> -----Original Message-----
> On 25 Jan 2010, at 16:54, Milton L Mueller wrote:
> 
> > A correction: there is no vertical integration of existing TLDs.
> > Existing TLDs, even when they employ JM/CO, are not vertically 
> > integrated.
> 
> Though this is academically speaking true, what remains to be analyzed is
> whether the joint marketing/co-ownership behave functionally in pretty
> much the same way as vertical integration and thus have similar if not the
> same effect.

In fact, you are the one who has gotten caught up in words and definitions and seem to have lost sight of the real issues, which are the effects on competition and on consumers. 

Here is a basic, factual distinction for you to consider:

Assume a new gTLD .foo. With JM/CO a consumer can buy TLD .foo from ANY registrar, not just the cross-owned one. So if the JM/CO registry-registrar combination engages in bad practices or offers lousy service, consumers can switch. No market power. 

With vertical integration, a consumer cannot buy .TLD .foo from any other registrar.

Yes, JM/CO allows a registry and registrar to be more coordinated, but there is still open competition. We get the benefits of a vertical relationship - a better chance for the TLD to survive and market itself - without the lack of competition. 

To say therefore that JM/CO has to be placed in the same category as VI is obviously wrong. So tell us again why JM/CO should be linked to the broader issues of Vertical integration? 

Another point you seem to have lost sight of while getting confused about the term vertical integration: when we talk about JM/CO in the DAG, we are talking about NEW gTLDs. New gTLDs have no market power, and indeed have huge handicaps relative to incumbent TLDs. To say that their ability to enter the market should wait 3 years while we figure out whether to allow vertical integration and carve out a regulatory distinction between private and public TLDs is absurd, frankly. It just makes no sense.