Thread Links | Date Links | ||||
---|---|---|---|---|---|
Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Howard,
I think everyone in the Speed ad hoc agrees with you completely that the study group/task force (when approved) should really be focused on one and only one speed.
In the short term, however, the ad hoc was formed with the charter to come up with a motion for a speed objective which had a reasonable chance of garnering the necessary 75% support. Based on reflector discussion and the straw polls on the conference calls, neither of the motions you suggest have that support. I do think we made decent progress in that we've narrowed it down to two numbers as opposed to the plethora of speeds that were on the table.
Having said that maybe it is a good idea to get one more straw poll done, by email this time.
Walter Thirion
Level One Communications
> -----Original Message-----
> From: Howard Frazier [mailto:hfrazier@xxxxxxxxx]
> Sent: Friday, July 02, 1999 2:37 AM
> To: stds-802-3-hssg-speed@xxxxxxxx; wthirion@xxxxxxxxxx
> Subject: Re: Speed motion
>
>
>
> Walter,
>
> I think we are supposed to be writing objectives for the
> project (i.e.,
> for the P802.3ae 10Gig Task Force, assuming one is formed), not for
> the Higher Speed Study Group, which will hopefully finish
> it's work some
> time before the Y2K bug renders us all unable to generate
> slideware and
> exchange email. Maybe we should have an objective which says:
>
> "The email thread from the stds-802-3-hssg-speed reflector shall
> be preserved on acid free paper, so that future generations can
> carry on the work."
>
> Okay. That's (more than) enough sarcasm for one night.
>
> We should opt for the "decide later what speed(s) we mean when we say
> '10Gig'" only as a last resort, and only in the form of a short term
> objective for the P802.3ae 10Gig Task Force. So, your motion could
> read something like:
>
> Decide between 10.000 Gb/s and 9.58464 Gb/s, or adopt both speeds,
> once a task force has been formed.
>
> We had a similar objective in the gigabit project. It read:
>
> 12. Decide between collision domain diameter of >= 50m or >= 200m
>
> at the time we finished the HSSG, up until the first meeting of
> the P802.3z Gigabit Task Force. At the July, 1996 meeting (in the
> lovely college town of Enschede, NL) we changed this to:
>
> 12. Support maximum collision domain diameter of 200m.
>
> by a unanimous vote. We took this course of action because
> there was no clear
> consensus prior to the Enschede meeting.
>
> So, we can defer the decision until later...But I Don't Think
> We Should.
>
> The speed is important. It goes beyond "marketing glossies", as
> product literature has been (unkindly) referred to. If we defer the
> decision, I believe that the press is going to lambast us. 802.3 has
> built up a solid reputation with the press as one of the few truly
> effective standards committees. Defering the decision could seriously
> errode our credibility, making us look mindless. Once you loose
> credibility, it is VERY hard to get it back. I think we need to
> find a way to get beyond the impasse we seem to have reached.
>
> Maybe we should put up two motions in Montreal. One which says:
>
> Adopt as an objective:
>
> Support speed of 10.000 Gb/s at the MAC/PLS service interface
>
> and another which says
>
> Adopt as an objective:
>
> Support speed of 9.58464 Gb/s at the MAC/PLS service interface
>
> If neither pass, our children can carry on the work....
>
> If both pass, we support both speeds, and decide just how to do that
> once we form a task force.
>
> If only one passes, then we know what speed we will be running at.
>
> Howard
>