Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [802SEC] Comments on Disbanding 802.4

Title:
Paul,
 
I very much appreciate the commentary.  I will share this with others on the EC reflector as I find it interesting from a historical and philosophical view point.  I did not participate in the EC back at the time where 802.4 was placed in Hibernation.  Clearly there was a lot of politics involved, that I don't care to comment on since I wasn't there.  However, you bring up a lot of philosophical issues regarding the definition of a WG.
 
Many people seem to take the view of "One MAC" per working group.  Some groups clearly enhance their MAC to the point where it could be argued that it constitutes multiple MACs, or chose to have multiple MACs out right I believe.  I fancy myself a purist.  I think it is all a question of scope.  I would argue that each time a WG goes to the EC and SA with a PAR, they modify their scope.  If that scope modification is allowed (and the WG stays within it's scope) then everything is legal.  I don't think the question of multiple MAC's etc should come into play. 
 
I do think the concept of a WG requires clarification.  Today, WG come into existence based on a single PAR.  People (or at least myself at times) often think of that first PAR as somehow defining the scope of the WG.  But clearly that is not really the case (at least I know of no rules that require it).  A sponsor on the other hand must have a defined scope.  If a PAR does not fit within that scope, it should be taken else where (though I suppose a Sponsor could chose to change it's scope with approval of appropriate authorities).  In the 'old days' my impression is that every new PAR tended to be a new WG, so there was a close alignment between the scope of a PAR, and the scope of the WG.  I don't see that to be the case any more.  For IEEE 802, some of our WG do reach such a large size and scope that they in effect act as mini-sponsors.  A new PAR could be some narrow subset of existing scope (based on the union of all previous PAR scopes for that WG), or could perhaps be adding new scope to the WG. 
 
When the time comes to develop a set of P&P specifically for setting up WG I think one of the requirements I will suggest is that the WG itself define a scope independent of but wholly containing whatever PAR initiates it, and that this scope be separately approved by the EC.  This may in part address the concerns you express, though I don't think it goes all the way.  Right now, we do not approve WG scope (or even their P&P though I sometimes review them).  Anyway, I think there might be some value to requiring WG to have an approved scope that is independent of the PARs they write.  I'm certainly open to other thoughts though.
 
Unfortunately concerning disbanding of 802.4 it is my opinion that they HAVE completed their work.  Once all standards are withdrawn, there is no longer a formal need to maintain a core of experts.  Without that need there is no need to continue the WG.  I agree that informal use of a standard continues, but pretty much by definition, IEEE is no longer committed to supporting the WG's standards since they chose to withdraw them.    Once all its standards are withdrawn the WG has complete its work.  At least that is my opinion.  I offer my apologies for voting against you, but my appreciation and thanks for sharing your thoughts which I very much enjoyed.
 
Hope to see you in person some day down the road.
 
Best Regards,
 
Mat
 

Matthew Sherman, PhD
Senior Member Technical Staff
BAE SYSTEMS, CNIR
Office: +1 973.633.6344
email:  matthew.sherman@baesystems.com

-----Original Message-----
From: Paul Eastman [mailto:paul@rfnetworks.com]
Sent: Friday, August 06, 2004 11:23 AM
To: Sherman, Matthew J. (US SSA)
Subject: Re: Comments on Disbanding 802.4

The thing you seem to be missing is that all of 802.4 was thrown out by the EC a few years back.  The arguments I am making now are the same arguments I made when the first motions were made.  Since 802.4 led the way (in developing the concept of hibernation, core of expertise, and other pseudo-sunset concepts) I felt it necessary to fight, not only for 802.4, but for other groups that follow.  I think that it is an important concept that standards complete their work.  I find it particularly abhorrent that 802.3 has tried (successfully) to perpetuate itself to the point where it is not a single standard but rather several individual and separate standards all tied together under a "marketing" umbrella of 802.3.  At one point it was announced in an 802.3 meeting that they had finally developed a CSMA/CD standard that didn't use CS, MA or CD.  When the big fight between 802.12 and the high speed group of 802.3 took place, I argued for an 802.13.  The 802.3 people made it clear that their technical solution was not strong enough to stand on its own and needed the marketing cover of 802.3 to survive.  Thus we ended up with a technically inferior standard that dominated through non-technical considerations.  

I don't particularly care one way or another which way the vote on this issue comes out.  However, I would like to caution you to consider the ramifications of your actions on long term effects.

Please feel free to pass any of my comments on to the rest of the EC.  My attempts at passing my comments through the reflector seem to have been blocked by the administrator.

-- Paul Eastman, Ph.D.
    Former 802.4 Working Group Chair

Sherman, Matthew J. (US SSA) wrote:
Paul,

It sounds like everything is handled informally since no standard actually exists right now.  What benefit does a hibernating WG provide to the current situation?  It's primary function in my mind it to provide a core of experts to field questions on a standard.  If it does not perform that function, then what is the point of keeping it around?  I'm sympathetic to that fact that there still seems to be a user base, but the user base is not using the WG facility.  So it does not seem to fill a purpose.  Am I missing something?

Thanks,

Mat

Matthew Sherman, PhD
Senior Member Technical Staff
BAE SYSTEMS, CNIR
Office: +1 973.633.6344
email:  matthew.sherman@baesystems.com



-----Original Message-----
From: Paul Eastman [mailto:paul@rfnetworks.com]
Sent: Thursday, August 05, 2004 1:37 PM
To: Sherman, Matthew J. (US SSA)
Cc: STDS-802-SEC@listserv.ieee.org; paul.nikolich@ATT.NET
Subject: Re: Questions on Disanding 802.4




Sherman, Matthew J. (US SSA) wrote:

  
Dear Paul,

I'm trying to determine how to vote on this issue.  I was hoping you can help me with some questions.

1)  Can you provide a list of the "core of experts" that this group maintains to resolve technical issues regarding 802.4 standards?

    
Since our standard was withdrawn in 2001 and our representation in the 
EC was terminated, I did not deem it necessary to continue active 
polling of the "core of expertise".  Should it be necessary, I could 
probably resurrect 50% or more of the "core of expertise" on very short 
notice.

  
2)  When was the last time a formal inquiry or request for interpretation was made of this group?

    
The 802.4 standard was very well written.  The last action of the 802.4 
group was to suggest that its 802.4L study group reform as 802.11. 
 There were no problems, resolved or otherwise, with the standard 
published jointly as ISO/IEC 8802-4 (ANSI/IEEE Std. 802.4) Token Passing 
Bus 1990

  
3)  How often do you receive informal inquiries concerning 802.4?

    
I personally have continued to receive a couple of inquiries a year, 
mostly from academic sources.  In the latest inquiry we even 
investigated and gave a rough order of magnitude quote on the 
resurrection of both MAC and PHY layers for a Department of Homeland 
Security application.

  
4)  Off hand, can you identify any specific users of the technology?  How big (in terms of number of nodes) do you estimate the total current deployment is? 

    
There is still a number of places where the carrier band version of 
802.4 is being used.  Most of the equipment is being shipped into Japan 
through Marubun, an importing company, and to various companies in 
India.  There is even some product being sold to Moore Products, a 
Seimens company located in Pennsylvania.  Specifically, Hitachi, 
Furukawa Electric, Yokagawa and possibly Toyota are Japanese companies 
with active systems

  
5)  What is the likelihood of any new nodes being added?

    
The answers to item 4 indicate where additional nodes are still being added.

  
6)  Is anyone actually building equipment, as opposed to just using equipment that is still around?

    
Relcom, Maris Graube's company (Maris was the first chair of the 802 
EC), is still manufacturing many nodes per year for the carrierband PHY. 
 My company, RF Networks, is still capable but not manufacturing nodes 
for the broadband PHY.

  
Thanks in advance for any answers you can provide.  Finally, in the quote Paul Nikolich provided, what did you mean by "old standards should have the right to protect their legacy"?  I'm not sure understand what you mean by this.

    
A lot of work was done to produce an extremely stable standard, both MAC 
and PHY.  Should some future work decide to do some work using a token 
bus architecture, I would hope that wheels would not be reinvented and 
that the pioneers of the work would get some acknowledgment.

  
Best Regards,

Mat





Matthew Sherman, PhD
Senior Member Technical Staff
BAE SYSTEMS, CNIR
Office: +1 973.633.6344
email:  matthew.sherman@baesystems.com




 

    

  

-- 
==================================================

Paul Eastman
RF Networks, Inc.
10201 N. 21st Avenue, Unit 9
Phoenix, AZ  85021
(602) 861-3652
Fax:  (602) 861-0251

  "Worrying about what's right is always more
    important than worrying about who's right."

==================================================

---------- This email is sent from the 802 Executive Committee email reflector. This list is maintained by Listserv.