There is an understandable focus on "how are you going to block French users" when we have also claimed that geoIP blocking is insufficient for selectively denying resolution for certain names - we understand this conflict, but we think it is energy spent in the wrong place.

It is probable that we will use geoIP filters to block French users from our services after November 24. We could apply this same geoIP filter list now to selectively block French users from reaching the sites that we've been told to make unavailable. But this would corrupt our core premise, which is that the DNS is not the right place for this to happen, and if we comply within France to meet the requirement then we accept the worst eventual outcome for users and we still do not have legally perfect coverage. We wish to first remain consistent with our core mission, and secondarily we must remain as close to compliance as we can to the legal constraints put upon us. By accepting the demands for blocking, we would admit that such demands are permissible within our mission - that we will act on behalf of companies (or others) to create ineffective and mis-directed regimes of content constraint.

We would be admitting that we think this solves the problem - it does not. We would be admitting that we will do this for an ever-increasing number of names - we will not. We would be permitting these methods to be applied for more areas of filtering such as moral, religious, social, or political reasons - we must not.

So we accept that there is the possibility that we will have errors in our blockade, but in this case (blocking French users) we maintain consistency with our stated mission of integrity of the DNS. We would rather there be a clear distinction made for users of our services: you get the DNS as you (the user) expects, or if Quad9 is forced to pollute the service with unexpected and arbitrary number of fragmentation demands, then we will not offer the services at all.

The underlying issue is not ONLY that geoIP blocking is ineffective, but that the entire premise of any of these demands is inappropriately applied to DNS recursive resolution providers. We would hope that this discussion continues to be on the most fundamental of the disagreement we have with the outcome in this court case, not on the details of the technical implementation which have yet to be entirely decided and which can be tuned over time.