On 1/26/25 2:33 PM, Pete Resnick wrote:
Sorry, a previous version was completely open ended; it's been hard to track this. But I still don't get why preferring backward compatible changes would lead to "nonsense". If the required changes lead to nonsense, sure let's not do that. But it doesn't seem a priori that those kinds of breaking changes are required. It's not like we didn't expect that there could be upgrades in the future from the start, after all. Why should we not prefer to work within the framework of what is already deployed if we can?On 26 Jan 2025, at 16:22, Michael Thomas wrote: On 1/26/25 2:16 PM, Pete Resnick wrote: |The working group will pursue incremental enhancements to DKIM and/or DKIM use, where possible. It will pursue parallel or replacement mechanisms only where incremental change is not feasible. | I agree with Richard here: The reformulation opens the door to a great deal of nonsense. I fundamentally disagree with making such a change. In what way? It seems to me that wholesale wheel-reinvention is what usually leads to nonsense.The original in Murray's version says, "The working group will observe the success of current technologies, primarily DKIM, reusing its techniques where applicable, to develop a new technology suite that seeks to address these concerns." That proscribes wholesale wheel-reinvention already; the WG will reuse the technology where applicable. The reversal in the proposed change is the problem: Only use different mechanisms when it is infeasible (I think some may read "impossible") not to. That's too high a bar.
I personally think there should be some general divide and conquer with some of the goals on the documents front where easier things get done faster, harder ones slower, and if existing DKIM deployments can take advantage of that, that would be great (I'm thinking of the mailing list change annotations in particular). If you're zeroing in on "where possible", fine that could be softened, but the general idea of allowing existing DKIM deployments to take advantage of some or all of the new work where it can seems like Good Thing, to me.
Mike
_______________________________________________ Ietf-dkim mailing list -- [email protected] To unsubscribe send an email to [email protected]
