On Mon, Nov 5, 2018 at 2:25 PM Scott Kitterman <[email protected]> wrote:
> I don't think it's something we should delay on. In my, admittedly > limited, > experience with these things, once something is in an experimental version > of > an RFC, then it 'has' to be preserved in the name of interoperability with > the > installed base. I think now is the time to remove obvious fluff. That's the opposite of my understanding of Experimental. Rather, I believe explicitly calling it an experiment means we reserve the right to mutate or even jettison portions of ARC we discover, through experimentation, interfere with interoperability either among ARC components, or with the operation of adjacent protocols, or provide more complexity than utility. If there's ambiguity around what "Experimental" means, we should take the time to sort that out and, if needed, add text to make it clear. This documents makes my head hurt enough as it is. > I've felt that on occasion too. Having reviewed the thread that Kurt pointed me to, it seemed like this is > something only one person wanted. It didn't appear to have a lot of push > behind it. > Based on my understanding of Experimental, I think a one-off feature is fine to include, again with the understanding that it could be omitted from a Proposed Standard version because it isn't widely useful. -MSK
_______________________________________________ dmarc mailing list [email protected] https://www.ietf.org/mailman/listinfo/dmarc
