Well that's a shame - I wonder if, given moxie is a member of the HN community, he'd clarify what is meant by that. I had a quick search and didnt find anything.
Given how explicit he was about this at the time (and the lack of comments to the contrary in the past 5 years), I honestly don't think any clarification is necessary. That's his opinion and while I may disagree with his opinion, we shouldn't pretend that he is likely to change it in the future (or to imply that isn't his opinion on the topic). If you feel strongly that you want decentralised instant messaging, use Matrix (or something like it) rather than hoping Signal will become something Moxie has said he is against it becoming.
[1]: https://github.com/LibreSignal/LibreSignal/issues/37#issueco... [2]: https://github.com/LibreSignal/LibreSignal/issues/37#issueco...
In short, his argument was largely chat ecosystems are constantly evolving but decentralized protocols tend to ossify. Thus, decentralized protocols will ultimately not be able to keep up with the demands of the users.
https://www.youtube.com/watch?v=Nj3YFprqAr8
Lots of people point to email as an example. I'd like to also point out the phone system and its codecs as another example. Most phone calls behind the scenes are using SIP to handle connectivity. SIP is codec agnostic, you could technically use any kind of codec in the RTSP stream. Its a very decentralized protocol, and highly extensible. And yet it seems like almost every time you try and place a call, the only codec anyone else supports is G.711, a codec released in 1972 and hasn't had a major update since 2000. We're slowly seeing G.722 adoption now that the patents have run out (HD voice on cell phones) but even then this is spotty. Meanwhile every other chat platform out there has moved on to more efficient and higher quality codecs because all it took was the chat platform owner to push one version update out.
The decentralized communications network is only just now finally adopting and rolling out a _33 year old_ codec. Sure sounds like a win for decentralized systems!
go to matrix.org > click "Try Now" > get redirected to element.io > click "Open in your browser" > allow it to use IndexedDB > create an account > start messaging
I have tiny grievances with every step of this process: matrix.org is filled with too much jargon; "try now" sounds like some kind of trial; redirections to totally different websites are scary; the relationship between Matrix and Element isn't clear; the stock photo-ish background on app.element.io is a bit tacky; asking for browser permissions on page load is rude; the account creation page makes the "edit homeserver" button fairly prominent but doesn't tell us what it does in plain language; I should have been told that you can invite people by email before I clicked "send a direct message"; adding an email to invite asynchronously pulls up a "terms of service" dialog.There's a lot to love about Matrix, so I hope it becomes easier for people to get started with it.
The UX still isn't "great" or "delightful" but I have had people sign up and find be by email or matrix ID with no problem.
Signal is far from an open standard. I use it and am quite happy using it, but with the knowledge that Moxie has publicly said many times that he does not believe in federation or third party clients (and will implement client blocking as well as using trademark suits if the problem gets big enough). And it's fine to have that opinion, I just disagree with it and you shouldn't sell Signal as something the lead developer has said they're explicitly against.
For folks saying (read: hoping) that Signal may become federated in the future -- I think it's a much safer bet to just use Matrix since it's already federated and has at least equivalent privacy guarantees (or even larger if you self-host your own server) and hope that Signal will peer with them in the future (which I personally think is about as likely as Signal becoming federated itself -- namely, it's very unlikely either way). But you do you.