It is a good precedent for moving to a decentralized naming system.

Maybe for sketchy stuff, but mainstream sites that stay out of legal trouble will prefer the regulated internet.

Until the legal troubles are brought to innocent mainstream sites by those who prefer to regulate the internet.

How does a decentralized naming system work when names must both be unique and user-chosen?

While I agree that our current root DNS system is a legacy artifact of the early days of the Internet, you're totally welcome to build your own ISP and not use the root nameservers. Whether people will want to interconnect with you is another question.

> you’re totally welcome to build your own ISP and not use the root nameservers

That’s a drastic overstatement. I believe what AJ007 is referring to is OpenNIC, which is a DNS service that provides custom and free domain registration using custom TLDs like .geek and .chan. It forwards existing TLDs to the root nameservers so that it’s interoperable with the current DNS system. It has its pros and cons- I used it for a few months without any major issues. It certainly doesn’t require building your own ISP to sidestep the current root nameservers.

OpenNic: https://www.opennic.org/

Medium sized AS backbone network operations speaking: nobody takes this seriously, and not one of more than a hundred individual ISPs we peer with uses opennic. It's dismissed as a bunch of dilettantes by people I know who run DNS services relied upon by millions of customers.

That’s fine and everyone is entitled to their opinions. But my point is that it’s an alternative root nameserver that anyone can choose to use without the need of starting their own ISP or breaking the current system. The technology exists and it’s been up and running for a while now. The seriousness of it all and whether or not important people dismiss the technology is a moot point.

What's their issue with it?

As I understand it, their issue is that it's not relevant for their customer base (nobody has requested access to the weird non ICANN top level domains), and it risks balkanizing the established single global Internet into things that can't talk to each other.

Essentially the same problem as if people started picking random non-RFC1918 /8 IP space ranges and numbering their networks into them, or choosing their own arbitrary AS numbers.

Do German courts even use precedent? I thought no?

The German courts still try to be internally consistent, they just don't feel as held to that standard as common law systems do. Precedent can still make or break a case under the right circumstances.

Interestingly, that's basically the same decision the English House of Lords (now Supreme Court) came to in 1966. While they tend to respect precedent, they aren't bound to it (though only in that Court):

> too rigid adherence to precedent may lead to injustice in a particular case and also unduly restrict the proper development of the law. [The Lords] propose therefore, to modify their present practice and, while treating former decisions of this house as normally binding, to depart from a previous decision when it appears right to do so.

https://en.wikipedia.org/wiki/Practice_Statement

Interesting, I didn't know. It seems to be more subtle though:

> German judges are not bound by precedent (Präjudizien), but, in pursuit of consistency, they are likely to consider previous decisions of the higher court(s).

https://libguides.bodleian.ox.ac.uk/c.php?g=422919&p=2888143