Re: [MIB-DOCTORS] normative references and REFERENCE clauses

["Romascanu, Dan (Dan)" <dromasca at avaya.com>]

+1

Dan
 

> -----Original Message-----
> From: mib-doctors-bounces at ietf.org 
> [mailto:mib-doctors-bounces at ietf.org] On Behalf Of Bert (IETF) Wijnen
> Sent: Tuesday, October 06, 2009 12:15 PM
> To: mib-doctors at ietf.org
> Subject: Re: [MIB-DOCTORS] normative references and REFERENCE clauses
> 
> Juergen Schoenwaelder wrote:
> > Hi,
> >
> > I have a MIB module in AUTH48 and the following question popped up.
> > The MIB modules has something similar to this:
> >
> > FancyString TEXTUAL-CONVENTION
> >     -- description etc. removed with a normative reference to 3629
> >     REFERENCE
> >        "RFC 3629: UTF-8, a transformation format of ISO 10646"
> >     SYNTAX OCTET STRING
> >
> > Since the content of this RFC to be normatively depends on 
> RFC 3629, I 
> > included RFC 3629 in the normative reference section. As RFC 4181 
> > points out, the RFC editor has a policy that you can't have a 
> > reference in an RFC that is not cited and they do not like 
> to cite the 
> > reference in the MIB module because the MIB module can be extracted 
> > and thus there is a dangling citation. Given this, it seems 
> I have two
> > options:
> >
> > a) Remove RFC 3629 from the normative references. This is 
> the easiest
> >    solution but then any tools analyzing normative dependencies
> >    between RFCs will likely miss a normative reference, which sounds
> >    like a bad idea.
> >
> > b) Add extra text somewhere outside the MIB module to establish a
> >    citation to RFC 3629. This adds some verbiage but 
> ensures that the
> >    normative dependency gets properly recorded.
> >
> > The third option, namely to try to change the RFC editor 
> policy, I did 
> > not consider seriously. ;-)
> >
> > I checked RFC 4181 but it only has guidelines on IMPORTs. So do we 
> > generally prefer option a) or b). If there is a clear preference, 
> > should 4181bis be extended to handle the situation where REFERENCE 
> > clauses refer normatively to other documents that are not cited 
> > outside the MIB module?
> >
> > /js
> >
> > PS: I am bringing this question up here because I am editing a much
> >     larger document where this issue is much more dominant and so I
> >     want to be consistent.
> >
> >   
> I think my preference would be to add a small piece of text 
> right before the module where you explain the normative dependency.
> 
> just my 2 cents.
> 
> Bert
> _______________________________________________
> MIB-DOCTORS mailing list
> MIB-DOCTORS at ietf.org
> https://www.ietf.org/mailman/listinfo/mib-doctors
>