Re: IETF interoperability testing

On 2007-12-18 12:49, Dave Crocker wrote:
> Ralph Droms wrote:
>  > Fred - to be clear, that DHCPv6 interop testing was not associated in
>  > any way with the dhc WG.  I'll let the organizers comment on any more
>  > general sponsorship arrangement or other association of the event with
>  > the IETF.
> 
> and
> 
> Randy Presuhn wrote:
>> I recall ones for SNMPv2 (various flavors), AgentX, and SNMPv3.  One 
>> might
>> quibble about whether they were organized by the working group, or 
>> whether
>> it was just a mighty coincidence that people from companies with 
>> employees
>> who attended these working groups hashed out mutual NDAs and held 
>> bake-offs
>> to verify that their implementations could interoperate, as well as 
>> experiments
>> over the open Internet.
> 
> 
> This is interesting.  People seem to be confused about formal 
> affiliation with IETF processes, versus independent activities that are 
> a useful adjunct to IETF work.
> 
> Let's be clear:  When an activity that relates to working group product 
> is not organized within the working group -- you know, overseen by the 
> chairs, minutes written, and other the other group process stuff of an 
> open IETF effort -- then it is not an IETF effort.

Correct. Not even if it takes place in the same hotel as the IETF
at the same time as the IETF, and not even if its results are later
used in an IETF interop report. But whoever is running the activity
had better say loud and clear "this is the not an IETF activity
and not subject to IETF rules".

     Brian
> 
> Apparently some folk thought I was making a slam at the IETF when I 
> noted we don't do interoperability tests.  It wasn't a slam.  The 
> lifecycle of a successful protocol involves quite a few phases, with 
> quite a few participants, at the micro and macro level.  The IETF is a 
> factor in only some of these.
> 
> But as anyone who has run a successful interoperability event for a 
> successful protocol knows, you don't just declare a time and place for a 
> party and see what happens.  That's the way the upcoming plenary was 
> presented to us.
> 
> At a minimum, it demonstrated that serious events for exploring utility 
> and problems are not commonly within the purview of IETF-specific 
> operations.
> 
> 
> d/

_______________________________________________
Ietf mailing list
Ietf@ietf.org
https://www1.ietf.org/mailman/listinfo/ietf