Absent somebody enacting a worldwide law (and enforcing it) there'sKarl:
nothing that can stop the creation, deployment, and use of distinct and uncoordinated root systems.
The essence of IAB's argument is summed up in this sentence: "it is not technically meaningful for there to be more than one root in the public DNS system."
Let's examine the rather interesting semantics of this statement.
Is IAB saying that alternative roots don't exist? No, that would be factually wrong. They do exist.
Is IAB saying that alternative roots cannot exist, that they are impossible? No, for if there were no possibility of such roots, IAB never would have issued the statement. One could refuse to acknowledge that alternative roots are properly called "roots," as Crocker seems to be doing, but that's just a semantic game, and unworthy of the anyone's time.
So what is the IAB saying? In a literal sense, it concludes only that "There is no getting away from the unique root of the public DNS." But, as noted before, people can and do get away from it, and if the IAB wasn't worried about that it's hard to understand why they issued this statement.
As best as I can tell, the real message is both tacit and non-technical: the IAB is trying to tell us that it is a *bad public policy* to for there to be more than one root in the public DNS system, because it would lead to "a very strong possibility that users of different ISPs who click on the same link on a web page could end up at different destinations, against the will of the web page designers." Most of us would agree wholeheartedly that fragmented connectivity is a bad thing.
The problem with this foray into policy advice is that while the IAB's opinions on the technical consequences of root administration are welcome, they are not experienced or accomplished public policy analysts. A technically informed policy position on this matter would involve much more than a statement that multiple roots might fragment Net connectivity. It would require an analysis of why multiple roots might form (economic incentives for creating them), it would propose and consider various policy responses to the problem, and it would weigh the benefits -- not just the costs -- of the various options. Such an analysis would also have to propose specific means of enforcing and implementing the desired policy. The IAB statement doesn't do any of this.
Just as a deliberately simple example, Here are two possible policy responses to multiple roots, lying at opposite poles of a spectrum of responses:
a) alternative DNS roots are made illegal and the FBI raids and
shut down anyone who runs one.
b) ICANN agrees to coordinate its root with all existing "rogue"
root operators.
Either policy is a *technically* viable method of maintaining a coordinated and unique root. As public policies, they would have radically different consequences, obviously.
To drive home this point, consider this little modification of the IAB statement:
"Put simply, allowing multiple public [browser softwares] would raise a very strong possibility that users of different [browsers] who click on the same link on a web page could end up [with different results, or no results at all] against the will of the web page designers."
This statement is true. Just as true as the one made about the root in the IAB Comment. Why did the IAB issue a statement about the one and not the other? If the IAB did issue such a "technical comment," would it be advocating that Netscape, Microsoft and other browser software developers be regulated?
My conclusion: if the IAB wants to participate in a policy debate on competing roots, it is welcome to do so, but I would encourage it to do so in a professional and comprehensive manner. If it wants to restrict itself to purely technical analysis of the consequences of various actions, that's fine too but it should be careful to refrain from normative judgements and to be more careful to delineate where "technical comment" ends and law, regulation and policy decisions begin.
Milton Mueller