On 3/29/2018 3:38 PM, Adam Roach wrote:
I still don't fully understand the nature of the objections I cite above
or the assertions that having separate tables for different RRTYPEs is
somehow broken. Based on my (admittedly lay) understanding of how DNS is
used by other protocols, I agree with your proposal that having distinct
tables for each RRTYPE makes far more sense than the current structure.
1. Another round of thanks. On re-starting the effort, I'd missed that
exchange. I think adding
"Scope is meant as a static property, not one dependent on the
nature of the query. It is an artifact of the DNS name."
from it, to the Intro will help a bit.
2. I think the latest round of discussion and change arguably implements
your view, albeit within a single actual registry.
d/
--
Dave Crocker
Brandenburg InternetWorking
bbiw.net
_______________________________________________
DNSOP mailing list
DNSOP@ietf.org
https://www.ietf.org/mailman/listinfo/dnsop
