Re: OPENPGPKEY RRTYPE review - [IANA #773394]

On 11 Aug 2014, at 11:01, Paul Wouters <[email protected]> wrote:

> Did I miss anything in particular that you thought needed to be added?

I think the most important thing that this document can do is present a clear and unambiguous specification, to promote interop between implementations.

Describing the design choices that were made has some value, but mainly for the historical record or to provide input into future design choices that other people might make in similar situations. There's next to no value added for implementors by doing this (note that I'm not saying it has zero value).

Similarly, use cases and examples are great if they provide useful clarity. If they don't (if the specification is clear enough without them), they're not necessary.

A lot of this is subjective.

I think in general the working group should be concerned with whether a document meets its reasonable objectives, and should not attempt to micro-manage the authors into adding orthogonal text if they don't feel like it. In this case, there's no actual need for the wg to be formally involved at all; this could easily progress on its own merits as an individual effort (and it's not a working group draft).

I like what you have done.

Not every RRType specification needs to turn into War and Peace.


Joe
_______________________________________________
dnsext mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsext