Re: [yang-doctors] SHOULD statements in rfc6087bis

[Juergen Schoenwaelder <[e-mail]>]

On Sat, May 26, 2018 at 12:15:03PM +0200, Martin Bjorklund wrote:
> 
> The whole point of having a SHOULD instead of MUST is that there are
> cases where the rule/recommendation simply doesn't need to be
> followed.  IMO this is a perfect example of when the description isn't
> needed.  The WG has decided that description for an enum is a SHOULD,
> not a MUST, so why do we not follow this guideline?
>

  (i) Because people want a clean compile.
 (ii) Because people do not want to have to enter a discussion every
      time whether an exception to the SHOULD applies or not.
(iii) Because people do not want to hold off important technical work
      debating weeks whether a non-harmful redundant description should
      be added or not.

And at the end of the day, if the SHOULD forces some redundant
descriptions, then this is likely less harmful than missing
descriptions and overall the SHOULD does its job even if it is
interpreted effectively as a MUST unless you have lots of time and
energy for a lengthy debate.

/js

PS: Have you never made changes to C code to get a clean compile even
    if the change was not strictly needed?

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
Fax:   [phone]         <https://www.jacobs-university.de/>