RE: [security-services] PE78: Reassignment of persistent identifiers

[robert.philpott@rsa.com]

> > Well, MUST NOT is a stretch for an erratum (as pointed out by Eve on
> > the call), and SHOULD NOT is not all that useful IMO, so perhaps a
new
> > format is needed, yes.
> 
> Without disagreeing, a MUST NOT is fine in an errata if the spec meant
to
> say that and just didn't. Then you ask for input as to whether anybody
> implemented or deployed based on assuming it wasn't a MUST NOT, and if
not,
> you're good to go. We've done it before.

I agree w/Scott.  Adding a MUST NOT to clarify the SSTC's intended
meaning is not unreasonable in the Errata.

I don't recall the specific issue, but as he said, we have done it in
the past.