I prefer the current registrar setup over anarchist blockchain hellscapes. You can sue Verisign, you can't sue 0xEf1c6E67703c7BD7107eed8303Fbe6EC2554BF6B.
I was recently quoted $250,000 to buy a .com that is not being used and as far as I can tell has never been used but some guy bought in the 90s when I was a young child. Aside from the level of extreme is that really different from "You could then probably buy readable domains for exorbitant amounts of cryptocoins."
You touch on what I think the core issue is with domains when you talk about suing Verisign. In some very limited cases, it's possible through litigation to get a domain you feel like you should have. But therein lies the crux of the whole issue: What standard needs to be met before you qualify for a domain name, especially one that is in contention?
In cryptocurrency, the criteria to receive the currency was originally solving problems with computing power because they needed some arbitrary measure. In the world of domain names, it's been a lot of different things but mostly just money and time.
That standard - what "qualifies" you to own a domain over someone else I think would have to be decided before truly effective regulation can be made or enforced, blockchain or not. Right now it's basically temporal, with some carveouts from lawsuits for trademarks.
The current regulations around how domain names work is woefully inadequate (like most regulation around technology) and gives outsized power to private entities. This is something that I think people don't talk about enough when they criticize blockchains.
If there was a better job being done with public policy, you'd likely have less people who think a system where there's no way of imposing real regulation is desirable.
It's especially bad with technology. My go to example is Apple's App Store, but this area is a good topic too.
To me, a happy mix of public/private policy would be best. The government allowing private TLDs, but regulating them. An anti squatting law, price capping new TLD name registrations, an enshrined "specificity system" to claiming an unused (or even used) domain would make the situation so much better, but it'll never happen.
Some TLDs have provisions against domain squatters and in case of copyright disputes, you can sue the squatters (or request ICANN to release the domain if that isn't possible for some reason). If you have a good reason to demand a domain currently unused, there are ways to get it.
With crypto-TLDs, nike.com and microsoft.com would be registered by a malicious player without any kind of recourse. The lack of enforceability makes the entire system pretty useless the moment it takes off.
The auction system is actually fairer since it gives enough time for every party to bid the highest amount according to their interest instead of relying on sniping services like in the current legacy systems.
> whereas a cryptographic system can do all that offline inside the data centers of the cryptosquatter.
what does this even mean?
>Some TLDs have provisions against domain squatters and in case of copyright disputes, you can sue the squatters
Sure, but this is also a double edged sword, since it also opens up abuse i.e. reverse domain hijacking which is a real and common problem. Just this week Charles Schwab tried to rob the legitimate owner of a domain that contained "Schwab", while they tried to hide the fact that the owner's name is also Schwab. [0]
>With crypto-TLDs, nike.com and microsoft.com would be registered by a malicious player
That's why good builders put in the time and effort to reserve these names from the start.
>The lack of enforceability makes the entire system pretty useless the moment it takes off.
Sovereignty has its price. No system is perfect, everything has its trade-offs, but I see a lot of people highlighting the best properties of the legacy system while ignoring the best properties of the new systems. The comparisons should be fair and compare the full range of positives and negatives instead of picking and comparing in a bad faith manner.
[0] https://domainnamewire.com/2023/01/24/charles-schwab-tries-t...
The on-chain contract could be written such that a buyer generates a token, supplied to an on-chain verification oracle that checks token.owned-domain.TLD exists, and if it does grants issuance of the new on-chain version.
This essentially allows for a fair transition from the old system to the new system, while eliminating squatting.
This system just seems to be traditional DNS with extra steps and a lack of manual recourse against bad actors, to be honest.
I suppose it could be configurable at purchase to allow it to be a recurring verification, if that was desired, but it would seem to overcomplicate for no reason. "Subscribing" to domains each year is legacy inefficiency.
I'm not sure what you mean about DNS, any chain-based system would be... chain-based, and therefore cryptographically-assured that a domain points where the owner chooses. "Not your keys, not your domain" applies.
Though ENS doesn't implement this, it'd be possible to add paid arbitration by humans for resolving trademark disputes. You couldn't get damages but you could get the domain.
