Re: Services and top-level DNS names (was: Re: Update of RFC 2606

[Dave Crocker <[e-mail]>]

John Levine wrote:
>> The problem isn't just "inability to use" -- it's that other parties
>> exist who may claim the usage right, and provide citations to RFCs to
>> back up their claim.
> 
> There are several ICANN documents describing the new process that
> include a set of recommendations to guide the process.  You must have
> read them, since you are concerned about this proposal.  Do you think
> that recommendations 3, 4, and 20 are adequate to address this
> problem?  If not, why not?


John, et al,

While I think it entirely appropriate to check whether any of us has a clue 
about what ICANN is actually doing, I do suggest that reference to a document 
warrants providing a specific citation, particularly when there are so many 
possible choices at the ICANN cite.

I'm guessing you mean:

<http://gnso.icann.org/issues/new-gtlds/pdp-dec05-fr-parta-08aug07.htm#_Toc43798015>

and specifically the Recommendations table.

If so...

No. 3 is about legal infringement.  That seems wholly irrelevant to the scope of 
IETF work, although I agree that the ietf list thread has started using language 
that sounds related.

No. 4 says "Strings must not cause any technical instability." which sounds 
exactly within IETF scope covers the gist of the technical aspects of the ietf 
list discussion.

No. 20 seems to have to do with failing a popularity contest.  Probably a good 
escape clause to include, but not all that relevant to our I discussion... I hope.

Let me stress that I do think language like "claim the usage right" makes No. 3 
sound appropriate, but that the scope of IETF RFCs is technical specifications, 
rather than "rights".  While yes, one can say that reserving a name or 
proscribing against the use of a name -- such as a single, top-level label as a 
standalone name -- can be interpreting as declaring a "right", I suggest that an 
IETF discussion will fare much better by focusing on the technical issues that 
lead to the constraints, rather than attempting a quasi-pseudo-meta-legal 
discussion.

Simply put:  If an IETF specification has gone through the IETF consensus 
process and been approved, the requirements and constraints imposed are almost 
by definition technical requirements.

Violating one of them invites instability, if only because it invites variable 
implementations.  At the least, treating such constraints as subject to creative 
interpretation by another body renders all output of the IETF fluid.

And just to press this to a logical conclusion, albeit not one that seems to be 
at issue... yet:  If ICANN believes that the IETF has issued a problematic 
specification, then ICANN needs to ask the IETF to fix it, rather than to have 
ICANN, or anyone else, issue a modification/clarification.

d/

-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net
_______________________________________________
Ietf mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/ietf