Re: [81attendees] casual attendees & WG process -- openness and continuity

[Steve Crocker <[e-mail]>]

I have been lurking with appreciation and occasional amusement.  After helping create the process and participating in multiple ways for over 40 years, I am truly empathic with all of the points of view.  Numbers do matter, and bigger groups are unwieldy.  Unprepared attendees are a drag.  I think these factors have been part of the reason for the emergence of design teams.  While all of these things matter, I want to focus and emphasize on the IETF's truly unique and outstanding heritage of openness.

First, the Internet architecture is open, and we've all seen enormous benefits.  Second, the IETF's processes are open, and we've all seen the power of free distribution of the documents and open attendance and introduction of new ideas.

In addition to these two key dimensions of openness, the fact that new people can enter right into the middle of the activity, both on the mailing lists and in the meetings, means that we are continually welcoming the next generation.  (And in this context, a "generation" might be measured in months or a couple of years, not decades.)  Many years ago there were some stats circulated that showed 37% were attending the IETF for the first time, and this statistic was essentially stable for several meetings.  That's a large number of new people.  Their average contribution may not be as high as the regular attendees -- though I have never seen any attempt to measure "contribution" -- but I have always thought this represented a very wise investment in our future.

Steve





On Aug 9, 2011, at 11:43 AM, Curtis Villamizar wrote:

> It is a different topic.  That is why I changed the subject line.
>  
> I don’t think this is an IETF governance issue.  In some WGs there needs to either be a filter (WG interest) or a lot more time allocated.  MPLS for example had mostly 10 minute slots and 4 hours (I think) of total meeting.  With 6 hours, at least the presentations would be 15 minutes on average.  Also if there was no interest at a prior meeting and not much has changed, then no time slot or a very short one or a Friday slot.  This would be up to the MPLS chairs.
>  
> The routing area could also make use of Friday.  We had some overlap problem trying to squeeze into Monday first thing to Thursday.  Only roll was on Friday and it seemed like that WG might be shutting down.
>  
> BTW – there didn’t seem to be a general comment session in the routing area open meeting, only a discussion of the paper on power optimization.  Otherwise this could have been brought up (again).
>  
> Curtis
>  
> From: [e-mail] [mailto:[e-mail]] On Behalf Of Dae Young KIM
> Sent: Tuesday, August 09, 2011 8:31 AM
> To: Curtis Villamizar
> Cc: Yoav Nir; [e-mail]
> Subject: Re: [81attendees] casual attendees & WG process
>  
> On Wed, Aug 10, 2011 at 12:11 AM, Curtis Villamizar <[e-mail]> wrote:
> Before asking any question the WG chairs should ask "how many people read or at least thoroughly skimmed through this document prior to the meeting such that they understand what it contains".  Then do the "favor/opposed" polling.
> 
> One of the problems with some of the WGs is too many junk drafts.  It seems like a number of authors just want to get their name on a WG draft or RFC so that they can cite it on their resume.  In at least one case an employer was rewarding employees for submitting a draft and having it presented and the result was a lot of junk drafts from that company.  This was counterproductive to the IETF WGs in which this was occurring.
> 
> The process used to be more streamlined because a small group of people would commit to deploy a simple protocol or extension on a significant provider network, a small set of people would agree to implement it, and the WG would provide a venue for open discussion.  In a number of WGs there are authors who persistently promote drafts with no provider or vendor interest or negative expressions of interest.
> 
> We have lost touch with the "running code" part of "rough consensus and running code" and we may not be doing well on the "rough consensus" part either.  We are headed toward the design by committee and voting methods used by other SDO along with including every half thought out feature anyone imagined during the process and no implementations.  This leads to protocols that don't work with features that no one has ever implemented or where no one has ever succeeded in getting to interoperate among vendors.
> 
> Curtis
> 
> You do have a point here, but this is far stretched off the topic of the thread on how to select venues and that with diversity.
> 
> I think you're raising a more fundamental issue of the governance/operational structure of IETF.
> 
> If your elite core group want to do work in a more cozy environment, you'd better redefine the membership of IETF and scrutinize the applicants to each meeting to filter out non-core members.
> 
> Is this what people want? Or is this what you ask the leadership to change the IETF to?
>  
> -- 
> DY
> _______________________________________________
> 81attendees mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/81attendees