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

RE: Dan- Again (response)

Hi Roger, 

I agree that we don't want to use the reflector to waste people's
time, so we should use it to drive decisions as to what technical
data would be relevant and useful for making decisions at the 
upcoming meeting. As we all know, there are tons of technical things
that can be done, but not all of them will be useful.
The reason I have to referred to things like "objectives" and
"decisions" is that this is a group making decisions to limit the
possible technical things we must consider. Measuring and analyzing
data that is not within the agreed upon scope of study will just
result in wasting of everyone's time.

This is not political maneuvering, it is practical standards development. 

>  I am gonna get busy here and try to deliver some input
>  (measurements at the meeting).  No intent to waste anyone's 
>  time here.
>  At this point I voiced my opinions...
>  and I am a bit surprised to see this "decided",  
>  "objectives",  and you shipped
>  a head song and dance come up over and over.  These are the 
>  words that  normally
>  sound  "Non Technical"  for a lack of a better word
>   to me, I delivered 10 reasons to use the signal pair, and 
>  so far the 
>  only arguments that I can see against that are:
>  0- The magnetic won't work (we can prove that it does)

I have not claimed that the magnetics won't work. I have stated 
that inserting them into the mid-span on the data pairs will 
impact Return Loss, Insertion Loss, and NEXT. How much we can 
tolerate is subject to specifications and measurement. I hope 
that you intend to bring in something that alleviates my concern
in this area. I am confident that using the 45,78 pairs would
substantially reduce the risk of impacting 10/100T because it
does not require insertion of impairments into the channel.

This is scientifically based, not politically.

And I have difficulty understanding how a differential detection
mechanism can be easily used to ascertain DTE-Power-Ready from the
center of the link. I consider myself reasonably knowledgeable, yet
it seems like a much more complex task than a common-mode DC
method would be. Perhaps you have an innovative method that will
surprise and impress me? I am willing to be impressed.

>  1- you shipped, proprietary stuff,  so it is not good, mine 
>  because it is in the
>      lab still, is better.  I don't want to hear that you 
>  passed all these other things (EMI... SPECS)
>      irrelevant, proprietary stuff, bad for all of us.  We 
>  were never out to force a solution on
>      the group, and for sure we can improve our scheme change 
>  it, and deliver an industry 
>      solution to make life easier for all of us.  FYI, I 
>  don't feel guilty for delivering a product
>      nor do I owe  anyone an apology.  And please remind me 
>  of a single statement made verbal
>      or otherwise on my part that shows disrespect for 
>  anyone's solutions or saying that 
>      Cisco's way is the way,  I still believe that what we 
>  have is valuable and given open
>      mindedness on the part of the group can be put to good use.

I have not stated anything negative about Cisco's technology because
I have not seen it. It may be the best thing since sliced bread and
I am willing to consider it, but if it uses differential detection
on the data pairs, it will have an uphill road to travel because 
those choices would defy my opinion of what the right solution should

As for "our scheme", the best solution I can think of would be to
let someone else build DTE-Power-Insertion panels and leave the
more interesting packet switching functions to me. I just don't want
the poor customer to be left holding the bag by seeing a standard
that does not meet their best interest.
>  2- The committee decided on x y z objectives.  and for sure 
>  that excludes what you believe in.
>      I enjoyed interacting with everyone here and got to know 
>  you and Larry a little better,
>      as you can tell I am new to the game here, but grateful 
>  to be a techie ....
>  see u there,
>  roger

I strongly suggest that you review the minutes and presentations
that have already been posted on the IEEE web site. Walking into a
large group and telling them that their decisions are not important
is not likely to be very persuasive or create a receptive tone for
your presentations.

I hope you take this as it is intended, in sincere hope that we can
work together to develop a useful standard.

Best Regards,

Dan Dove
HP ProCurve Networks