RE: [gnso-vi-feb10] Question to WG on RAA

[Milton L Mueller <mueller@xxxxxxx>]

> -----Original Message-----
> a policy of not expanding VGRS's registry market share 

Such a policy approach makes no sense in this WG. Structural separation policy 
per se will have little effect on VerSign's market share in the near term. The 
best it can do is to facilitate market entry by new actors with ideas and 
services strong enough to gradually grow in relation to .com and .net. You do 
not make new competitors' ideas and services any better by restricting 
VeriSign's ability to offer new TLDs.

As you noted, the best and most efficient way to reduce its market share is by 
divesting VRSN of established TLDs. If you want to divest .net, fine, but such 
a process would have absolutely nothing to do with the remit of this WG. So 
let's drop it. 

As I have already noted, the other major competition policy issue is mergers 
and acquisitions. I.e., should VeriSign should be prevented from acquiring any 
and all new competitors as they succeed? Probably so - but again this has 
nothing to do with this WG. Do I think VeriSign should be prevented from 
offering new TLDs or services? No. The former restriction could be construed as 
pro-competitive; the latter restriction would be anti-competitive.

You do not do consumers any good by placing artificial restrictions on an 
existing firm's ability to compete in new domains. Or if you think it does help 
end users, please explain how. We know how it makes life easier for the 
protected suppliers. 

> as long as competition policy exists
> as the controlling issue, market share remains relevant.

An understanding of competition policy grounded in economics would suggest that 
market share is relevant to acquisitions policy and to divestiture policy, but 
not to the issue of structural separation and the encouragement of new, viable 
market entry, which is what we are concerned with here. 

> From a "pure consumer perspective" the regime in place in 1985 worked
> pretty good. Domain names were free. 

And there were only 200 of them. Frankly, I would like to see an ICANN new TLD 
program that made it possible for some 21st century Jon Postel to run a TLD and 
give out domains for free. ICANN makes that impossible now. Structural 
separation requirements are part of that problem. One reason I am interested in 
cross-ownership and VI is that it might make it a little easier for something 
like that to happen. 

> Do you really think we can reason
> from where we are, to some place realizable, illuminated only by the
> light of "pure consumer perspective"?

Yep.