Re: [urn] Are ISO 3166 country codes stable enough for URN use?

Ok.

It sounds like the durability and persistence requirements for URNs are 
being relaxed somewhat, compared to the old days.

I'm fine with that, carry on.

Sean

On 2/21/2016 12:35 PM, John C Klensin wrote:
> Sean,
>
> Very briefly and having spend far too much time immersed in a
> well-known use of 3166 and it implications, two observations:
>
> (1) The "we can, and probably will, reassign a code after five
> years" rule have been dead for some years.  You should be
> careful about what you treat as an authoritative reference.
> Because of (2), I may regret even mentioning that.
>
> (2) The definition of "persistent enough" or "stable enough"
> inevitably has to be per-namespace.  One of the earliest
> examples in the discussions that led to URNs was "the weather
> in..." which can be as stable as location names are as a
> reference but whose object is typically entirely unstable.
>
> This isn't a WG topic.  When the NDN spec is revised, you can
> certainly have the discussion in the context of that revision
> (assuming they don't simply decide to register something based
> on a document published elsewhere which, fwiw, is certainly what
> I would be inclined to do if I were in their position).
>
>     john
>
>
> --On Sunday, February 21, 2016 11:45 AM -0800 Sean Leonard
> <dev+ietf@seantek.com> wrote:
>
>> In reviewing the latest draft-ietf-urnbis-ns-reg-transition, I
>> noticed the dependency on NBN [RFC3188], which in turn depends
>> on ISO 3166.
>>
>> An example is <URN:NBN:fi-fe19981001>.
>>
>> URNs are supposed to be stable and durable over time...
>>
>> but are ISO 3166 country codes stable enough for URN?
>>
>> I have not actually read all of the details of ISO 3166. But
>> ISO 3166 grapples with an intensely political question: how to
>> name countries. I read this PDF:
>> http://unstats.un.org/unsd/tradekb/Attachment194.aspx
>>
>> And it says that "withdrawn [codes] may not be reused for five
>> years". That means that they can be reused in five years. I
>> think that violates some URN principles.
>>
>> NBN isn't the first time we have seen ISO 3166 country codes
>> in URNs. There is the urn:lex proposal (not really sure where
>> that is now).
>>
>> There are also my proposals for urn:xmlns and urn:rdf. However
>> in my proposals, the entire NSS is assigned on a first-come,
>> first-served basis: the relationship between two- and
>> three-character ISO 3166 country codes being part of the NSS
>> is purely a coincidence and therefore reassignment of a
>> country code does not affect the durability and immutability
>> of the assigned string.
>>
>> Sean
>>
>> _______________________________________________
>> urn mailing list
>> urn@ietf.org
>> https://www.ietf.org/mailman/listinfo/urn
>
>
>