Unconfirmed Minutes
IEEE 802.3 Plenary
Vancouver, British Columbia
November 11-15, 1996

Andy Luque
Open Communications

Monday, November 11, Administrative
Geoff Thompson convened the meeting at 15:00 as IEEE 802.3 chair. Andy Luque, IEEE 802.3 secretary, recorded the minutes. Participants introduced themselves. The attendance sign-up list was circulated. The attendance list for the November 11-15 Vancouver plenary is included (Attachment A2).

The IEEE 802.3 voters list was presented (Attachment A2). The qualifications needed to become an IEEE 802.3 voter were reviewed. The potential voters list was presented (Attachment A2). Participants need to indicate if they want to become voters. The following were added to the voters list: Kevin Daines, Stefan Wurster, Mitsuji Okada, Amrit Kalla, Ben Speiser, Rabih Makarem, Hiep Tran, David Van Goor, Johannes Hansen, Joe Pelissier, Dave Fifield, Ian McKay, Douglas Thomson, Rich Taborek, Nand Aggarwal, Steve Augusta, Mehdi Hatamian, John Bestel, Brian Dugan, Richard Grenier, Vivek Telang, Carlos Tomaszewski, and Wen-Tsung Tang. The at-risk voters list was presented. None were removed.

The Enschede Minutes were distributed. They will be considered on Thursday. The Monday/Tuesday Agenda was presented (Attachment A1). The Flatman and Nikolich presentations were deleted. This information was well summarized at the IEEE 802 plenary.

Motion: Approve the Monday/Tuesday Agenda as modified.
M:Howard Frazier S:Andy Luque; Approved by acclamation.

Executive Committee Liaison - Geoff Thompson
Geoff reported on the actions of the IEEE 802 Executive Committee. The published ISO/IEEE 8802-3:1996 standard is being distributed this week. Most members received a copy in the mail after Enschede. Need to make a list of those qualified to receive a copy at Vancouver. The IEEE 802 Executive Committee considered future meeting sites and Sponsor ballots.

IEEE 802.3 Operating Rules - David Law
David reported on the IEEE 802.3 Operating Rules (Attachment A3). There is a requirement for IEEE 802.3 to develop a set of Operating Rules. This is important when conflict arises.

A letter ballot was mailed in January of 1996. At this time, we still need 20 more ballots to close. The comments will be collated and a new draft will be prepared. Wednesday, a one hour meeting will focus on a new draft and will make a recommendation regarding a future ballot. A report will be provided Thursday.

Executive Committee Liaison - Geoff Thompson
IEEE 802 is planning to take over the relevant ISO secretariat. This has some financial ramifications. 40% of our registration fees are used to operate this TAG.

The Fiber Optic PAR has expired. This TAG has been working for some time but a Recommended Practice has not emerged. A proposal has been made to extend the PAR. Need to make a decision on extending or withdrawing the PAR. A letter ballot would be required to disband the group.

Revisions to the Conditional Approval process are still under consideration. The current Rules have created problems in the past. So far, a review of the Rules has been inconclusive. A temporary Procedure is being renewed at every plenary until the new Rules are adopted. This issue will be discussed at a meeting Tuesday afternoon.

No tutorials are scheduled for Vancouver.

Some groups are planning to request the acquisition of PC overhead display equipment by 802.

IEEE 802.1 Liaison - Andy Luque
Andy reported on the status of IEEE 802.1. The IEEE 802.1Q, VLAN, and IEEE 802.1p, Enhanced MAC, efforts are of particular interest. According to the discussions Monday, only legal size MAC frames will be allowed. It does not appear likely that the MAC frame size will need to be increase. Additional octets are needed to carry the VLAN header. Nonetheless, the situations still needs to be monitored.

Progress continues to be made with regard to IEEE 802.1p. There was another WG ballot. The IEEE 802 Overview and Architecture revision is not moving as rapidly. There has been no progress since the last plenary.

There will be an IEEE 802.1 technical plenary Wednesday morning.

IEEE 802.12 Liaison - Pat Thaler
Pat provided an update on IEEE 802.12. The group has 5 active PARs.
o Redundant Links - LMSC Sponsor ballot.
o Full Duplex - new draft to be reviewed.
o High Speed - participate in IEEE 802.3z. No active work.
o 2 pair Cat 5 - no work at Vancouver.
o 802.1D - review draft at Vancouver.
Geoff Thompson, asked for details on the Full Duplex scheduled. Pat, not sure of the schedule. A WG ballot may occur after the March plenary.

Administrative - Geoff Thompson
The schedule for the week was briefly discussed. The Gigabit Ethernet presentations have been divided into four tracks. The task force will meet after the IEEE 802.3 plenary ends today for about one hour. The plenary will resume Tuesday at 8:30.

Tuesday, November 12, Administrative
The IEEE 802.3 plenary re-convened at about 8:35. Geoff Thompson chaired the meeting. Andy Luque, IEEE 802.3 secretary, recorded the minutes. The attendance list was circulated.

The voting rules were reviewed. The voters list was presented (Attachment A2). New qualified voters were solicited. Todd Vafiades and Steve Metzger were added to the voter list. The at-risk voter list was presented. No one requested to be removed.

No tutorials are scheduled. The regular social is planned for Wednesday evening.

ISO/IEEE 8802-3:1996 will be distributed this week. 70 copies are available. Copies were mailed to IEEE 802.3 voters. New voters may qualify for these books.

Steve Swanson's SWL presentation was removed from the Agenda. Not ready to do it yet.

Interpretation Request - David Law
An interpretation request has been received from LeCroy (Attachment A4). It concerns the 10BASE-T templates. Need to formulate a response. It will be reviewed Thursday and a decision will be made on the need for a letter ballot. A discussion is planned for Wednesday 13:00-14:00 in room 704.

IEEE 802.3 State of the Standard - David Law
David reported on the status of various IEEE 802.3 documents (Attachment A5). The 10BASE-5 PICS to be published in November 1996. The 1802.3 Conformance Test standard will be dropped.

SC25/WG3 Liaison - Alan Flatman
Alan reported on cabling standards (no Attachments). SC25/WG3 has not met since his report at Enschede. The next meeting will be the week of January 13 in Barcelona. A liaison letter has been received from SC25/WG3. Alan or Geoff will report on the letter. A reply is needed by Thursday.

Liaison TR.41 - Chris Di Minico
Chris reported on the TIA Ottawa meeting (Attachment A6). The Delay Skew specification was approved. It was modified from 50 nsec to 45 nsec per 100 meters of cable. This accounts for channel variability. It allows for temperature and connector deviations. The delay is specified in the standard and variability is allowed in an attachment. This approach avoids delays in completing the TR41.8.1-96 specification. The ballot will close on December 10.

What categories of cable are covered? Chris, it covers Cat 3, 4, and 5. It also applies to screened twisted pairs.

The group continues to look at patch cord and cross connecting hardware. Currently, there are no specifications in this area.

The group sent a letter to the ATM Forum regarding plastic fiber.

An effort is underway to harmonize the EIA568-A and the ISO 11801 standards. Some differences need to be reconciled.

A new document is being developed to deal with Cable Installation Practices. Relevant material will be extracted from EIA568-A.

A set of revisions to EIA568-A will be considered at a meeting next week.

Fiber Survey - Chris Di Minico
Chris reported on a fiber optic cable survey (no Attachment). This study complements the earlier work done by Compaq. It provides information to help converge on a distance specification. The survey was distributed to members of a professional installers group. Out of 2000, about 200 members responded. They included designers and installers instead of end-users. The results still need to be fully collated.

A more detailed presentation will be provided during the Gigabit Ethernet task force meeting.

Alan Flatman, are you recommending against plastic fiber? Chris, it is not a recognized category. Alan indicated that he has made little progress in his European cable survey. He is still looking for sponsors.

Gigabit User Survey - Rich Gardner
Rich reported on a Gigabit end user survey (Attachment A7). A survey was mailed to a large number of end users. Network redundancy is important to users. Conformance with standards is also important. The results have been weighted. The data is still being analyzed. There were several questions about the data collected by the survey. More details will be provided during the Gigabit Ethernet task force meeting.

Enhanced MAC Algorithm - Mart Molle
Mart reported on the status of the Enhanced MAC task force (Attachment B). Prototype BLAM adapters have been tested. Experimental data appears to validate the simulations.

The BLAM draft is being updated. Need to recognize the Full Duplex standard and the changes adopted in ISO/IEEE 8802-3:1996. The State Machines have been modified to clarify several areas per discussion with Dave Davies.

Substantial testing has been conducted on the BLAM adapters. A network has been set up with 10 BLAM adapters using Pentium machines running Linux. The test results indicated that:

1-It solves the capture effect problem.

2-Short term fairness has been improved.

3-It does not hurt classic stations in a mixed environment.

The plans for the week were discussed. The test results will be reviewed. The updated draft will also be considered. A plan will be formulated for the future.

100BASE-T2 - Scott Carter
Scott reported on the status of the 100BASE-T2 task force (Attachment C). Two ballots were conducted since the July plenary. No technical comments were received in WG confirmation ballot.

Draft 6 underwent a LMSC sponsor ballot that closed in late October. 90 ballots were mailed and 74 were returned for a 79% return rate. 66 approve and 4 disapprove ballots were received for a 89% approval rate. A total of 115 comments were submitted. 21 of these were technical. The comments will be reviewed Wednesday morning.

Geoff Thompson, any technical changes? Scott, none that are substantial. None of the issues raised affect interoperability. A report will be provided Thursday.

Full Duplex - Rich Seifert
Rich reported on the status of the Full Duplex task force (Attachment D). A LMSC sponsor ballot was conducted on draft 3. There were 78 approve, 3 disapprove, and 3 abstain ballots. A total of 161 comments were submitted. Several were technical.

The task force met Sunday evening to work on the comments. Four key technical submissions have been identified:

1-Increase the maximum frame size from 1518 to 1522 bytes to support VLANs. A note would alert the reader about future work that may modify the maximum frame size. This solution may be reconsidered based on the IEEE 802.1Q report on Thursday.

2-Two additional Management objects were requested. They would facilitate the detection of congestion. The request can not be accepted because these variables are not observable. These measurements can be undertaken at a Higher Layer.

3-Inconsistencies have been identified between the Managed Objects in the normative clause and the GDMO in the informative Annex. The sections will be reconciled, the GDMO is correct. 100BASE-T2 is also affected by this issue.

4-Change the definition of the pause_time quantum from a slotTime to 512 bit times. This would allow Gigabit Ethernet to use a different slotTime without affecting the PAUSE clause.

Three items are on the agenda for the week. Resolve the outstanding comment from the Sponsor ballot. Requests that the disapprove voters change their vote. Request that draft 3.1 go out for a LMSC confirmation ballot or to the IEEE Standards Board.

A short break was taken about 10:00 AM.

Gigabit Ethernet - Howard Frazier
Howard reported on the status of the Gigabit Ethernet task force (Attachment E). There was an interim meeting September 9-11 in Coeur d'Alene. 31 technical presentations were made on user requirements, repeaters, system issues, MAC/PCS/GMII, optical PMD, and copper PMD. The objectives were reviewed and the distance for SMF was extended from 2 km to 3 km. A timeline was adopted for the development of the Gigabit standard. The first draft is targeted for January.

Four parallel tracks are planned for this week to deal with the large number of technical presentations. The group will try to identify the key items that will go into the first draft.

100BASE-T Maintenance - Geoff Thompson
Geoff reported on the status of Maintenance (no Attachments). A total of 10 Maintenance requests have been received. One new request since Enschede. A meeting is planned for 17:00. A PAR request will be considered Thursday. Clarification is needed on some requests.

Administrative - Geoff Thompson
The task force schedule was discussed and meeting rooms were assigned. The Gigabit meetings will start at 12:30. Concern was expressed about the concurrent tracks. All major decisions will be made Wednesday afternoon in a task force plenary session.

802.3 Radio Link - John Cafarella
John Cafarella solicited interested parties to contact him about the development of a 10Mb/s IEEE 802.3 WLAN standard. A simple header would be added instead of the IEEE 802.11 format. He can be reached at (617) 246-0103 or Johncafarella@worldnet.att.net.

The meeting adjourned about 11:00 AM.

Thursday, November 14, Administrative
The IEEE 802.3 closing plenary convened at 8:30. Geoff Thompson presided over the meeting. Andy Luque, IEEE 802.3 secretary, recorded the minutes. The Thursday Agenda (Attachment A1) was presented. No changes were requested.

The potential voters list was presented (Attachment A2). Kelly McClellan was added as a voter. The at-risk voters list was presented. None were removed. The attendance sign-up list was circulated.

The Enschede Minutes were considered.

MOTION: Approve the IEEE 802.3 Enschede minutes.

M:Andy Luque S:Alan Flatman; Approved by acclamation

Two corrections were requested. In pages 1, 5, and 8, the meeting dates need to be in July instead of March. In page 2, Alan Flatman should appear as the SC25/WG3 liaison. It should also be noted that v. Pattay interrupted Alan's report and gave his version of Alan's report. The Minutes were approved by consensus.

There will be a call for patents after the morning break.

IEEE 802.3 Operating Rules - David Law
David reported on the status of the IEEE 802.3 Operating Rules (Attachment A8). There was a meeting Wednesday. Draft 3.1 will be produced based on the comments submitted. Volunteers were solicited to review the updated draft prior to a general mailing. Rich Seifert, Andy Luque, Alan Flatman, Bill Quackenbush, and Douglas Thomson volunteered. Draft 3.2 will be produced before the March plenary. A vote is expected Thursday at the next plenary.

Interpretation Request - David Law
David reported on the LeCroy interpretation request (Attachment A9). They had questions about the 10BASE-T voltage templates. The request was broken into five items.

TECHNICAL MOTION: IEEE 802.3 approves the LeCroy responses developed by the ad-hoc as appropriate and correct and without the need of a 30 day letter ballot.

M:Rich Seifert S:Dave Fifield; Y:45, N:0, A:3; Approved

ISO JTC1/SC6 Liaison - Gary Robinson

Gary reported on the status of the IEEE 802.3 projects at ISO JTC1/SC6 (Attachment A10). SC6 met last month. 100BASE-T and MB#4 ballot responses were approved at Guernsey.

Publication is being converted to PDF from Framemaker.

The Full Duplex and 100BASE-T2 standards are moving ahead.

The SC6 TAG is redoing its administration. It is moving to the IEEE. IEEE 802 will now have a formal relationship to SC6.

JTC1 is going through re-engineering. How can it improve? This review has been going on for a year. It will probably continue for another year. Make it more efficient.

MOTION: Move to recommend that the US vote to support the three New Projects for IEEE 802.3r, IEEE 802.3x, and IEEE 802.3y and the drafts of each.

M:Gary Robinson S:Bill Quackenbush Y:38 N:0 A:3; Approved

Administrative - Geoff Thompson
Several administrative issues were considered (Attachment A11).

MOTION: IEEE 802.3 directs its chair to not support the acquisition of video projectors. IEEE 802 should not get into the AV equipment business. Groups should furnish their own tools as they see fit. The technology is immature and not suitable for large groups, as demonstrated at the opening plenary.

M:Andy Luque S:Rich Seifert Y:38 N:5 A:7; Approved

MOTION: IEEE 802.3 directs its chair to not support the purchase of new laptops for the IEEE 802 Executive. Laptops are common tools furnished by most companies. Simple 1 page overheads and executive minutes do not justify new laptops. IEEE 802 should not be in the equipment business. We support having printers at meetings.

M:Andy Luque S:Rich Seifert Y:47 N:2 A:4; Approved

A friendly amendment was accepted to delete reference to "Simple 1 page overhead..." The motion was approved, as amended.

Maintenance 100BASE-T - Colin Mic
Colin reported on the status of Maintenance (Attachments F). The group met Tuesday from 17:00-18:00. A total of 15 requests have been received. 6 were closed with the publication of ISO/IEEE 8802-3:1996. 3 are ready for ballot. 6 were returned for additional information. The Maintenance PAR is ready. The group will meet at the January interim to prepare a ballot document. Information will be available before the next plenary. Some of the changes would be technical. No graphs are involved. It is all text.

Someone noted that these revisions open up the entire book for comment.

TECHNICAL MOTION: Approve PAR for 100BASE-T Maintenance revision.

M:Colin Mic S:Gadi Lahat Y:36, N:0, A:0; Approved

The proposed scope was modified. The scope is limited to updating IEEE 802.3u to identify and correct possible inaccuracies.

Administrative - Geoff Thompson
A short break was taken between 9:45-10:00. There was a call for patents. According to IEEE Rules, patents need to be made available under reasonable and non-discriminatory terms. A letter to that effect is needed. In addition, Geoff requests a letter on patents that are under application. These letters can go to Geoff or to the IEEE. The letters do not need to include specific details on the patent. A standard can be withdrawn, because of patent issues.

Full Duplex - Rich Seifert
Rich reported on the status of the Full Duplex task force (Attachment G). There were no formal meetings since Tuesday. One disapprove voter has agreed to change his ballot. The second disapprove voter needs to be contacted. Do not expect any technical changes.

The third disapprove voter requested an increase in the MAC frame size from 1518 to 1522 bytes. A Note, warning about a possible increase in the frame size, would be added to IEEE 802.3x to satisfy this voter. This issue was discussed extensively at the IEEE 802.1 technical plenary. IEEE 802.1 is unclear as to what they want to do. Representatives of IEEE 802.1 were invited to come to IEEE 802.3. No one in IEEE 802.1 had any time to do this. No formal request has been received from them.

Alan Chambers, IEEE 802.1 favors the 4 extra octets in the MAC frame. It might be prudent to allow for even more octets. Make 8 or 16 octets the upper limit. Applications are difficult to change. There are legacy problems.

Geoff Thompson, this seems to be a serious problem. There has been a failure to communicate. The depth and breath of the problem is not clear. There is a large legacy problem. Gigabit excludes frame growth. This new standard creates another legacy problem. There is a window of opportunity. This situation provides for discussion.

The minimum frame size, of 64 bytes, may also increase. No technical request has been made.

Andy Luque, adding Geoff's Note to IEEE 802.3x is not an appropriate solution to the frame size issue. There are better ways of publicizing this problem.

TECHNICAL MOTION: IEEE 802.3 does not support the modification of its frame size (maximum or minimum).

M:Andy Luque S:Douglas Thomson; Tabled

Geoff Thompson, a review of the standard is appropriate. Wider review than in the WG.

Mic Seaman, disagrees with the liaison statement. What would be an appropriate way to deal with the problem? We have a standard in progress that may impact the MTU or the maximum frame size. Need feedback on what is the impact to the standard if the maximum frame size is exceeded. Hold back until the related draft is completed.

The 4 extra bytes were not incorporated into the document that has passed the first WG ballot. There are two options. One option would increase the MAC frame size. The other would reduce the frame size available to the applications. The intent in IEEE 802.1 is to make you aware of the changes that are likely to be requested in the future. Seek information before making the decision. The Note would be a good part of the communication mechanism.

Howie Johnson, against the motion. The Note addresses the communications problem. It would reach a wider audience. Use these tools to solve the communications problem. Comment in the Gigabit Ethernet draft.

Geoff Thompson, Bridged and Switched VLANs are quite popular.

PROCEDURAL MOTION TO TABLE: Table the Luque/Thomson motion.

M:John Payne S:Gadi Lahat Y:30, N:17, A:8; Approved

Geoff Thompson, the commentor is willing to accept the text in the Note. This is not a technical change. Need to publicize this issue.

Andy Luque, disagrees. This is not a good way to achieve this objective. There are better ways to publicize this issue.

TECHNICAL MOTION: That IEEE 802.3 approves the resolution of Sponsor ballot comments as determined by IEEE 802.3x

M:David Law S:Bill Bunch Y:35, N:7, A:10; Approved

Andy Luque, offered a friendly amendment. "Except that the resolution of Geoff's comment would be in the recirculation document, but not in the published standard". David and Bill did not accept it as friendly. It was offered as an unfriendly amendment, but was not seconded.

Shimon Muller, offered a friendly amendment. The Note in IEEE 802.3x would say "a few". It was not accepted. He then proposed "several bytes". David and Bill accepted this suggestion as friendly.

The question was called. There were no objections. The Law/Bunch motion, with the Muller modifications, was approved.

TECHNICAL MOTION: IEEE 802.3 requests that the IEEE 802 Executive Committee approve the forwarding of IEEE 802.3x draft 3.1 for LMSC Sponsor recirculation ballot.

M:Rich Seifert S:Bill Quackenbush Y:45, N:0, A:9; Approved

Comments in this ballot are limited to the issues raised during the initial LMSC ballot. Recirculation was one of Geoff's conditions for withdrawing his negative ballot.

TECHNICAL MOTION: IEEE 802.3 request the IEEE 802 Executive Committee approve the placement of IEEE 802.3x, after successful recirculation, on the Revcom agenda and provide updated drafts to the ISO process.

M:Rich Seifert S:Bill Quackenbush Y:42, N:0, A:4; Approved

Xerox has officially transferred the assignment of the Type field numbers to the IEEE. A procedure needs to be written to govern their assignment.

Enhanced MAC - Mart Molle

Mart reported on the status of the Enhanced MAC task force (Attachment H). No plans to make any motions at Vancouver. A report will be provided on the experimental results and the updated draft. The Gigabit Ethernet effort consumed most of the time this week. Compatibility with Packet Bursting needs to be examined. Limited attendance at the Tuesday BLAM meeting.

Mart reported on some of the testing with the 10 BLAM adapters. Several foils presented some of the experimental results. A variety of scenarios were considered.

All the testing has been on 10BASE-T networks. So far, no 100BASE-T results. The 100BASE-T software drivers are not stable.

Will a larger LAN been examined? Yes, would like to explore a larger setup, but more adapters are needed.

Where multiple repeaters used? Yes, at 10 Mb/s there were many repeaters.

The experimental results verified that latency improved.

100BASE-T2 - Scott Carter
Scott reported on the status of the 100BASE-T2 task force (Attachment J). There were 115 comments. 21 were technical. All the comments have been resolved. Some of the comments also concern the Full Duplex specification. Interrelated comments have been addressed. No technical changes. A number of editorial changes. Plan to recirculate the pages that were modified. Howard Frazier's PICS statement was accepted.

TECHNICAL MOTION: IEEE 802.3 authorizes that the changed pages of the IEEE 802.3y 100BASE-T2 draft 7, that will be modified, as documented to resolve the comments from the LMSC Sponsor ballot, be submitted to the IEEE 802 Executive Committee for LMSC sponsor ballot recirculation.

Furthermore, conditional upon a successful recirculation requiring no technical changes, IEEE 802.3 requests that the IEEE 802 Executive Committee approve the placement of IEEE 802.3y on the Revcom agenda, and provide updated drafts to the ISO process.

M:Scott Carter S:Rich Seifert Y:46, N:0, A:1; Approved

Gigabit Ethernet - Howard Frazier
Howard reported on the status of the Gigabit Ethernet (Attachment K). Over 200 attended the 45 technical presentations at Vancouver. 5 task forces have been created. Consensus was reached on the content of the first draft. A total 16 motions were considered. Several optical PMD proposals. A separate task force will deal with Long Haul Copper.

The first draft will be distributed in January. It will include all of the proposal adopted so far. An interim meeting is planned for January 27-29 in the San Diego area. A PAR and 5 Criteria will be developed for Long Haul Copper. Parallel tracks are expected. The draft will be distributed a week before the interim.

TECHNICAL MOTION: IEEE 802.3 authorizes IEEE 802.3z to produce a PAR with supporting documentation, to facilitate the establishment of a separate task force to conduct the standards work on the 1000BASE-T (Long Haul Copper) PHY and that IEEE 802.3z task force forward this PAR and supporting documentation to the IEEE 802 Executive Committee, at least 30 days in advance of the March 1997 IEEE 802 meeting, for consideration at the March 1997 IEEE Standards Board meeting.

M:Howard Frazier S:George Eisler Y:56, N:1. A:0; Approved

The new PAR will be reviewed by IEEE 802.3 at the March plenary.

George Eisler, 1000BASE-T is planning an interim meeting for December 17-18 in New Jersey. No hotel site has been selected. Special dispensation was granted for this meeting, with regard to the 30 day announcement rule.

Connector specifications will not be included in the first Gigabit Ethernet draft. An IEC standard is underway in this area. The group will monitor Fiber Channel progress. The group does not plan to discuss the connector at the January interim. Proposals will be entertained at the March plenary.

The Full Duplex and 100BASE-T2 task forces do not plan an interim unless new issues are raised by the confirmation ballots. 100BASE-T Maintenance plans to meet concurrently at the Gigabit Ethernet interim.

The IEEE 802.3 plenary meeting adjourned at 12.18.

Respectfully submitted February 15, 1997.

Andy Luque
IEEE 802.3 Secretary
Open Communications
131 McCartney Dr.
Bend, Oregon, 97701
(541) 389-6512
andy_luque@hotmail.com