Re: #481, was: WGLC: p7 MUSTs

[Alex Rousskov <[e-mail]>]

On 06/30/2013 10:48 AM, Julian Reschke wrote:
> On 2013-06-09 20:49, Alex Rousskov wrote:
>> If you say "server MUST NOT send X", the proxy becomes responsible for
>> not forwarding X. If you say "server MUST NOT generate X", the proxy
>> forwarding behavior is not restricted by that specific requirement. When
>> you say "request MUST NOT have X", the specs become ambiguous: some will
>> claim that a proxy forwarding X is in violation and some will claim that
>> the requirement is not applicable to proxies.

> The trouble is that what you're asking for a change in requirements, and
> that most definitively is *not* an editorial change. 

Whether polishing how these ambiguous requirements are worded actually
changes those requirements depends on whether the reader believes that
the proxy must police the given aspect of the message. Some readers may
indeed decide that your polishing is not editorial in nature, depending
on how you change the specs. The very fact that you suspect there will
be protocol changes essentially implies that the current requirements
are ambiguous and ought to be fixed.


> As such, I'm not
> too enthusiastic to make these kind of changes without feedback from the
> working group.

On the other hand, it is difficult to provide feedback without seeing
the changes.


> Do people agree that these requirements need to be rephrased? Do we have
> concrete proposals about *how* to change them?

FWIW, I do: Reword them to name the actor (client or server, usually
obvious) and use "generate" instead of "send". When that default does
not seem appropriate to you or others, let's discuss!


Thank you,

Alex.