Hi Med,
> +1 to Quifang's comments. Effectively, I think that this was a limitation in
> the original NMDA architecture that is being corrected.
I agree that the <system-config> document should be seen as correcting RFC 8342
(NMDA).
Separately, I think the WG made a mistake when publishing RFC 8808, which
should have also *updated* RFC 8342. It is unfortunate (dare I say tragic), as
the <factory-default> datastore remains hidden to many. IDK if there exists a
mechanism to change the "update" status of a document after publication. One
might ask, how exactly RFC 8808 updates RFC 8342? As, technically, deriving an
"identity" off of a base identity doesn't seem to raise to the level of an
"update". That said, I wish that RFC 8808 had in it a diagram, like the one
below, which could be construed as an update?
+-------------+ +-----------+
| <candidate> | | <startup> |
| (ct, rw) |<---+ +--->| (ct, rw) |
+-------------+ | | +-----------+
| | |
+-----------+ | +-------------------+
| <running> |<--------+ | <factory-default> |
| (ct, rw) |<--------------| (ct, ro) |
+----+------+ +-------------------+
|
<SNIP/>
Kent // as co-author of RFC 8342 (NMDA)
_______________________________________________
netmod mailing list -- [email protected]
To unsubscribe send an email to [email protected]
