Masataka Ohta wrote:
Hallam-Baker, Phillip wrote:
I don't see the value of running code quite as others do.
I agree. It was valuable in good old days, when implmenting a protocol
was purely voluntary with no budget.
Existence of multiple independent implementations, then, meant the
protocol was widely accepted by the society.
However, after the overly introduction of standardization engineering
to IETF, people satisfy requirements merely because they are required.
So, existence of required running code does not mean much.
It means the specification is implementable.
Since the goal of our work is to produce specifications
that will allow multiple independent implementations to
inter-operate successfully, I can think of no more valuable
review input towards this goal than comments from implementers.
I think adequate procedures exist for gathering implementation
experience for the IESG to evaluate protocol interoperability.
Ietf mailing list