What's the path to actually make these domains usable? Does a browser just have to start supporting Handshake?
Too many blockchain pitches say things like it "would be great" if everyone used it. Yeah, it'd be great if we all moved to a decentralized everything, but users don't care enough.
Could a browser like Brave or Firefox decide to start supporting this? Then some websites would only work on Brave/Firefox, eventually forcing Chrome and Safari to support it too.
You can browse Handshake sites with Firefox by using the NextDNS [1] resolver (it's an option in Firefox's Settings). You can also use hnsd [2] or hsd [3] as your resolver.
There are a lot of sites and you can see who has claimed their names and/or put up resolvable names on DNS Live [4].
The web navigator (browser) has had "registerProtocolHandler"[1] api, which has been around since at least 2008[2]. The history has been a little complicated, but it lets one write a page that can handle addresses of this sort. That could be a handshake relay page, that loads the handshake content, then renders a page with it.
There is also similar work in WebExtensions to allow this to happen in the extension layer. Here's[3] a list of p2p-related handlers which were whitelisted (similar was done in Chrome), but "web+handshake" would work without whitelisting. These extensions would obviously be more useful if any mobile browsers had extension support. Firefox seems to have just shut the door on most extensions![4] What the hack?!
That list of Open Source "Internet Good Guys" is a list of groups that Handshake decided to sponsor, not groups that endorse or have anything to do with Handshake.
The actual sponsors of Handshake include the dogpile of VCs and financiers that you'd expect would gamble on a blockchain project, confusingly (deliberately?) listed below the definitely-not-sponsors list.
Look, it's great that they want to support the open source community that they and everyone else relies on, but it looks to me like they're trying to trick careless readers into thinking that these well regarded open source / community organizations support the Handshake project. Even with the big disclaimer, that list is the biggest part of the page. It almost seems like they're trying to look good by association, when there isn't even any association... they just wrote a check using VC funds.
To me, this sends a strong negative signal. If anything, it makes them seem more overhyped, not less.
Eh, I think you just read it wrong. This doesn't seem confusing nor deliberate. Right above the list of the sponsored open source projects, it says in bold:
"""
The inclusion of the pledge recipients on this page does not constitute or imply any endorsement of Handshake on the part of recipients but simply reflects gratitude for the grant recipients' contributions to FOSS.
Some of the Current Pledge Recipients (Net 10.2MM USD + coin grants to some recipients)
"""
This is the biggest issue I have with Handshake. I have a couple Handshake TLDs, but I can't really use them at all. What Handshake is trying to achieve is in essence similar to trying to establish a new set of root servers. And just like that idea, it will never realistically happen.
Infrastructure changes on this scale are basically impossible, as evidenced by the IPv4 to IPv6 transition. The only reason that even has a chance of succeeding is economic incentives caused by rising IPv4 prices, however, Handshake does not have that type of incentive.
In my opinion, Handshake will lead to positive changes in our current DNS system. However, I don't see Handshake ever replacing our current system.
Sure I can use it for some things, but no one else can. I can't even get an SSL cert for them, or control DNS using an API like I can with Cloudflare (I can setup my own BIND server, but its a lot more work for not much gain).
I can use a gateway like HNS.to to share it with other folks, but thats uglier than a normal domain and bad practice due to shared cookies and the like.
> What Handshake is trying to achieve is in essence similar to trying to establish a new set of root servers. And just like that idea, it will never realistically happen.
Yes, it will need to find a foothold in a parallel domain (no pun intended). IPC is one example. Combine something like in-process REST with the ability for any app to trivially bind to a name, and we'd be getting somewhere interesting.
Skynet leverages HNS names to shorten its skylink hashes. So path to usage here would be in the form of a shortened URL. IE: https://[traditional TLD]/hns/[HNS domain]
Too many blockchain pitches say things like it "would be great" if everyone used it. Yeah, it'd be great if we all moved to a decentralized everything, but users don't care enough.
Could a browser like Brave or Firefox decide to start supporting this? Then some websites would only work on Brave/Firefox, eventually forcing Chrome and Safari to support it too.