Re: [dmarc-ietf] "psd=" tag early assignment

[Douglas Foster <[e-mail]>]

I see no personal attack.   John was clear, and has been clear, that he has
no intention of documenting any limitations or risks associated with the
tree walk, because in his judgement, they are not important.   My concern
is about a document that creates a new vulnerability, then fails to
document it.  The private registries DO create complexities for the tree
walk, complexities that have been trivialized instead of being mitigated.

But if you wish to endorse the current document.  Let's do so honestly:

"The tree walk is vulnerable to false PASS if certain combinations of
present and missing data occur.   Even though this has happened in the
past, it is the firm opinion of some work group members that this will not
ever occur in the future.   Therefore the details of a non-problem are of
no interest to DMARC participants."

My issue is with the document.

Doug


On Fri, Jul 8, 2022 at 2:50 PM Barry Leiba <[e-mail]> wrote:

> >> So John has confirmed that it is his intent to hide any information
> about private registries, because the
> >> private registries create complexity for his algorithm which he does
> not wish exposed.
> >
> > I submit that equating "this is not worth explaining as it's a corner
> case" to "we should hide this detail
> > because I don't want anyone to know about it" is logically absurd as
> well as baldly antagonistic.
>
> Agreed, and thanks, Murray.
>
> Doug, I've called out others for similar things, and you get it here:
> Please do not attribute bad intent to other participants, and please
> do not put things in terms of accusations or in ways that seem only to
> pour tar.
>
> A response such as, "John, I strongly disagree: I think it's really
> important to talk about at least some uncommon cases in order to make
> the situation clear.  Discussing private registries in one of those
> important cases, as they create complexities for the algorithm that
> need to be shown," gives your technical opinion without being
> insulting or inflammatory.
>
> Barry
>