[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: stds-802-16: Re: Agenda




Demos:

I agree that there was a lack of communication between the TG1 and TG3 group
that lead to some misunderstandings.  However, to be completely honest, I
would have to say that the major factor that lead to the misunderstandings
were that not enough people in the TG3 group actually read through the TG1
MAC to comment on it properly and not enough people in the TG1 MAC group,
myself included, read through the TG3 Functional Requirements.  Although
Roger does set the main schedule for the 802.16 Work Group, it is based on
input from the various Task Group officials and from members with concerns.
I'm sure that if someone could recommend a schedule that will allow all of
the task groups to accomplish their required goals for this session and
still allow for joint sessions to occur, Roger would not be against it.
Personally, based on the amount of work outlined by Rogers email yesterday,
I'm at a loss in trying to create such a schedule.  Although, I too would
like to see more communication within the groups I can't see how it can be
done this session and I don't see sarcasm helping out either.  One possible
way to increase the communication would be via this email reflector.
Another one may be a joint interim meeting between the TG1 and TG3.  If
there are enough people that believe that this would be required, then maybe
it should be brought up during one of the plenary sessions next week.

my $0.02 (US$0.0131)

Phil

-----Original Message-----
From: Kostas, Demos [mailto:dkostas@adaptivebroadband.com]
Sent: Friday, November 03, 2000 12:29 AM
To: 'trinkwon@compuserve.com'; 'IEEE - Marks, Roger'
Cc: stds-802-16@ieee.org
Subject: RE: stds-802-16: Re: Agenda



One item I remember is that when it was discovered that the poor
communication between TG1 and TG3 was primarily responsible for the TG3's
misunderstanding on what the TG1 MAC does or does not do, our chairman
accepted the responsibility for this situation and apologized.  I then hoped
that this realization was going to result in some changes in our Task
Groups' interactions. I guess that was too much to hope for. It seems that
we back a business as usual standardization process. 
Too bad! 
Demos  

Dr. Demosthenes J. Kostas
Director, Industry Standards
Adaptive Broadband Corporation

3314 Dartmouth Ave
Dallas, TX 75205  USA

tel: 214 520 8411
fax: 214 520 9802


-----Original Message-----
From: David Trinkwon [mailto:trinkwon@compuserve.com]
Sent: Thursday, November 02, 2000 3:04 AM
To: 'IEEE - Marks, Roger'
Cc: stds-802-16@ieee.org
Subject: RE: stds-802-16: Re: Agenda



I appreciate the nature of your problem. It's not so much at the High Level 
agenda, its more to do with how the individual TG Chairs (with yourself) 
designate and coordinate the time within the parallel sessions so that we 
members can make intelligent plans for hopping to and from the different TGs

and sub-groups. From a TG3 perspective, we need to see a specific time when
TG1 
/ TG3 MAC issues are being worked which does not conflict with the TG3 PHY
and 
Evaluation Criteria work. TG1 and WHUMAN will have their own overlapping 
priorities / concerns. This jig-saw needs to be worked out between the TG/
SG 
Chairs, and presented for approval at the Opening Plenary. It is NOT 
practicable for members to sit through every parallel session on every day
not 
knowing when the specific sub-topics will come up, and then discovering a 
conflict.

Other standards bodies manage to publish more detailed advance WG/TG agenda
as 
to which topics / contributions will be discussed on which morning /
afternoon 
/ evening etc, enabling conflicts to be avoided or identified beforehand and

then resolved or worked around. 802.16 seems to prefer building walls
between 
TGs/SGs and then complain afterwards that the right people weren't at the 
appropriate sessions. This also causes difficulties when agreements have to
be 
undone or reworked because the best/complete input was unavailable
originally. 
Please help us to plan our individual times more productively for the
overall 
benefit of all TGs/SGs. These other standards bodies also recognize that 
"unplanned" time slots should be deliberately left ( or built around 
break-times etc) to enable various ad-hocs, drafting groups or "serendipity"

discussions to take place for the benefit of the overall process, or to
allow 
individual members to "browse" through other TGs on an opportunistic basis, 
improving mutual awareness and cross-fertilization of information, ideas
etc.

AS I see it, 802.16 has moved rapidly from a single-issue Work Group (10-66
GHz 
with sub-groups formed for MAC, PHY and Coexistence) to a Multi-issue WG 
(10-66GHz, 2-11GHz and WHUMAN) without an associated improvement to its
meeting 
logistics / procedures.

I notice that the high level agenda contains a "common" MAC group. Please
note 
that at present we do NOT have a "common MAC" and have not yet discussed a 
"common MAC" development plan or "common MAC" TG structure / chairmanship
etc. 
We are in a transition mode from a TG1 MAC to a common MAC which also
embraces 
the needs of TG3 and WHUMAN (and any future air interfaces). In accordance
with 
Motions agreed in Denver, TG1 has agreed to provide a generic baseline at
Rev4 
+ the outstanding convergence layers, but so far that hasn't quite happened,

and we (the WG) haven't yet decided how the overall work should be managed 
beyond Rev 4. The assurances given in Denver that Rev 4 would be PHY 
independent etc are not fully correct, and the more recent contributions
within 
TG3 have identified additional aspects for discussion / resolution. As a 
consequence of the Denver Motions, the 802.16 MAC no longer "belongs" to TG1

only, or to the existing TG1 MAC sub-group.

David Trinkwon
e-mail : trinkwon@compuserve.com
Telephone :  UK  (+44) (0) 7802 538315   USA   (+1) 972 345 5226
Fax :                UK (+44) (0) 20 7681 1695   USA   (+1) 602 532 7013

----------
From: 	Roger B. Marks[SMTP:marks@boulder.nist.gov]
Sent: 	Thursday, November 02, 2000 05:41
To: 	trinkwon@compuserve.com
Cc: 	stds-802-16@ieee.org
Subject: 	stds-802-16: Re: Agenda



>It seems that we will have more than usual conflicts on our time next week,
>with many of the same people involved simultaneously with TG3 MAC issues,
TG1
>MAC issues, TG3 PHY and Traffic proposals etc, WHUMAN PAR / 802.11 issues
and
>the unresolved ISTO stuff. I hope that you will be trying to do some
creative
>agenda building, rather than relying on the normal process of "like it or
lump
>it". I made some suggestions at the Denver meeting, lets see some new
>initiatives.
>
>David Trinkwon


David,

I understand that people who are interested in multiple projects
would like to see the Session agenda structured to allow them to
participate in more than one. On the other hand, it will be a very
busy week for all. Based on your request, I have taken another hard
look at the situation. The best I can do, I am afraid, is to
apologize for the fact that this problem is beyond my ability to
solve.

The current schedule
<http://ieee802.org/16/meetings/mtg10/agenda.html> gives each of the
Task Groups the following allocations:

Monday:     2 hours
Tuesday:    8 hours (assuming 2 hours for lunch and breaks)
Wednesday:  6 hours (assuming 2 hours for lunch and breaks)
Thursday:   5 hours (assuming 2 hours for lunch and breaks)
TOTAL:     21 hours

If necessary, groups can use Tuesday night, and possibly part of
Monday night (though an important WirelessHUMAN session starts at
8:30 pm Monday). The social takes up Wednesday night. {By the way,
the 802 organizers have run short of meeting rooms and have still not
100% committed to providing space for all of this.}

Now please consider some of the activities for the week. Without
specifying the details of each group's agenda, I can take a good
guess based on the contribution that have come in:

*TG1 has about 600 comments to resolve in order to prepare for
initiating a letter ballot. We will run the MAC and PHY groups in
parallel to maximize the time, but the MAC team has the bulk of the
comments (around 500, which comes down to about 150 seconds per
comment). The resolution of each comment, including votes when
necessary, has to be fully documented. In addition, 9 contributed
documents have been received; these will reduce the time available
for comment processing. In summary: TG1 will be busy.

*TG2 has many comments to process in order to prepare for initiating
its own letter ballot. It also expects to address some serious issues
arising from documents that do not recommend specific changes. This
may require the group to write quite a bit of new text. LIke TG1, TG2
will need to provide a fully documented change list.

*TG3 has received 24 contributions. Around 18 of these are PHY
proposals to be evaluated. If, for example, these were to be handled
on Wednesday, each would have 20 minutes, including questions and
setup time. The evaluation criteria and scoring system still need
work, channel and traffic modeling papers need to be discussed, and
four contributions were made regarding TG3 considerations for the
MAC. I've probably forgotten some other things too.

*WirelessHUMAN has a very full plate, as I have indicated in other
messages. This revolves around discussions with other Working Groups
and subsequent PAR revisions. The Study Group may also find itself
engaged in a number of discussions with voting members of the
Executive Committee. The 5 pm PAR deadline on Wednesday is critical.
On Thursday, the group needs to find time to prepare a Call For
Proposals, as described in the PAR.

Then there is the 802.16 Plenary activity. I've allowed two hours
each on Monday, Wednesday, and Thursday. If we can find a way to give
some back to the Task Groups, we will. However, we have a lot of
major issues to address.

I know of two ways to allow people to increase their multi-project
participation:

(1) Hold "joint sessions". The problem here is that each Task Group
has orthogonal requirements and a brutal task list. In my mind, any
joint session means that at least one group basically suspends its
work during that time. Speaking for TG1, we have very little spare
time for this. If other groups want to do it and think they can fit
it into their schedule, they may. However, I can't accept the role of
coordinating it; Task Groups will have to do it on a bilateral basis.

(2) Arrange the agendas so that people can duck out of one meeting
when they aren't interested and be sure that something interesting
will be happening in another room. Again, there is no way I can
predict who will be interested in what. Furthermore, much of each
group's agenda is sequenced based on external requirements or
internal logical ordering needs, so there isn't much flexibility.

The bottom line here is that I am not going to attempt to coordinate
the schedules of the Task Groups. If individual Task Groups want to
work something out with each other, they may, but I am personally
concerned that each group has too much on its plate and no time to
spare.

What we CAN do to help is to make the Task Group agendas available on
the server, keep them updated, and stick to them. I hope the Task
Groups will cooperate.

Regards,

Roger