soon.
>> >
>> >Using 00-00-00 doesn't do anyone (including Xerox) any harm. So why
>> >don't we just leave well alone and get on with more important issues?
>> >
>> >Mike Moreton
>> >Synad Technologies Ltd.
>> >
>> >
>> >-----Original Message-----
>> >From: Johnston, Dj [mailto:dj.johnston@intel.com]
>> >Sent: 01 October 2003 19:35
>> >To: David Halasz
>> >Cc: stds-802-11@ieee.org; IEEE 802.1
>> >Subject: RE: [802.1] TGi use of OUI 00-00-00
>> >
>> >
>> >I think Xerox would be the people to answer that.
>> >
>> >Just because it was done before doesn't mean its right. OUI 00-00-00 is
>> >assigned to Xerox, we don't have the right to just go and use it.
>> >
>> >DJ
>> >
>> >
>> >David Johnston
>> >Intel Corporation
>> >Chair, IEEE 802 Handoff ECSG
>> >
>> >Email : dj.johnston@intel.com
>> >Tel : 503 380 5578 (Mobile)
>> >Tel : 503 264 3855 (Office)
>> >
>> > > -----Original Message-----
>> > > From: David Halasz [mailto:dhala@cisco.com]
>> > > Sent: Wednesday, October 01, 2003 11:23 AM
>> > > To: Walker, Jesse; tony@jeffree.co.uk
>> > > Cc: Floyd Backes; mick_seaman@ieee.org; Johnston, Dj;
>> > > stds-802-11@ieee.org; IEEE 802.1
>> > > Subject: RE: [802.1] TGi use of OUI 00-00-00
>> > >
>> > >
>> > > I wasn't convinced yet that a new OUI was needed. I fail to
>> > > see how 802.11i
>> > > usage would be confused with RFC1042 or RFC1390. To the contrary, it
>> > > appears consistent that 00 00 00 be used.
>> > >
>> > > Stated another way, why wasn't it a problem for Xerox, when
>> > > 00 00 00 was
>> > > used in RFC1042?
>> > >
>> > > Dave H.
>> > >
>> > > At 01:08 PM 10/1/2003, Walker, Jesse wrote:
>> > >
>> > > >Floyd,
>> > > >
>> > > >No disagreement. My question is what OUI should we be using.
>> > > What is there
>> > > >in the spec was put there trhough our collective ignorance,
>> > > and we need to
>> > > >fix the problem.
>> > > >
>> > > >-- Jesse
>> > > >
>> > > > > -----Original Message-----
>> > > > > From: Floyd Backes [mailto:fbackes@propagatenet.com]
>> > > > > Sent: Wednesday, October 01, 2003 8:48 AM
>> > > > > To: Walker, Jesse; mick_seaman@ieee.org; Johnston, Dj;
>> > > > > stds-802-11@ieee.org
>> > > > > Cc: 'IEEE 802.1'
>> > > > > Subject: RE: [802.1] TGi use of OUI 00-00-00
>> > > > >
>> > > > >
>> > > > > I would rather TGi not use 00-00-00 because that OUI has
>> > > > > special meaning
>> > > > > in other uses, most notably RFC1042.
>> > > > >
>> > > > > I know that TGi has nothing to do with RFC1042. This fact
>> > > > > could be used
>> > > > > either as an argument in favor of, or against its use by TGi.
>>The
>> > > > > architect in me says it's a stronger argument against.
>> > > > >
>> > > > > Let's keep it clean and not overload 00-00-00 with yet
>> > > > > another meaning.
>> > > > > Why not ask the RAC to allocate another OUI specifically for
>>this
>> > > > > purpose?
>> > > > >
>> > > > > Floyd
>> > > > >
>> > > > > -----Original Message-----
>> > > > > From: owner-stds-802-1@majordomo.ieee.org
>> > > > > [mailto:owner-stds-802-1@majordomo.ieee.org] On Behalf Of
>> > > > > Walker, Jesse
>> > > > > Sent: Wednesday, October 01, 2003 9:44 AM
>> > > > > To: mick_seaman@ieee.org; Johnston, Dj; stds-802-11@ieee.org
>> > > > > Cc: IEEE 802.1
>> > > > > Subject: RE: [802.1] TGi use of OUI 00-00-00
>> > > > >
>> > > > >
>> > > > > Mick,
>> > > > >
>> > > > > 802.11i uses OUI 00-00-00 as a prefix indicating the
>> > > cipher suites and
>> > > > > authenticated key management suites it defines. What do
>> > > you suggest we
>> > > > > 802.1 as a replacement?
>> > > > >
>> > > > > -- Jesse
>> > > > >
>> > > > > > -----Original Message-----
>> > > > > > From: owner-stds-802-1@majordomo.ieee.org
>> > > > > > [mailto:owner-stds-802-1@majordomo.ieee.org]On Behalf
>> > > Of Mick Seaman
>> > > > > > Sent: Tuesday, September 30, 2003 6:11 PM
>> > > > > > To: Johnston, Dj; stds-802-11@ieee.org
>> > > > > > Cc: 'IEEE 802.1'
>> > > > > > Subject: RE: [802.1] TGi use of OUI 00-00-00
>> > > > > >
>> > > > > >
>> > > > > >
>> > > > > > David,
>> > > > > >
>> > > > > > The authority over the allocation of 00-80-C2 is 802.1. This
>> > > > > > OUI is and has been used to allocate standard group MAC
>> > > > > > addresses on request from any standards defining group
>> > > > > > (subject to certain procedural and status criteria which I
>> > > > > > won't go into here). In the past these allocations were
>> > > > > > published by ISO in a TR (Technical Report). They are now
>> > > > > > published by the IEEE and requests can be made to the IEEE
>> > > > > > RAC (Registration Authority Committee) which vectors (or
>> > > > > > rather the IEEE assigned staff do) the request to 802.1. If
>> > > > > > 802.1 ceased to exist then a replacement technical
>> > > > > > verification committee would be found and the assignment
>> > > > > > moved. Strictly speaking the allocation is not to "IEEE 802
>> > > > > > COMMITTEE" but to 802.1 for the purposes of 802. Tony and I
>> > > > > > discussed the fine print of the detail here during the course
>> > > > > > of last meeting, and he is straightening the record. The
>> > > > > > address given is for the previous chair of 802.1, Bill
>> > > > > > Lidinsky. It should be "802.1 chair" so the record does not
>> > > > > > need revisiting if and when the chair changes.
>> > > > > >
>> > > > > > All this means that you can't use 00-80-C2 as a dumy entry.
>> > > > > >
>> > > > > > I don't know what 00-50-C2 does, and I should. I am sure it
>> > > > > > is not a dummy entry. Tony??
>> > > > > >
>> > > > > > Geoff and I discussed 00-00-00, following my suggestion that
>> > > > > > it could be held to mean "no OUI present in this field".
>> > > > > > Geoff was reasonably concerned that this might infringe on
>> > > > > > Xerox's rights to fully use the value. On the other hand the
>> > > > > > value has already been assumed to have limited applicabiulity
>> > > > > > by its incorporation into the LLC SNAP SAP designator as the
>> > > > > > key for "an Ethertype follows". Don't expect further
>> > > > > > authorative clarification until the RAC meets again (probably
>> > > > > > November) and examines the precise wording of past
>> > > > > > correspondence on this subject.
>> > > > > >
>> > > > > > Mick
>> > > > > >
>> > > > > >
>> > > > > >
>> > > > > > -----Original Message-----
>> > > > > > From: owner-stds-802-1@majordomo.ieee.org
>> > > > > > [mailto:owner-stds-802-1@majordomo.ieee.org]On Behalf Of
>> > > > > Johnston, Dj
>> > > > > > Sent: Tuesday, September 30, 2003 5:05 PM
>> > > > > > To: stds-802-11@ieee.org
>> > > > > > Cc: IEEE 802.1
>> > > > > > Subject: [802.1] TGi use of OUI 00-00-00
>> > > > > >
>> > > > > >
>> > > > > >
>> > > > > > I'm a little concerned about the use of OUI 00-00-00 in
>> > > 802.11i. The
>> > > > > > IEEE OUI list http://standards.ieee.org/regauth/oui/oui.txt
>> > > > > shows this
>> > > > > > as belonging to Xerox.
>> > > > > >
>> > > > > > 00-00-00 (hex) XEROX CORPORATION
>> > > > > > 000000 (base 16) XEROX CORPORATION
>> > > > > > M/S 105-50C
>> > > > > > 800 PHILLIPS ROAD
>> > > > > > WEBSTER NY 14580
>> > > > > >
>> > > > > > There appears to be no OUI set aside to mean
>> > > "Everyone", "No OUI" or
>> > > > > > "Not Organizationally Specific".
>> > > > > >
>> > > > > > There are the following two IEEE OUIs. Their purposes are a
>> > > > > mystery to
>> > > > > > me:
>> > > > > >
>> > > > > > 00-50-C2 (hex) IEEE REGISTRATION AUTHORITY
>> > > > > > 0050C2 (base 16) IEEE REGISTRATION AUTHORITY
>> > > > > > 445 HOES LANE
>> > > > > > PISCATAWAY NJ 08855
>> > > > > >
>> > > > > > 00-80-C2 (hex) IEEE 802 COMMITTEE
>> > > > > > 0080C2 (base 16) IEEE 802 COMMITTEE
>> > > > > > FERMI NAT'L ACCELERATOR LAB
>> > > > > > M/S 368 P.O. BOX 500
>> > > > > > BATAVIA IL 60510
>> > > > > >
>> > > > > > Unfortunately, the IEEE list does not give a pointer to
>> > > > > > whatever defines
>> > > > > > the use of allocated OUIs, nor does it come with a
>> > > driver's manual.
>> > > > > >
>> > > > > > If either of these IEEE OUIs are appropriate, we should use
>>it.
>> > > > > > Otherwise we should either attach a bit to signify the
>> > > > > validity of the
>> > > > > > OUI in the table entry, or we should request an OUI for non
>> > > > > > organizationally specific uses in 802 standards. This is
>> > > > > assuming that
>> > > > > > Xerox do mind us using their OUI.
>> > > > > >
>> > > > > > 00-80-C2 appears to be used elsewhere to identify 802 frame
>> > > > > > types. E.G.
>> > > > > > RFC 1483
>> > > > > >
>> > > > > > So asking the following series of questions might resolve
>> > > > > the issue..
>> > > > > >
>> > > > > > 1) Is it OK to use 00-00-00 to mean "Non Organizationally
>> > > > > Specific" ?
>> > > > > > 2) If not, should we use 00-80-C2 instead?
>> > > > > > 3) If not, is there an OUI we should use for the purpose?
>> > > > > > 4) If not, then can we have an OUI for the purpose please?
>> > > > > >
>> > > > > > Regards,
>> > > > > > DJ
>> > > > > >
>> > > > > >
>> > > > > > David Johnston
>> > > > > > Intel Corporation
>> > > > > > Chair, IEEE 802 Handoff ECSG
>> > > > > >
>> > > > > > Email : dj.johnston@intel.com
>> > > > > > Tel : 503 380 5578 (Mobile)
>> > > > > > Tel : 503 264 3855 (Office)
>> > > > > >
>> > > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > >
>> > >
>> > > Dave Halasz
>> > > Cisco Systems, Inc.
>> > > 4125 Highlander Parkway
>> > > Richfield, OH 44286
>> > >
>> > >
>>
>>Regards,
>>Tony
Dave Halasz
Cisco Systems, Inc.
4125 Highlander Parkway
Richfield, OH 44286