On Aug 5, 2015, at 9:39 PM, Mark Andrews <ma...@isc.org> wrote: > If the attacker has a good cookie then you have a high degree of > confidence that the IP address is correct even if it a UDP request > and you can take steps like contacting the operators of the network > / server.
After some pretty intense criticism of this document, I went radio silent last week for two reasons. The first is that Mark said this, and it caused me to rethink my position on the document. The second is that my mail server exploded in a rather exciting way, and I spent the weekend writing an SMTP server so that I didn’t have to install Postfix again. So the short response to this is that I can now see how the draft is useful, but I think the draft fails to communicate why it is useful, and communicates some things that it’s purportedly for that aren’t useful. However, it may also be that I simply didn’t read it carefully enough. So I want to give it another thorough read and possibly suggest some changes that I think would result in readers having a clearer understanding of what it does that it useful, and not coming away with the impression that it does other things that I think could actually be harmful, but at the very least are not useful. I think the changes I would propose would be to the explanatory text, not to the specification. I think it would be useful to consider these changes before concluding the WGLC, but I am sure that the authors would like to see forward progress, and I’m not notoriously a fast reviewer. So I mention this for the chairs’ information, but have no real expectations as to what they might do in response. Thanks to Mark and Donald for their patient responses to my questions.
_______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
