Speaking of the devil in the details…
On Dec 4, 2012, at 3:59 AM, "Andrew G. Malis" <agmalis(_at_)gmail(_dot_)com>
wrote:
Stephen,
Your goal is laudatory, but the devil will be in the details. For example,
you wrote:
Note also that this experiment just needs an implementation that
makes it possible for the WG chairs and responsible AD to verify (to
the extent they chose) that the implementation matches the draft.
Will this require WG chairs and/or document shepherds to do a code review to
verify that the implementation and code match? A better criteria might be
that there be at least two independent implementations that successfully
interoperate. That would also show greater WG interest than just a single
individual or organization.
Define "independent". If the draft author contributes to two open source
implementations, does it count as independent? What if one author contributes
to an open source implementation, the other author implements it in her
company's product, and they claim interoperability? That doesn't say much
about the quality of the spec, other than that the authors understand it.
Open source code is a plus, but shouldn't be a requirement, as such a
requirement might discourage some vendors from implementing.
Thanks,
Andy