I ventured over to 1&1 to see about those gTLD preregistrations. I spent about an hour seeing what I could find that made sense (example: car.insurance) and can tell you that 99% of the stuff I checked on was already preregistered. I would venture that the majority of these are pregistered by domainers so even though you hear that impressive 2 million plus preregistrations I think if we could see the underlying data the majority would be preregistered to a nominal amount of people (i.e. domainers).
Another thing to consider with the preregistration numbers is that there is no cost to preregister, so you have nothing to lose by pregistering a domain, or thousands perhaps. I preregistered about 10, whether I’d actually buy them is another story, but I have that cozy feeling that I’m first in the 1&1 line in the event I want to purchase them. While 2M+ preregistrations sounds impressive, how many sales will that translate to? Averaged across how many new gTLD extensions? My call is that this data tells me that at least half of the extensions will die and shrivel up pretty quickly, which I think anyone can call when you see all the ridiculous extensions out there.
I opened an article today that is talking about the possible extinction of Black Friday (http://business.time.com/2013/09/26/black-friday-is-facing-extinction/?hpt=hp_t3). I guess that .blackfriday extension looks even worse then I initially thought it did.