[jimmysong]

This has been tried. It's called Namecoin and it failed to get much traction.

I suspect that using Bitcoin instead would improve adoption dramatically instead of creating yet another token. This can be done using sidechains or something similar.

[verdverm]

It is not the same as Namecoin.

Handshake at it's core is a decentralized root zone of trust, the equivalent of ICANN. It's up to people building on it to create the registrars and such.

Previous experiments were for one TLD like .bit or .eth

With Handshake, every possible TLD is released over 52 weeks

Even more so, it works with existing DNS infrastructure and tools. All it is doing is allowing for all TLDs and resolving them (with preference for ICANN / HNS) where the overlap occurs. It was built from the start to exist along side today's systems.

[michaelbuckbee]

Yeah, domain registration seemed like one of the few problems that potentially (lots and lots of handwaving around the details here) could actually have been a good fit for a blockchain solution.

- It's all public - It doesn't change that often - It is a data storage issue (as compared to processing) - On a per record basis it's pretty small (as compared to trying to store PNGs in the blockchain)

[troquerre]

One of the issues with Namecoin is the issuance. You can register any name for a set price. This leads to easy squatting behavior and allows whales to buy up all the good names. Handshake built in mechanisms to improve the issuance of names (detailed in danShumway's comment) which is critical when you're working with the issuance of scarce, non-fungible assets. It's like drawing a comparison to bitcoin and bit gold. Both had the same idea, but the specifics matter.

[sdtsui]

Hey Jimmy - I hope the bitcoiners that eventually read this thread can get both sides, as many project teams have received some version of the "do it on bitcoin with sidechains/lightning/liquid/rsk" refrain. I think in practice it's more complicated than that.

First - I tend to think an idea with clear similarities being tried before and failing is default a non-negative (sometimes good) sign: shows us what didn't work, and deepens everyone's understanding/validation of the problem space.

I would also say that it's more similar to Bitcoin than most people would think. Looking at the software, I think it's pretty cool that handshake's first implementation is a fork of a bitcoin implementation that has seen production usage for years (bcoin), one important change is the addition of opcodes to support covenants, so that blinded vickery name auctions can be done fully on-chain.

Basically, Handshake can build on top of the security and reliability that Bitcoin-like systems have validated for us over the last few years: namely, UTXO systems, proof of work, user knowledge/habits around self-custody.

I think Bitcoiners could consider looking at Handshake from this perspective: there is a lot of core technology, architecture decisions, and spirit that is similar. Perhaps if you believe in the potential of public blockchains like many early Bitcoiners do, the solution might not be things like sidechains (which are mentioned as a potential scaling option in the design notes), but extending the core technology in a minimal, single-purpose way (to support auctions)?

Perhaps it is possible that Handshake goes down as one of the best examples of an ambitious new project actually "using Bitcoin" (albeit in a way most Bitcoiners would not immediately agree with) instead of creating "yet another token"?

----

p.s. I've been a fan of yours for years and your content was some of the first I ran into when I was entering crypto full-time. I guess at the end of the day I'm simply curious what you thought of the other projects so far (including Namecoin and ENS) - do you own any names on other decentralized naming systems?

What do you think prevented Namecoin from getting sufficient traction from your perspective?

[rglullis]

ENS is running on top of ethereum just fine

[tynes]

I think ENS is queried more often in a different context than DNS, since many people access it via JSON RPC over HTTP. The interface is an Ethereum contract, although records could be cached in a DNS server if the server can hit an Ethereum node upstream. The two projects feel pretty different for that reason. Handshake could be accessed over DoH potentially.

[pinhead26]

There is currently no such thing as a decentralized side chain. No one wants all this data on Bitcoin.

[tynes]

Handshake is very different than Namecoin. Its possible to come to this conclusion by doing any amount of research. Please consider thinking before speaking in tribalism. There is no good trustless sidechain technology. Sovereign names can exist in parallel to sovereign money and both can benefit from each other.

Also HNS is a coin, not a token.