[radext] Re: Roman Danyliw's Block on charter-ietf-radext-07-03: (with BLOCK)

[Alan DeKok <[e-mail]>]

On Jun 12, 2026, at 3:07 PM, Roman Danyliw via Datatracker <[e-mail]> wrote:
> 
> Roman Danyliw has entered the following ballot position for
> charter-ietf-radext-07-03: Block
> ...
> Per "Define and publish minor extensions to RADIUS, such as new attribute
> definitions.  If the RADIUS attributes are defined outside the RADEXT working
> group, RADEXT will be responsible for reviewing those.", can the following be
> clarified:

  I think if there's any issue with the text, there is likely consensus to just delete it.  There is significant new work which is waiting for charter updates.

> -- Does "outside of the RADEXT WG" mean outside of the IETF?  Assuming yes, is
> RADEXT waiting on liaisons statements?

  Yes, see recent emails from the WBA.

>  How does the WG know to take action if
> not such statements?  If the attribute definition is happening outside of the
> IETF why is there any work to do at all?  Wouldn't code points be handled by
> the DEs?

  If the text is removed, this issue is moot.

  However, I do want to highlight a related issue with the IETF process.  IANA manages protocol allocations, and does have Designated Experts for some registries.  However, the main RADIUS Attributre Type registry is marked "IETF Review".

  Note that this is *not* Designated Expert Review, or even "Working Group Review".  This is a hole in the IETF process.

  I've seen situations where a protocol registry has updates requested from outside of the WG.  Since there is no process to notify anyone, even a live WG is not notified that the allocation is taking place.  This practice avoids a clearly useful process of WG review.

  The result is that the specification does not follow WG practices, or even previously published BCPs / guideline documents for that protocol.

  The text in the charter was an attempt to suggest that the RADEXT WG should be consulted for issues related to the RADIUS protocol.  Since this is controversial, it's best to simply remove that text.

  Alan DeKok.