[ietf-dkim] Splitting the DKIM base doc
lear at cisco.com
Sun Mar 26 00:28:12 PST 2006
Michael Thomas wrote:
> A split is guaranteed to cause that date to slip, fullstop.
> As I've said, from a developer's standpoint, multiple documents
> suck. But here's what I don't understand: if we split the documents,
> we will have to submit all of the former parts of -base as a bundle
> since they will necessarily have cross normative references. And
> even if we could, a -base document that doesn't define, say, the
> query mechanism and/or the contents of the RR would be completely
> pointless -- and an invitation for people just winging it.
I heartily agree with Mike on this one. I think a split of the base
document along the lines discussed in the WG is a HUGE mistake. The
purported reasoning for the split at the meeting was to carve out the
DNS RR. Without this information defined the spec is useless, which
means that the base spec would have to gate on the RR definition
*anyway*. Hence, if we are going to define an RR (and we should) we
have two choices:
* define it now, include it in the base spec and be done with it; or
* define it later and expect that the next RFC will be classed an
update to the standard.
More information about the ietf-dkim