On 4/18/2018 5:07 PM, Suzanne Woolf wrote:
Second, does this work add significantly to the complexity of the DNS protocol, or the work of implementers and operators?
A slight respin of this bullet: Does this work change the way the infrastructure of the protocol works, or is this work targeted at users of data stored within the DNS? The former class is things like DNSSEC, CNAMEs, DNAMES, CDS/CDNSKEY; the latter is something like DANE.
I'd also add a 4th bullet: "Fourth, does this work require wide deployment to be useful?" - this question taken together with my respin of the second item is trying to suggest that the class of proposals that just add data to the DNS might be acceptable even if the expected deployments are limited, whereas a proposal that requires the servers, resolvers and clients to know something more than just serving the data should require a consideration of whether the deployment be wide, if not, then why not, and how do we propose to get to a wide deployment?
Later, Mike _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
