Interesting move but not sure about the benefits

I'm not sure about the benefits, either. Maybe they won't stick, and that will make the land grab irrelevant. But then again, we don't know what new TLDs will do to search or discovery of sites. Imagine the SEO on kippt.app.

Most likely they will just work as super nice urls for regular pages

Consider for a moment how many URI's you know off the top of your head that aren't in Alexa's top 100. The goal of DNS is to allow humans to access web resources without search.

gTLDs will drive down search traffic for websites (I won't have to use Google to find redshirts.book), but most discovery today is aimed at individual pages, for which a URL shortener is still the best option.

It's also worth noting that especially with IPv6, gTLDs will provide much more convenient device specific URLs (chrisnicholson.iphone, for example). You'd have to be a math major to memorize IPv6 addresses.

I think that's a good read, Chris, but it requires that the good generic TLDs are really available to register. If Amazon (or Google, if they strike a good deal) will *let* the author/publisher register redshirts.book, it will be an elegant shortcut for the brain. But if they lock up good domains as silos for owned content, that will mislead people.

Which is why "generic" gTLDs (.book, .app, .movie) will be adjudicated last (greatest number of claims). Redshirts.amazon will work almost as well (how many authors' websites have you visited in the past year?) Perhaps you know the answer to this: can ICANN reject every gTLD application for the reason you mentioned (essentially, nobody gets .book because its too generic?)