Re: [Json] Benoit Claise's No Objection on charter-ietf-json-00-01: (with COMMENT)

[Barry Leiba <[e-mail]>]

> Probably a detail, but it puzzles me. Maybe I read too much into this...

I think you are.

> "Any changes that break compatibility with existing implementations of
> either RFC 4627 or
> the ECMAScript specification will need to have very strong justification
> and broad support."
>
> Versus
>
> "Any changes that break compatibility with the RFC 4627 or
> the ECMAScript specifications will need to have very strong
> justification
> and broad support."
>
> Is this intentional that you mention the existing implementations of RFC
> 4627?

Yes.

Consider that this is an essentially similar situation to what we had
in going from PS to DS... except here we're going from Informational
to PS.  The point, though, is that we have a mature protocol that's
moving in the Standards Track (here, *into* the Standards Track).  The
justification required for a change increases with the level of
disruption it would cause to what's out there.

Errata are easy.
Useful changes that are fully compatible with current implementations
are still easy; the conversation should be brief.
Important changes that break compatibility are still within scope, but
require a very strong justification, and will probably involve a lot
more discussion.

Your version (the second above) doesn't make much sense: changes to
the spec are changes to the spec.  The point is to consider the effect
those changes will have on implementations.

Barry