Glen Zorn wrote:
This document is 36 pages long, yet devotes only a single paragraph to the
use of extended RADIUS attributes.
The goal was to finish the design document significantly before the
extended attributes document was finished. Due to normal WG issues,
this didn't happen.
Since the extended attribute set is
likely the one to be most used in the future, this seems a rather gross
oversight.
The extended attributes document is referenced from the guidelines
document. It suggests that designs not meeting the criteria of the
guidelines document use the practices documented in the extended
attributes document.
I would suggest that the authors try to design a few extended
attribute sets and then document that experience.
How does this help the guidelines document?
The extended attributes document meets WG consensus, and follows the
traditional RADIUS data model for attributes. It would seem that
everyone agrees that the specification it proposes is satisfactory, and
meets their needs.
I mention this at this
late date because I expected this note to become an RFC long before the
extended attributes doc, but (FWIW) the latter has completed WG "Last"
Call...
The guidelines document has seen multiple last calls, with last-minute
comments at each one.
Alan DeKok.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf