> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of John 
> Levine
> Sent: Tuesday, September 13, 2011 8:07 AM
> To: [email protected]
> Cc: [email protected]
> Subject: Re: [marf] Revisiting reporting addresses
> 
> At abuse.net, that's handled completely manually, both the evaluation
> and the selection of addresses.  This is policy stuff that is utterly
> outside the scope of a fricking reporting format.

I agree that what's being discussed isn't a protocol matter, but it's not 
impossible for material like this to be in a BCP or other advisory document.  
The question is really whether or not this is stuff we want or need to say 
beyond the BCP we're already producing.

I don't think we could codify, in protocol at least, anything at all about 
WHOIS.  In its current form it isn't a reliable source of any kind of 
information.  Sometimes what you need is there, sometimes it's not; sometimes 
it's there but not in a form an automated system can extract.  This might be 
helpful to include in advice someplace though if we can find text on which we 
can agree.  All of that might change if the WEIRDS effort is successful, but 
for now WHOIS is a wild west.

I think it's clear we want not to include domain names in any reporting 
addresses for the reasons John stated.  The dkim-reporting document has already 
been updated to say so, and in my opinion the spf-reporting and 
reporting-discovery documents should follow.

This all leads into the issue of whether or not we want to do some advisory 
text about FBLs outside of the context of private agreements.  I'll comment 
about that on a separate thread.

-MSK, as participant
_______________________________________________
marf mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/marf

Reply via email to