Re: [sipcore] draft-ietf-sipcore-rejected Terminology

[Eric Burger <[e-mail]>]

DeMorgan’ theorem for Standards, which I forgot:
x UNLESS y is the same as FOR ALL NOT y, MUST x.
Thanks!

Anyone else?

> On Mar 6, 2019, at 9:28 AM, Brian Rosen <[e-mail]> wrote:
> 
> 5) Cut the second Terminology paragraph.  Reword the instances into If <condition> MUST else SHOULD, but be explicit about the condition.
> 
> 
> 
>> On Mar 5, 2019, at 5:24 PM, Eric Burger <[e-mail]> wrote:
>> 
>> Grumble, grumble. That’s my civil disobedience to (1) put the 2119 language into active voice and (2) point out that SHOULD is useless without UNLESS. The various levels of folding on my part would be:
>> 
>> (1) leave it as is
>> 
>> (2) cut the second Terminology paragraph and turn the ‘UNLESS' instances into ‘unless' instances.
>> 
>> (3) cut the second Terminology paragraph and remove the unless clause, turning the SHOULD into the useless SHOULD.
>> 
>> (4) cut the second Terminology paragraph and use the evil passive voice version of the 2119 Terminology paragraph, written by non-writers
>> 
>> How low shall I go? ;-) I’m willing to go all the way if demanded, but my vote would be to keep the language as is. However, it’s only one vote - I’m the editor not author and it’s a work group document, so I will bend to the will of the work group.
>> 
>>> On Mar 5, 2019, at 2:01 PM, A. Jean Mahoney <[e-mail]> wrote:
>>> 
>>> Hi Eric and Bhavik,
>>> 
>>> I've been working on the Document Shepherd write-up for the sipcore-rejected draft. The draft's Terminology section and the normative text that uses the term "UNLESS" need updating:
>>> 
>>> 2.  Terminology
>>> 
>>> This document uses the terms "MUST", "MUST NOT", "REQUIRED", "SHALL",
>>> "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and
>>> "OPTIONAL" as described in BCP14 [RFC2119][RFC8174] when, and only
>>> when, they appear in all capitals, as shown here.
>>> 
>>> As a matter of principle, this document uses the term "UNLESS" to
>>> distinguish when a "SHOULD" means "MAY".  Otherwise, the "SHOULD"
>>> means "MUST".  Without UNLESS, SHOULD is meaningless.  A.k.a.
>>> Burger's Law of Protocol Meaningless Terms.
>>> 
>>> 
>>> The following two statements are covered by the "UNLESS" addition:
>>> 
>>> [1]
>>> If an intermediary issues a 608 code, the intermediary SHOULD include
>>> a Call-Info header in the response, UNLESS there are indicators the
>>> calling party will use the contents of the Call-Info header for
>>> malicious purposes (see Section 6).
>>> 
>>> [2]
>>>        As such, the intermediary SHOULD play the announcement,
>>> UNLESS the caveats described in Section 3.5 and Section 6 hold.
>>> 
>>> 
>>> Would you rewrite these statements so that they only use the key words described by [RFC2119][RFC8174]? As the statements are currently constructed, it would be somewhat difficult for an implementer to know what to do since both statements point to largish sections of descriptive text.
>>> 
>>> Thanks!
>>> 
>>> Jean
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>>> _______________________________________________
>>> sipcore mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/sipcore
>> 
>> _______________________________________________
>> sipcore mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/sipcore
>