delivery receipts => XEP-0184: Message Delivery Receipts
    optional read receipts => same as above
    user presence information => XMPP RFC
    binary serialization for efficiency and extensibility => XEP-0231: Bits of Binary
    end-to-end encryption => XEP-0373: OpenPGP for XMPP, XEP-0384: OMEMO Encryption, XEP-0378: OTR Discovery
    WebRTC signalling for negotiating VOIP and video chat => XEP-0343: Signaling WebRTC datachannels in Jingle
    signed introduction tokens to reduce spam => ?
    a standard extension mechanism => https://xmpp.org/extensions/

Doing decentralized social media, on XMPP, is possible. Take XEP-0060: Publish-Subscribe, add Atom 1.0 (yes it's the power of XML, you can put a standard in another) and boom you have a social network with feeds, comments, subscriptions and everything, fully ready.

I'm doing that for years with Movim https://movim.eu/. The several major XMPP servers are handling that perfectly as well. How many Matrix servers are out there ? One, that is still in beta (Synapse).

XMPP is standard (IETF wise), is already massively deployed, is used by universities, governments, companies…, is exensible, is stable and is maintained by a big and motivated community.

Don't reinvent the wheel once more, just implement the standards.

(1) The address book.

(2) The easiest possible signup and setup.

More generally for #1, whatever method(s) you use to find people's contact information and connect with them.

Without that, you could have created the perfect chat experience in every way, but people still won't use it. Because their friends aren't on it and/or they can't find their friends.

A ton of chat apps fail because engineers sit around talking about how to architect things, when the truth is that once you've gotten past the almost-insurmountable hurdle of getting two users onto the system and starting a conversation, the actual mechanism for actually sending messages back and forth is not much more than an implementation detail.

Not that those things aren't important, but they are like 10% of the formula for success, and the other 90% is getting to the point where they can start a chat session.

People's primary need is to communicate. With the people they want or need to communicate with. And via the contact information they have available. (Exchanging contact info for a new service offline is a huge hurdle.) A better experience is better, but at the end of the day, these other concerns will nearly always win out. So you don't get very far trying to use a superior experience as leverage to get people using it. You need to attack that problem more directly.

I used to call it ‘email for mass communication’ but it confused people, since it’s not based on email, so I stopped calling it that. That’s the goal though.

I also gave a talk on it at the Internet Archive last week, if you want a quick intro in video from. https://archive.org/details/12120iadweb (My part starts at 1:13:30). If you have any questions, happy to answer.

“Secure Scuttlebutt (SSB) is a peer-to peer communication protocol, mesh network, and self-hosted social media ecosystem. Each user hosts their own content and the content of the peers they follow, which provides fault tolerance and eventual consistency.[5] Messages are digitally signed and added to an append-only list of messages published by an author.[6] SSB is primarily used for implementing distributed social networks, and utilizes cryptography to assure that content remains unforged as it is propagated through the network.”

Read more: https://en.wikipedia.org/wiki/Secure_Scuttlebutt

Join us! There is a fully functioning client, and bustling Solarpunk community: http://scuttlebutt.nz

[1] - https://delta.chat/en/

Counter-intuitively, the current goal is to centralize the messaging interface, providing a common platform for users who are willing to pay for a consolidated system that saves time and confusion, and to reduce the waste of niche use cases (e.g. having the entire [name here] messaging app on your phone because one weird family member or friend insists on using it)

Bringing email, text, and (aspirationally) all other messaging systems into one.

In doing this, we cannot possibly hope to uniquely integrate each of the countless hundreds of services with their weird quirks and formats, so I aim to establish a standardized language for async messages equipped with the basic common features and extensions for prevalent (but not universal) features like read receipts, "liking/reacting" to messages, etc

It would be really cool to open up that standard to the world if we do a good job of designing it, or collaborate with anyone else who wants to participate in this goal. Our core service will be a niche, paid, closed system, but the underlying language I would like to be open source and interoperable with whatever else turns out to develop (Twitter's new endeavor and Matrix are the two efforts I am most aware of)

After all, Valyrio means Language

I've also been working on an adjacent concept with [https://jwmza.com/polymath], which is a new comprehensive markup language meant to define long-term compatible, human readable documents and websites/document structures which intermix existing standards like (La)TeX, HTML, CSS, Markdown etc.

Perhaps the underlying language and concepts here will mix with the effort to decentralize messaging as well, as messaging is fundamentally the asynchronous bidirectional transfer of documents of mostly text and sometimes other media or information.

This is patently false. Look no further than the other comments in this thread. Nobody can agree on what they want out of the messaging experience. Some want a rich experience like telegram and others want a minimal one like IRC. Should this protocol's features maintain parity with the proprietary and centralized services? Better develop them rapidly to keep up. Should they stagnate to maintain interoperability across an ecosystem? Good luck being competitive. We haven't even gotten to the features themselves and I promise you there will be irreconcilable differences. Does the protocol maintain message history or is it ephemeral? What is the privacy model? Can we delete messages on other servers due to GDPR? When you spend time at the business end of the messaging space, these dichotomies flow endlessly, without consensus, and they cover the entire horizon of the space.

If you think you know the answers to these questions, you don't, because you're solving the wrong problem. If you want to decentralize messaging you have to solve the problem of how to agree to disagree while maintaining interoperability for a shared experience. Let me just make that last point clear: you cannot design a messaging protocol where one party sees emojis and the other doesn't. That doesn't work because you can't lose social information.

Matrix has failed to solve this problem. We need a new approach. What will emerge to finally conquer this space?

COI is chat over IMAP[0], I think it's one of the best solutions I've seen in the sapace.

[0]: https://www.coi-dev.org/

He said the move is to specifically reduce fragmentation in public conversation.

I'm in the decentralized (dWeb) community (we have 10M+ users monthly, in-production, running thru GUN) and admittedly, I think we all get distracted with the war cry of privacy, p2p-for-sake-of-p2p, etc. but is not many other people's end goals.

At this point the closest thing to a standard for person to person messaging is SMS... which like the old guard of ICQ, uses a number rather than a screen name.

That said, I would love to see a good standard for person to person chat, especially one that doesn't rely on magic numbers.

The future of messaging, decentralized or not, is certainly something with a tractable protocol.

The biggest problem with social media is how it is ruining our mental health.

It’s sad that almost no one looking in to this.

I have helped thousands of people exchange contact info after meeting in various niche chat rooms first, but you can flip that and have your irl friends meet you in decn-chat dot org click the 808game room, we can meet in there and pm/dm a code word or something and make each other contacts in various ways.

Your point is something to keep in mind though, ultimately I think we also need apps that can display a qr code like a one time pad to give someone contact details or any chat rooms to meetup in, Other methods of exchanging info about contact and chat room / chat app options should be added to whatever is going to be really good.

2 - I really agree with this, I spent some money and time on rocketchat development to give an optional 'open lobby' to anyone who enters, just pick a name and you are in.. then if you want extra abilities you can add an email or verify or whatever.

easy to get in is important indeed.

Now, I recognize that regular e-mail uses proof of work as a spam prevention measure these days, so I don't want to be too harsh on a decentralized alternative.

That said, I've watched the energy expenditures of Bitcoin etc. with alarm, and I've become very concerned about the sustainability of anything using proof of work if it should become popular. Fundamentally proof of work makes people waste energy in order to accomplish tasks such as sending messages, at a time where we need to use as little energy as possible to reduce emissions.

Have you considered sustainability at all with Aether? Have you considered any alternatives to proof of work for dealing with spam?

This is something that will probably take some puzzling to get to a sustainable mechanism. The strategy I've heard talked about most often is to append metadata that marks it as 'deleted', so when things populate or 'gossip' further, it won't show up in the user interface. Instead it's labeled as 'archived' or 'deleted', or has a link to it's replacement.

Two services don't need to have identical feature sets to interoperate, because you don't need perfect, full interoperability, if you can send most of your messages across platforms, that's still pretty good.

> Let me just make that last point clear: you cannot design a messaging protocol where one party sees emojis and the other doesn't.

Which service doesn't show emojis? Literally every mainstream mobile operating system and desktop browser supports emojis as regular text.

To date, no one ever developed XMPP chat applications as a product, which can be easily deployed and will work consistently on every platform. Client and server developers were always disjointed, working separately from each other. This lead to great inconsistencies in implementations of even such basic functions like adding a contact. Also, it often happens that when a client developers need some feature that honestly should be done by a server, a developer still does this on a client, because it's all he has, with subpar results.

Absent leadership from XSF also plays a role. This club now mostly cares about bureaucracy and following a set of self-imposed rules instead of developing a set of working standards that would allow XMPP apps to compete with the best messaging apps out there. That's why it is unlikely for any great product to appear under such guidance.

We're trying a different approach, maybe we'll even succeed. If so, you'll hear about it on HN.

I actually thought about using matrix, but it apparently "modern" means "resource hungry". My xmpp instance server runs on a small server with 2GiB Ram, with more than 1.5GiB being unused.

It was not used by many people to talk outside of Google servers (federation ability went almost entirely unused), and it suffered from tons of inbound spam problems, so they shut it off.

To answer your question directly though, XMPP doesn’t seem to be that great of a protocol. I’ve heard repeatedly that implementing it is a big mess, and that XML was a poor choice. I’m not sure that this is the reason that users don’t use it, though. It could be that these days, the client-server model is mostly obsolete, as due to battery constraints, phone users mostly rely on Apple or Google push messages to notify.

1) not knowing what is supported on the server, or the other users server.

2) not handling mobile clients well (heavy battery use, weird deliverability issues.)

Yes, I'm full of anger. We had xmpp, pidgin, miranda, trillian. Now we have gazillion mobile only (no, a web gw through a server to your phone is NOT a desktop client) messaging, so let's throw matrix in to solve it, instead of building an xmpp client with up to date xep support. Xkcd 927 on steriods.