Re: [Idr] WG Last Call on Proposed IANA procedures for BGP Well-known communities

["Tony Li" <[e-mail]>]

 
Poll response, with commentary in line...

Largely, I think this whole discussion is a semantic quibble because of a
misunderstanding of the ability to declare something 'well-known'.  If we
simply select some other arbitrary term (e.g., global) and define it
reasonably, this entire discussion becomes moot.


|Which way do you support Brian's proposal?
|No selection limitation
|   [X] Support generally the adoption of such a proposal
|   [  ] Do not support generally the adoption of such a proposal
|   [X] If yes to the first, support Standards Action plus FCFS
|   [  ] If yes to the first, support just FCFS


I have no objections to Brian's technical content.  However, I'm concerned
about the semantic usage of the term 'well-known'.  As has been discussed,
this bit of terminology has been around for a very long time and we can't
easily go redefining it without creating mass confusion.  Thus, I believe
it's very difficult to add new well-known communities (or path-attributes)
at this late date.  We should simply accept this reality and move on.


|Well-Known vs. Standards-Action
|You may select 1 option
|   [  ] Rename Well-Known to Standards-Action?
|   [  ] Add an 'Also Known As Standards-Action' to Well-Known?
|   [x] Neither one


Why not simply have a "standards action" (or similarly named) class of code
points.  One then says that compliance with a given RFC requires a specific
interpretation of a given community value?


|Assignment strategy options
|You may select 1 option
|   [x] Support Early Assignment of Standards-Action 
|Communities, per RFC
|          4020?
|   [  ] Do not support Early Assignment of Standards-Action 
|Communities,
|          per RFC 4020?
|   [x] Allow Early Assignment for Individual Internet Drafts?
|   [  ] Do not allow Early Assignment for Individual Internet 
|Drafts (WG
|          drafts only)?
|
|What does 'Well-Known' status represent?
|You may select 1 option
|   [  ] multiple implementations
|   [  ] other -> please post a reply with the criteria


As defined earlier, well-known implies that it is supported by ALL
implementations that support the base specification.  As those
implementations are already in the field, you cannot subsequently declare
those implementations non-compliant.


|Allotted time for achieving 'Well-Known' (permanent) status 
|for any Early 
|Assignments?
|You may select 1 option
|   [  ] 6 months
|   [  ] 1 year
|   [  ] 18 months
|   [  ] 2 years
|   [X] other -> please post your criteria
|   [  ] never


Why not just allow the standards track to progress?

Tony


_______________________________________________
Idr mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/idr