> In my opinion, a "just okay" .com is better than a good domain with some other TLD. I suspect this is a pretty common opinion.
I agree with that, but finding even a "just okay" .com is getting pretty damn hard. I use .com for most of my projects, but I've also got some .io's and was looking at .ai (which looks like a massively missed opportunity right now - you need to pay $100 just to register to be able to buy the domain, and then $100 per domain; if they made it a bit easier/cheapr, it'd seem the perfect time to profit from lots of deep learning / AI tech projects...)
> I thought about this when I wrote my comment, but as an American it's not something I'm especially familiar with. Our own ccTLD .us isn't commonly used.
Yeah, .us had a very brief period of limited popularity, but was effectively stifled by the silly policy of geographic subdivision.
> I think the only reason these were ever popular is because they were priced so cheaply. It seems like .info is/was mostly used for throwaway domains, and .biz is just a terrible name. It's supposed to be for businesses, but it sounds like it'd only be used by shady used car dealerships. Nevermind that .com is officially for commercial entities anyways.
Yeah, it really got off on a bad start by effectively feeling like cheap knockoffs of .com, to the extent that at some point it was a clear signal that you were probably a small two-bit operation or a scammer if you had the .info/.biz and someone else had the corresponding .com. And if you had a name that somebody didn't have on .com you'd be an idiot not to buy the .com anyway. This is very different from e.g. the ccTLDs, where someone might very well choose to have the ".co.uk" (you can get ".uk" too now, but they're not common) and not bother with the ".com" if they're genuinely geographically limited to the UK (personally I'd snap up the .com anyway, and redirect it, just in case, but a lot of businesses don't).
> I think .io is a fad that's going to die along with infinite scrolling websites.
Maybe, maybe not. I think .io makes a lot of sense as a "easy way to get a decent name" for projects that are not necessarily intended to be directed at the general public. If you're aiming for a tech savvy audience, the .com isn't so important. So <some random project on github>.io might very well persist. I agree that most who use .io for public-facing projects will probably end up moving to .com's as/when they find/can afford to buy suitable alternatives, though.
> Side note: would .tv still be big if Netflix used netflix.tv instead of netflix.com?
That's an interesting question. I think a lot of these domains come and go largely based on how good the registry's are at marketing them. Part of the challenge is that the individual TLDs for the most part are not huge money-spinners, so don't have the budgets to try to turn it into valuable brands.
For .name one of the challenges we faced was exactly that: We were contracted to sell at a relatively low price (the market wouldn't have been prepared to bear a higher price anyway for .name at that point, but for others, possibly), and we were dependent on registrars that in most cases had no clue, and the ones that had a clue had little incentive to invest hugely in it because to them we were just one of a bunch of other TLDs, and we couldn't give them anything extra of value (like exclusivity) compared to the other registries.
So unless you make it big, or can show projections that you will, the revenue potential drastically limits the amounts you can spend building a brand around the TLD. Which again limits the growth. And around in circles we go.
The painful irony is that we'd probably have done far better if we stuck to our original business (vanity hosted e-mail) - the e-mail service was bought by a company backed by Marc Cuban; I don't know if I can mention the sales price but it wasn't huge (we had ~2 million registered accounts; the sale happened after the .com crash) but as far I know they made the purchase sum back in 6 months by converting the accounts to paid accounts and ditching any user that wouldn't pay (a shame, but it was that or closing the service down entirely).
Our original business plan was actually to stick with paid accounts like they did, and we identified them early on as our only realistic competitor, but then we got seduced by the potential of building up a huge free user base and sell, with the enthusiastic agreement of the VC's we spoke to (this was at a time where there had been crazily ludicrous transactions selling free users for amounts accounting to decades of actual revenue per user, in the expectation that you'd be able to sell to someone else who would find magical ways of monetising them better).
Based on the .name registrations in early years I think it's safe to say that it took many years to get to the same revenue numbers as they did basically over-night. But we gambled on registration numbers for .name several magnitudes higher than that, and that's the price you pay sometimes.