| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Dear IEEE 802 EC members, I am providing the attached, as IEEE P802.3cg Task Force Chair, regarding the process and the technical concerns Mr. Thompson has expressed. Mr. Thompson has voiced concerns over the scope and CSD of the IEEE P802.3cg draft and has asked the EC to consider blocking the progress of the draft to sponsor ballot. As the Chair of the IEEE P802.3cg Task Force, I would like to point out that Mr. Thompson has voiced these concerns since at least the initial working group ballot and the group has carefully considered them throughout the ballot process. His concerns have been aired both with considerable task force time, and in direct access to the larger IEEE 802.3 ballot pool. In all this, both the ballot resolution committee (IEEE P802.3cg Task Force) and the larger ballot pool have disagreed technically with Mr. Thompson’s conclusions. Architecturally, PLCA provides the physical layer’s inputs to the IEEE 802.3 Clause 4 MAC and performs functions, such as alignment, consistent with architecture in IEEE Std 802.3-2018, and existing within other IEEE 802.3 reconciliation sublayers. These are the roles delegated to the Physical Layer as a reconciliation sublayer, consistent with the PAR and CSD responses. The Physical Layer provides carrier sense, collision detect and alignment to the physical layer transmission characteristics, while the MAC provides frame-level addressing, error detection, and responds to the collision detect and carrier sense to avoid collisions, as per the text of IEEE Std 802.3 clauses 4.1.1 (MAC Overview) and 4.1.2.1.1 paragraph 2, which covers normal operation of the half duplex MAC. Additionally, the resulting sublayer has been shown to be compatible with and depend on the functionality of the detailed specification of the IEEE 802.3 MAC found in subclause 4.2. However, this is not, as Mr. Thompson asserts, a matter of implementation. It is easy to blur high-level distinctions of 'media access' functionality. However, as engineering distinctions, understanding the relationship between the physical layer functionality in PLCA and the MAC functionality in clause 4 required and benefitted from Task Force members' careful study of the IEEE 802.3 text and requirements, and are not high-level classifications. While the distinctions are subtle, individuals both by themselves and collectively in meetings have taken a significant amount of time and effort in considering Mr. Thompson’s concerns. References and more discussion may be found in the attached. The ballot pool and the ballot resolution committee have expressed overwhelmingly, disagreement with Mr. Thompson. Members of the group have listened and labored long and hard considering Mr. Thompson's objections and attempting to make clear the points of disagreement, referencing the text of IEEE Std 802.3, the history of other 802.3 projects, ISO/IEC layering models, consistency with other IEEE 802.3 reconciliation sublayers, compatibility with the physical layer portion of the text and the existing IEEE 802.3 MAC specification, as well as demonstrated compatibility with existing IEEE 802.3 networks and equipment. Selected references are in the attached. Based on these analyses, the ballot resolution committee (IEEE P802.3cg Task Force) has consistently reaffirmed their disagreement on the key technical issue - Mr. Thompson's assertion that the PLCA functionality constitutes a new MAC. As an individual, I believe it is not a proper exercise of our process for the EC to block the progress of a ballot on a technical disagreement which has been duly considered in the ballot process, especially after the ballot process has resulted in a clear decision of the ballot pool affirming the draft, the CSDs, and its scope. George Zimmerman, Ph.D. Chair, IEEE P802.3cg 10 Mb/s Single Pair Ethernet Task Force President & Principal CME Consulting, Inc. Experts in Advanced PHYsical Communications george@cmephyconsulting.com 310-920-3860 -----Original Message----- From: ***** IEEE 802 Executive Committee List ***** <STDS-802-SEC@LISTSERV.IEEE.ORG> On Behalf Of Law, David Sent: Friday, March 01, 2019 2:22 PM To: STDS-802-SEC@LISTSERV.IEEE.ORG Subject: Re: [802SEC] Scope and CSD re-approval of the P802.3cg project Dear IEEE 802 EC members, I will let the IEEE P802.3cg leadership address the technical concerns that Mr Thompson has. As IEEE 802.3 Chair I however wanted to make you aware of the steps that have been taken within IEEE 802.3 to ensure that the concerns raised receive as broad consideration as possible within the IEEE 802.3 Working Group. I also wanted to provide you with the current status of the ballot. Best regards, David ----- The concerns below have been presented to the Working Group balloting group (in substantially similar form) as unsatisfied comments since the initial ballot, please see <http://www.ieee802.org/3/cg/comments/index.html>. In addition, Mr Thompson was given the opportunity to present his concerns during the November 2018 IEEE 802.3 Working Group closing plenary meeting as part of the IEEE P802.3cg meeting report, please see <http://www.ieee802.org/3/minutes/nov18/1118_cg_close_report.pdf#page=12>. As a result of the discussion after the presentation at the November 2018 IEEE 802.3 Working Group closing plenary, the IEEE P802.3cg Task Force leadership agreed to provide a presentation at the January 2019 IEEE 802.3 Interim meeting series. This presentation provided background on the project and in particular the grounds for disagreeing technically with the assertions made by Mr Thompson. This presentation was arranged at a time when no other IEEE 802.3 activity was meeting to ensure all IEEE 802.3 meeting series attendees had the opportunity to attend without competing demands on their time. I encourage you to access this presentation at <http://www.ieee802.org/3/cg/public/Jan2019/Tutorial_cg_0119_final.pdf>, in particular, slides 28 to 33. The ballot results from the initial Working Group ballot through the 3rd Working Group recirculation ballot are provided in the table below. Please note that the IEEE P802.3cg Task Force held an interim meeting on 19th and 20th February 2019 in addition to meeting during the January 2019 IEEE 802.3 Interim meeting series. This has enabled two completed recirculation ballots, the 2nd and 3rd, since November 2018 when the concerns were presented at the IEEE 802.3 Working Group closing plenary. I have also included the current results of the ongoing IEEE P802.3cg/D2.4 10 Mb/s Single Pair 4th Ethernet Working Group recirculation ballot as an additional row to the table. As I announced to the IEEE 802 EC reflector, this ballot will close on Friday 8th March 2019 at 23:59 AOE <http://www.ieee802.org/secmail/msg23179.html>. Please note that every ballot notice included the text 'One of the responsibilities as a balloter is to ensure that draft is consistent with the criteria for standards development (CSD) responses which are available at <https://mentor.ieee.org/802-ec/dcn/18/ec-18-0079-00-ACSD-802-3cg.pdf>. An Approve vote indicates your agreement that the draft is consistent with the CSD responses.'. +------------+-----------+--------+----------+-----+-----+-----+--------+----------+---------+----------+ | Stage | Close | Voters | Returned | App | Dis | Abs | Return | Approval | Abstain | Comments | +------------+-----------+--------+----------+-----+-----+-----+--------+----------+---------+----------+ | Initial | 15-Aug-18 | 249 | 139 | 100 | 24 | 15 | 55.82% | 80.65% | 10.79% | 739 | | 1st recirc | 26-Oct-18 | 249 | 150 | 115 | 20 | 15 | 60.24% | 85.19% | 10.00% | 502 | | 2nd recirc | 18-Dec-18 | 249 | 156 | 131 | 9 | 16 | 62.65% | 93.57% | 10.26% | 351 | | 3rd recirc | 09-Feb-19 | 249 | 164 | 142 | 4 | 18 | 65.86% | 97.26% | 10.98% | 128 | +------------+-----------+--------+----------+-----+-----+-----+--------+----------+---------+----------+ | 4th recirc | 08-Mar-19 | 249 | 170 | 148 | 3 | 19 | 68.27% | 98.01% | 11.18% | 0 | +------------+-----------+--------+----------+-----+-----+-----+--------+----------+---------+----------+ ----- From: ***** IEEE 802 Executive Committee List ***** [mailto:STDS-802-SEC@LISTSERV.IEEE.ORG] On Behalf Of Geoff Thompson Sent: 28 February 2019 23:19 To: STDS-802-SEC@LISTSERV.IEEE.ORG Subject: Re: [802SEC] Scope and CSD re-approval of the P802.3cg project Colleagues- This is to complete my action item resulting from the following Conference Call item: 8.00 II Scope and CSD re-approval of the P802.3cg project -- a heads up Thompson 5 02:23 PM Thompson gave verbal update. There was discussion regarding Thompson’s update. I promised to provide material on the topic in advance of the EC Meetings in Vancouver so you would be prepared for discussion there. I asked to hold off on sending out such material until after the interim to see if any progress was made on my issues. That meeting was held last week. While some progress was made, things are not yet resolved. There were proposals at the close of that meeting to make some changes in the project documentation (PAR, CSD plus the project objectives which are internal to 802.3) but I haven't seen any proposals yet. THE PROJECT and TITLE P802.3cg Standard for Ethernet Amendment: Physical Layer Specifications and Management Parameters for 10 Mb/s Operation and Associated Power Delivery over a Single Balanced Pair of Conductors THE PROJECT SCOPE 5.2.b. Scope of the project: Specify additions to and appropriate modifications of IEEE Std 802.3 to add 10 Mb/s Physical Layer (PHY) specifications and management parameters for operation, and associated optional provision of power, using a single balanced pair of conductors. And, as per our 802 Operations Manual 9.2 The CSD statement shall be reviewed and approved by the WG and the Sponsor as part of the approval process for the following: • Forwarding the PAR to NesCom • As part of the WG ballot to assure the draft is consistent with the CSD • Forwarding the draft to Sponsor ballot • Forwarding the draft to RevCom It is this review for "Forwarding the draft to Sponsor ballot" that is expected to be on the EC agenda in Vancouver. The issue: I do not feel that the draft is consistent with the project documentation (PAR, CSD plus the project objectives which are internal to 802.3) THOMPSON ISSUES WITH THE PROJECT PAR Thompson asserts that the addition of PLCA (what the project has labeled: Physical Layer Collision Avoidance) belongs by statements elsewhere in 802 (Ref: 5.2.3) and 802.3 (Ref: 1.1.3) in the Media Access Control (MAC) Sublayer. Thompson asserts that since "PLCA" provides MAC functionality, it belongs in the MAC sublayer and is thus out of scope for the approved PAR The Task Force and text placement in the draft asserts that PLCA is in the Physical Layer and that is where it belongs (Largely, I believe, for ease of early implementation integration with existing silicon). CSD items General There was no mention whatsoever in the PAR, CSD or the project objectives that a new Media Access Control method would integral or even part of this project Broad Market Potential There is no mention of access method, old or new having an impact either way on the Broad Market Potential. That would seem to preclude introducing a new access method without notice. Compatibility The statement "The proposed amendment will conform to the IEEE 802.3 MAC" is not true. The Ethernet MAC is only used when the new MAC doesn't preclude its operation with its new access method either by disabling it or by a fault operation. Its half-duplex shared media operation is CSMA/CA, not the CSMA/CD with which Ethernet is so closely identified. Distinct Identity This one is tricky. I assert that the project is not asserting what should be its very distinct identity as something other than Ethernet. What is in the CSD is just the usual boilerplate for Physical Layer speeds and feeds. When the proposed network is in "PLCA" mode it has more in common with 802.4 or 802.12 than it does with Ethernet. I believe it needs a name and a separate standard to fully leverage its distinct identity. This is a challenge since 802.3 is the only wired group left in 802. For this reason I think development and maintenance of the new standard should operate within 802.3. POSSIBLE OUTCOMES THOMPSON'S ISSUES ARE UPHELD • Draft is modified to conform to project paperwork (this would be the case if PLCA is removed from the draft) • PAR & CSD are modified to describe the draft being developed THOMPSON'S ISSUES ARE NOT UPHELD • >50% of the EC decides that Thompson is wrong or his opinion doesn't matter or ??? • EC decides that this review is not procedural and removes this requirement because it is "technical" OTHER POSSIBILITIES • No decision, stalls start of Sponsor Ballot • Some compromise crops up which makes things better DISCLAIMER I have no ax to grind here about product. I care that we follow our rules, I think they are there for a reason. I care that Ethernet means Ethernet instead of any wired LAN. I do think that a PLCA standard should be developed with a distinct identity. REFERENCES: P802.3cg PAR http://www.ieee802.org/3/cg/P802_3cg_PAR_140518.pdf P802.3cg CSD https://mentor.ieee.org/802-ec/dcn/18/ec-18-0079-00-ACSD-802-3cg.pdf P802.3cg Objectives http://www.ieee802.org/3/cg/objectives_3cg_0318.pdf P802.3cg Draft http://www.ieee802.org/3/cg/private/8023cg_D2p4.pdf P802.3cg TF web area http://www.ieee802.org/3/cg/ Thank you, Geoff Geoffrey O. Thompson IEEE 802 Executive Committee GraCaSI S.A. Mountain View, CA 94043-5286 <thompson@ieee.org> +1.540.227.0059 ________________________________________ To unsubscribe from the STDS-802-SEC list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-SEC&A=1 ---------- This email is sent from the 802 Executive Committee email reflector. This list is maintained by Listserv. ---------- This email is sent from the 802 Executive Committee email reflector. This list is maintained by Listserv.
Attachment:
EC-802d3cg-PLCA-Thompson_issues_zimmerman.pdf
Description: EC-802d3cg-PLCA-Thompson_issues_zimmerman.pdf