Re: [yam] deployment of 8BITMIME?

[John C Klensin <[e-mail]>]

--On Sunday, August 16, 2009 17:50 -0400 Tony Hansen
<[e-mail]> wrote:

> Some of us have been having a side conversation about
> interoperability reports and deployment reports.
> 
> There's a point that John and Dave have been making that I
> think needs to be re-stated again:
> 
> When you're looking at the PS => DS stage, you're concerned
> with interoperability, that is, "how well do different
> implementations of the protocol get along with each other?".
> 
> But in the DS => FS stage, you're concerned with deployment,
> that is, "how far reaching has this protocol become?"
> 
> YAM really needs to be looking at the deployment of the
> standards.
>...

Tony,

One more thought on that subject that occurred to me only while
reading your note.

I think it would be helpful to both us and the IESG --assuming
it is true-- to be able to state explicitly that the deployment
of the specification did not turn up any new interoperability
issues.  Without any issues of "interoperability testing",
deployment experience might indicate interoperability or clarity
problems that were not detected by the PS->DS testing.  We would
certainly be obligated to report such problems if they were
detected (and presumably move to remove the spec from the YAM
agenda so it can be improved, recycled in grade, or even bounced
back to Proposed). It is probably desirable to affirmatively
state that no previously-undetected problems have been found
with implementations that do not interoperate smoothly but that
are both consistent with the spec.

     john