Additionally, I always found it interesting how the .xxx TLD never took off (was essentially DOA), even after they were so popular and valueable in the pre-sale auctions icann and others conducted b4 their introduction/release several years ago.
One of the original TLDs is .arpa, and I don't see any of that uniformity in a world where second-level domains like .co.uk domains are pervasive.
> Additionally, I always found it interesting how the .xxx TLD never took off (was essentially DOA)
The usecase driving .xxx was not from site operators but from third-parties seeking an easier way to censor porn sites.
https://en.wikipedia.org/wiki/.xxx
The concept was flawed from the start given the same site can be served through multiple domains, and it relied on site operators located in any jurisdiction from any corner of the world to voluntarily serve their sites thorough that TLD.
The longest vanity TLD I could find was .travelersinsurance. Two dozen characters are nothing more than a rounding error in the 2k character limit. You routinely pass 10 times more characters in path and query parameters than you do in the domain name.
Also, how often do people type in URLs instead of clicking links?
Also, keep in mind that some browsers are starting to hide path and query parameters in the address bar and instead only show the domain name. I presume that makes domain names and TLDs more relevant in terms of name recognition.
> The Google Chrome browser supports a maximum length of a web page URL of 2 MB (2048 characters) in size.
taken from [0]. Do note that this is a restriction created by the browser developers. In theory a URL can be of arbitrary length. Some browsers (like Firefox) do comply with this, though they might now show the full URL to the user, cutting it off after a certain number of characters.
[0]: https://mywebshosting.com/what-is-the-maximum-url-length-lim...
For example, the following domain renders just fine, despite being ~2076 characters long: https://postman-echo.com/get?foo1=aaaaaaaaaaaaaaaaaaaaaaaaaa...
See the upstream chromium docs: https://chromium.googlesource.com/chromium/src/+/refs/heads/...
I don't think adding ~10 or so characters to the TLD meaningfully impacts this limit.
Heck, let Big Tech take .google/.apple/.amazon/.comcast etc if that's what it takes to get this done. Point being to keep the root pollution to a finite amount rather than the ongoing sell off of any name imaginable.
And honestly we need better mechanics for machine-intended references regardless. It's ridiculous that you resolve and load a webpage, only for that webpage to require you to resolve a bunch more human readable (ie non-decentralized) names for loading subresources. For example, going to a bookmarked page shouldn't result in any DNS queries for human readable names.
And why should that distinction be made? "New gTLDs" have been active since 2013. That's nearly 10 years of usage you're still arguing should be invalidated.
To be clear I'm not picking on things like .biz etc. Even though those were obvious cash grabs, they're at least widely applicable and thus widely adopted. But things like .christmastrees .business .companyname .morebiglongwords etc. Over "nearly 10 years of usage" I can still count on my fingers the number of times I've seen things like this used.
Granted, the domain name system is not ideal, but adding new TLDs, however long, is a better solution than dealing with having to name your company so you can get a .com that matches it.
I doubt this assertion is grounded in reality. Sites that resort to puns from cc TLDs like lobste.rs don't suffer from name recognition problems, and no one ever had any problem remembering twitch.tv.