On 11/11/2010 12:25 PM, Dave CROCKER wrote:
To establish the base: It is not possible to achieve widespread use on the
Internet without having multiple components interacting. That's called
However, the interoperability might be among components that are clones of a
single code base.
So our language needs to be enhanced to cover multiple implementations. And as
long as the language hood is up, we might as well put in a turbo-booster that
asserts the higher octane 'interoperability' word.
A hallway conversation with Russ added an item that simply had not occurred to
There might be multiple implementations that rely on on undocumented
modifications of the spec. This means that an additional, interoperable
implementation cannot be made purely from the specification.
Again, I believe the requirement for the document is "merely" to get the wording
right. I do not believe any of us differ on the actual meaning we are trying to
achieve. That is, I have not seen anything that indicates we have disparity
about the intended requirement.
Test language: (*)
(Full) Internet Standard:
The Internet community achieves rough consensus -- on using
the multiple, independent implementations of a specification
3.3. [Full] Internet Standard (IS)
This is the existing final standards status, based on attainment of
significant community acceptance, as demonstrated by use of multiple,
independent implementations that conform to the specification.
ps. I just realized that the original language that Russ cited said "on using
the running code of a specification". "Of a specification" explicitly means
that the stuff that is running is the spec and, therefore, can't really mean
that it's using hallway agreements. (However I think it's dandy to make the
Section 3.3 language bullet-proofed against creative misunderstanding.)
(*) This is just from me; it hasn't been vetted with my co-authors.
Ietf mailing list