I just had an interesting conversation with Kathryn Kleiman, who's involved in technology law and co- founded the "Domain Name Rights Coalition" in additional to a bunch of other cool stuff.
The main area we don't see eye-to-eye on is the need for new top-level domains. As I understand her argument, we need to break the ".com" monopoly up and give everybody a chance to have a "short" name. In the current system, "mcdonalds.com" goes to the company with the golden arches and everybody else loses out.
I understand the concern, I just don't see how TLDs solve the problem. Suppose I get "mcdonalds.consulting" in a brand new TLD. Even if McDonalds doesn't successfully sue me to take it away, in what sense am I better off than registering "mcdonalds-consulting.com"? If my web browser tries ".com" by default, then the difference is just a period instead of a dash.
Introducing new TLDs is not going to deprecate the worth of ".com" as long as it is seen as the default. The ugly truth, as I see it, is that there really is only one namespace. Somebody is going to win and get the preferred name as long as it is available; chopping up the bits differently just rearranges where the money goes. All we can really do is ensure there are still (less-desirable) alternatives for everybody, and we don't need TLDs to do so.
(One idea might be to auction off names every couple of years--- no automatic renewals. But that needs more thought.)
It's also worth pointing out that even with 1000 TLDs, the cost of McDonald's registering in all of them is miniscule. The only solution seems to be setting up yet more rules for what you're allowed to register in each TLD, but I don't see that as a winning solution.