On 1/20/25 4:42 PM, Murray S. Kucherawy wrote:

    I guess the reason I ask is whether there is actually some
    immediacy to
    this. It's been 20 years since DKIM came around and has been pretty
    quiet as far as I know. If there is -- or at least some of it is
    -- it
    might be better to separate tasks that need to be immediately
    addressed
    so as not be boat anchored to things that can wait.


I wouldn't say it's been quiet.  There's a trail from STD 76 littered with abandoned drafts attempting to address one issue or another, but none of them have achieved enough critical mass to reach publication or broad implementation. I would agree though that there hasn't been any urgency in all that time.

One way to look at this effort is that it's a collection of those abandoned concerns all being addressed, finally, in one shot, with the momentum of a handful of large operators.

I mean, I'd expect this to take 5 years at a minimum. The original took about that and nothing much changed once IIM and DK were merged. That and I wasn't being facetious about second system syndrome: with computer engineering, there are those who have read Fred Brooks (rip) and those who repeat the mistakes blindly (the rest of us repeat the mistakes knowing better :).

But my larger point is that it might be better to divide and conquer on the documents front. The "message changes" algebra seems like a tractable problem -- it could just be a header that is signed by whatever -- but other things like the so-called replay problem could take years and then still flop. Requiring an entire package of changes in one document could needlessly delay implementation and deployment of the easier things.

Mike
_______________________________________________
Ietf-dkim mailing list -- ietf-dkim@ietf.org
To unsubscribe send an email to ietf-dkim-le...@ietf.org

Reply via email to