Re: [v6ops] I-D Action: draft-ietf-v6ops-ula-usage-recommendations-04.txt

[David Farmer <[e-mail]>]

> On Nov 5, 2015, at 00:40, Mark Andrews <[e-mail]> wrote:
> 
> In message <[e-mail]>, David Farmer writes:
>>> On 11/4/15 19:41 , James Woodyatt wrote:
>>> Everyone-
>>> 
>>> This draft A) uses normative requirements language with RFC 2119
>>> keywords and B) calls for publication in the Informational category. Is
>>> the document really intended for the Best Current Practice (BCP)
>>> category? If not, could we please either remove the citation of RFC 2119
>>> and all the usage of upper case keywords, or insert an explanation for
>>> why the document uses normative requirements language despite being an
>>> Informational category document?
>> 
>> The only use of use upper case RFC2119 keywords in the document are in
>> quotes from other documents.  Such use seem appropriate, even if you
>> think upper case RFC2119 keywords are not normally appropriate in an
>> Informational category document.  Since quotes using upper case RFC2119
>> keywords are included I think the citation of RFC 2119 is also
>> appropriate and clarifies their use.
> 
> Then you have to think about all use of "must", "should" etc. in
> terms of RFC2119.  For reference I've got drafts which quote
> documents with RFC2119 semantics but have not cited RFC 2119.

Personally, I don't like the inclusion of any capitalized  RFC2119 key words without the inclusion of a RFC2119 reference to clarify their meaning.

> There is also nothing preventing the use of RFC 2119 in BCP or INFO
> documents.  The use of RFC 2119 does not imply standard.

I agree, however, I decided not to argue that point.  In the document in question the author choose to only capitalize the RFC2119 key words in the quotes from other documents and therefore included the reference, this seemed reasonable to me in the context of the document and well within the discretion of the author.   I expect even for those who think Informational documents should not use capitalized RFC2119 key words, they would recognize and exception for quotes from other documents. 

I believe it is reasonable to use use capitalized RFC2119 key words in any document anywhere and reference RFC2119.  For Informational documents, within reasonable limits, this should be a style question left to the discretion of the author.  For Standards Track or BCP documents there is an expectation to use capitalized RFC2119 key words when make normative statements.  But even this isn't an absolute.  In my opinion even here, within reason, the author should have discretion, but for clarity I believe it is necessary for an author to explain any deviation from this expectation. 

-- 
===============================================
David Farmer                          Email: [e-mail]
Office of Information Technology
University of Minnesota    
2218 University Ave SE         Phone: [phone]
Minneapolis, MN 55414-3029   Cell: [phone]
===============================================