7 ms·

Well, nothing, but it does assume that Verisign doesn't have an unstated plan to do something else with .name. Domain registration is the real estate of the Int

by bityard 13d ago

Well, nothing, but it does assume that Verisign doesn't have an unstated plan to do something else with .name. Domain registration is the real estate of the Internet and very little has happened in the last 30 years of domain name policy that has been for the benefit of anyone outside the the registrars. For all we know, this is simply the first step in a series of moves for Verisign to better monetize the .name TLD in some novel fashion. My evidence for this is that Verisign's own arguments for terminating the third-level domains are highly dubious. They claim that the third-level domains are too hard for them to manage. Bollocks: there are only 22,000 of them in use. That is a VERY small database that practically fits on a calculator and I refuse to believe that any manual labor around it is an outsized burden compared to pretty much any other semi-popular TLD. The second claim is that "the majority of those are not in use." Okay, if that's true, then where is the management burden coming from? That means tens of thousands of people are giving them money and getting nothing in return, isn't that the definition of an ideal business model? It just doesn't pass the sniff test. And finally, having read both of the linked documents, it sounds like people who have registered their .name for years in the future are not getting their money back. Are they likely to pay a second time, to a sub-registrar with no history?

xamde 13d ago
Also, someone managed to run all this 20 years ago. Has technology gotten so much worse? Is RAM now THAT expensive?
- toast0 13d ago
  Not just someone, Verisign managed to run this registry since its delegation. Restrict future 3rd level registrations, offer a path to upgrade a 3rd level registration to a 2nd level registration for those 2nd level domains with a single registrant and the burden will decrease over time.
troyrollo 9d ago
There is a problem with registrars dealing with the third level domain because it requires a different UI at the very least and many of them just decline to do it. It has become increasingly hard over the years to find one that will do it. Verisign could create a work-around by defining a format at the second level - such as ml--2-4-johnsmith.name in which the first number is the number of levels, then there are n - 1 numbers to identify the number of characters at this level, so in this case it would map to john.smith.name at the registry level. There are only about 5 or 6 existing second level domains that are too long for that to work and they could be remapped as, say, ml--id-1 through to ml--id-6. Since the double dash at that position is reserved for registries (the registry should reject a registration with double dash at that position outside of its own defined uses and xn-- for punycode domains) , this would allow Verisign to facilitate those domains being managed (or even created afresh as long as they are not too long) as if they were second level domains while continuing the third level domain service, avoiding the registrar implementation problem. They won't do it though because Verisign has never been known for the customer-focus or corporate responsibility.

by bityard 13d ago

xamde 13d ago

Also, someone managed to run all this 20 years ago. Has technology gotten so much worse? Is RAM now THAT expensive?

toast0 13d ago
Not just someone, Verisign managed to run this registry since its delegation. Restrict future 3rd level registrations, offer a path to upgrade a 3rd level registration to a 2nd level registration for those 2nd level domains with a single registrant and the burden will decrease over time.

troyrollo 9d ago

There is a problem with registrars dealing with the third level domain because it requires a different UI at the very least and many of them just decline to do it. It has become increasingly hard over the years to find one that will do it. Verisign could create a work-around by defining a format at the second level - such as ml--2-4-johnsmith.name in which the first number is the number of levels, then there are n - 1 numbers to identify the number of characters at this level, so in this case it would map to john.smith.name at the registry level. There are only about 5 or 6 existing second level domains that are too long for that to work and they could be remapped as, say, ml--id-1 through to ml--id-6. Since the double dash at that position is reserved for registries (the registry should reject a registration with double dash at that position outside of its own defined uses and xn-- for punycode domains) , this would allow Verisign to facilitate those domains being managed (or even created afresh as long as they are not too long) as if they were second level domains while continuing the third level domain service, avoiding the registrar implementation problem. They won't do it though because Verisign has never been known for the customer-focus or corporate responsibility.