Re: [dmarc-ietf] ARC questions

[Michael Thomas <[e-mail]>]

On 11/23/20 12:48 PM, Dave Crocker wrote:
>> This recent article also goes into things that DKIM signatures imply:
>> https://blog.cryptographyengineering.com/2020/11/16/ok-google-please-publish-your-dkim-secret-keys/ 
>> <https://blog.cryptographyengineering.com/2020/11/16/ok-google-please-publish-your-dkim-secret-keys/>
>
> The level of condescension, ignorance, and error throughout that 
> article is impressive.  Given that it was written by someone whose 
> profession requires extreme care about complex matters, the level of 
> carelessness in the article is especially unfortunate.
>
> Conveniently, he put his biggest error in bold font:
>
>      "*DKIM provides a life-long guarantee of email authenticity that 
> anyone can use to cryptographically verify the authenticity of stolen 
> emails, even years after they were sent."*
>
> DKIM does no such thing.
>

Yeah, that was pretty bad. "DKIM can be used to verify a piece of mail 
due to operator practices, but there are absolutely no guarantee that a 
signature will verify in the future due to those same practices."

>
> ps. making sure that DKIM signature become invalid  relatively soon -- 
> I think that removing the keys is simpler and just as effective as 
> publishing the private keys -- seems like a reasonable suggestion.
>
>
Stephen Farrell is threatening to write an ID on the subject of 
publishing private keys. Frankly the stakeholders -- providers and users 
-- are not very well aligned on when where and why a provider would do 
such a thing. And writing an ID to say how to invalidate key when just 
unpublishing old selectors when you rotate keys is an easy second best 
shows that inertia is the actual issue, not the technical shortcoming.

Mike