Thoughts on name squatting

Play stupid games, win stupid prizes

Context: the Farcasterverse is a flutter because a namesquatter was kicked from squatting on the '/bankless' channel name, and was caught trying to sell the name, all while channel names are a beta feature (not decentralized), thus subject to revocation upon bad behavior. Squatting is dubious, but this is a dumb move to pull when like fnames, they are not onchain yet.

Domain name squatting is perhaps the internet's 2nd oldest profession (next to the 1st, which is the same one IRL).

Domain names are property. Digital property...and the idea of digital property rights are contentious and exactly the kind of thing Web3 writ large is trying to solve for. But like the real world, there's a perpetual debate on whether property rights are rooted in the libertarian 'homesteading' principle or a result of pure social consensus. Often there's no right answer here, but a space of tradeoffs.

You've all probably heard of Zooko's triangle. Yes, we love our trilemmas in the world of distributed systems (everything is really isomorphisms of the CAP Theorem TBH...). I'll just tip my hat here and acknowledge others have done a lot of noodling on this topic before me, rather than repeat boilerplate acknowledgments. That being said, dealing with tradeoffs is where the rubber really hits the road and I've rarely found focusing on trilemmas to be useful, except as a device to pillory moonboys who have claimed to 'broken' said trilemma.

Consider Namecoin, perhaps the only non-Bitcoin blockchain to be endorsed by none other than Satoshi Nakamoto himself. Namecoin matches Bitcoin's proof-of-work aesthetic, requiring miners to expend energy, find a low hash value & produce a block in order to secure a name. Apparently, Namecoin has since pivoted to being 'merge mined' with Bitcoin. Merge mining is basically 'vibe mining' and the merge mined change doesn't inherit any of the double spend protections of the main chain (see Section 3.3 from the Eigenlayer whitepaper, if you're curious):

Proof-of-work, on the surface, seems to align with the libertarian homesteading principle. In a nutshell, the idea of homesteading is that one who 'mixes their sweat and labor' with a natural resource has more of a claim to that natural resource than anyone else. This applies to everything in the libertarian worldview: from ranch land, to radio frequency spectrum, to water rights, to aircraft noise abatement. Well enough, but of course, this doesn't really solve the problem, since we are just shifting the question of ownership to: how much 'sweat' do you need to mix in order to secure real property? And what claims can you make against the state or other individuals when you own property? Often 'the crown' was the ultimate guarantor who would attest to the fact that you wrote down the boundaries of your ranch. In Alta California, receiving a land grant (diseño) was done by submitting a rough handdrawn survey to the governor often measured by laying down a 50-foot cord on horseback (la reata or lariat in English). Of course, you also had to be Catholic and on the good side of the governor (sorry Native Americans!) Not a lot of sweat equity to secure acres of prime real estate.

However, domain names are not real property. They are intellectual property: you cannot hold them, they are pure bits...just information. Holding a domain name boils down to a social claim more than a real property claim. As such, social property is only as strong as the governance behind it. Is ENS backed by Proof-of-Stake? Yes, but Ethereum itself ultimately boils down to the social layer, choosing to run the Beacon Chain node software and not Ethereum Classic, for example.

Social property is not as simple as 'finder's keepers' from now to infinity, there has to be "proof of social work" that backs legitimate claims to a name in the here and now, and an agreement to maintain that claim over time, otherwise, the claim will be reallocated. Both ENS & Namecoin require ongoing registration fees. But money is perhaps not what we mean by social proof, in fact, domain name squatters always hide behind the "but I paid for this!" defense.

We have to solve the hard problem: governance. Consider that Reddit has had this issue with subreddit names being secured by bad-faith trolls for decades, with nil solutions in sight. I'm hopeful that decentralized judicial systems like Kleros could be used to help with moderation disputes on social platforms like Farcaster. But another method for determining social proof could simply be by using PageRank. Let there be multiple 'bankless' channels, but the one with the greatest followership (perhaps augmented by anti-sybil mechanisms), will be the one to secure the canonical '/bankless' on a client like Warpcast.

Loading...
highlight
Collect this post to permanently own it.
DFern Longform logo
Subscribe to DFern Longform and never miss a post.