Re: [core] [Last-Call] [art] Artart last call review of draft-ietf-core-problem-details-05

["Martin J. Dürst" <[e-mail]>]

Hello John,

On 2022-06-25 00:15, John C Klensin wrote:
> 
> 
> --On Friday, June 24, 2022 06:20 -0400 Ira McDonald
> <[e-mail]> wrote:
> 
>> ...
>> CORE - please delete *all* of your CDDL details for language
>> tags and just use one of the several excellent libraries that
>> correctly parse language tags, when needed.
>> ...
> 
> Two problems with that.  The general one is that IETF standards
> track specs rarely (I wish I could say "never") say "go use that
> library" if only because libraries change and are hence not
> stable references.  If this were intended to be an Informational
> document, then maybe.  But, as a Proposed Standard, I don't
> think that flies.

I don't think Ira meant that standards should reference libraries. I 
think what he had in mind was the following diagram:

Tag 38 spec     --------------------> BCP 47 (language tag spec)
      |              references                   |
      |                                           |
      | implemented by                            | implemented by
      |                                           |
      V                                           V
Tag 38 library/function ------------>  language tag library
                           uses

Regards,   Martin.


> The one more specific to this spec is that,
> because there is no standard way to specify directionality in a
> language tag itself, either that functionality gets lost (and
> must hence still be included in those CDDL details) or we need
> to extend BCP 47 to include a language tag specification
> mechanism.  The CORE effort explicitly decided to not use/depend
> on such a change and no one (yet) has challenged that decision.
> There are also substantive arguments why such an addition would
> not be a good idea.

>      john
> 
> 
> If those libraries actually
>