Re: [Lsr] When is an IANA Registry Required

["Les Ginsberg (ginsberg)" <[e-mail]>]

Alvaro -

Inline.

> -----Original Message-----
> From: Alvaro Retana <[e-mail]>
> Sent: Wednesday, March 17, 2021 7:04 AM
> To: Les Ginsberg (ginsberg) <[e-mail]>; draft-ietf-lsr-isis-srv6-
> [e-mail]
> Cc: John Scudder <[e-mail]>; [e-mail]; [e-mail];
> Christian Hopps <[e-mail]>
> Subject: RE: When is an IANA Registry Required
> 
> On March 16, 2021 at 6:24:22 PM, Les Ginsberg wrote:
> 
> 
> Les:
> 
> Hi!
> 
> 
> > But one thing I find missing in your response is some info on what problem
> > YOU think needs to be addressed?
> 
> 
> I simply think that the specifications are not complete without
> guidance on how to use/assign the unused bits.  I rather try to close
> that door before a potential problem may come up.
> 
[Les:] Specifications typically say:

"Remaining bits are RESERVED. SHOULD/MUST be transmitted as 0 and MUST be ignored on receipt."

And the same wording is used for bits covered by a registry.
Consider https://www.rfc-editor.org/rfc/rfc7794.html#section-2.1 (for which we do have a registry):

"Undefined bits MUST be transmitted as 0 and MUST be ignored on
   receipt."

> In the extreme case anyone can make use of the bits, through the ISE
> or a different SDO -- ideally we will be paying attention, but may
> not.  Sure, a registry doesn't stop implementations from squatting on
> codepoints either (even inside the IETF), but at least people have to
> want to bypass the allocation rules.
> 

[Les:] When a specification says "all other bits are reserved, MUST be sent as 0 and MUST be ignored on receipt", I do not see how anyone can assume they can use them - nor why the existence of a  registry would prevent such a squatter from doing whatever they want outside of normal channels.
Are you suggesting that normative statements in RFCs simply don’t mean anything?

Sorry, don’t mean to be argumentative. Just expressing why I am not understanding/agreeing w your POV.

   Les

> My 1c.
> 
> Alvaro.