Re: [Json] Consensus call: establishing name equality

[John Cowan <[e-mail]>]

Carsten Bormann scripsit:

> Hmm, a bit too much MUSTard for my taste.  Only the first MUST in 1
> is really needed.

Not so, and for this reason:  in 1 and 2 it says "comparisons MUST be
conducted, after all unescaping is done, by comparing" etc.  That does
not positively require that unescaping be done.  It can be read as
"comparisons MUST be conducted, after all unescaping is done (if it
is done at all)".  Only 3 positively requires unescaping to be done
before comparing.

It has been said that a standard (and this is a Standards Track RFC)
is a contract between the implementer and the user.  We ought not to
disregard the possibility that some implementers will be acting in bad
faith, trying to find whatever loopholes they can, and that some users
will be suing them for non-compliance.  Insofar as any form of words
can capture it, we should make sure that implementers MUST do what users
can properly expect them to do.

So why didn't I say that I couldn't live with 1 or 2, you may wonder?
Because I'm willing to swallow a certain amount of lack of precision in
the search for consensus.

-- 
You are a child of the universe no less         John Cowan
than the trees and all other acyclic            http://www.ccil.org/~cowan
graphs; you have a right to be here.            [e-mail]
  --DeXiderata by Sean McGrath