It feels strange to say this in 2019, but one of the biggest tech companies in the world is betting its future on interoperable messaging. In a major announcement on Wednesday, Mark Zuckerberg announced a shift in focus for Facebook, moving from the News Feed to encrypted, ephemeral messaging — or as he put it, from the town square to the living room. After years of pushing for more data shared with more people, Facebook wants to give users a quiet place to talk.

It’s a fascinating plan for lots of reasons — in part because of the break from Facebook-as-usual — but it’s hard to know what it might look like in practice. The relatively simple part is knitting together the products Facebook already has. Facebook, WhatsApp, and Instagram all operate direct messaging services, and The New York Times had already reported that the company was looking to connect them, letting you reach a WhatsApp user from an Instagram account and vice versa. But the plan Zuckerberg described Wednesday goes further, changing how those apps interact with protocols like SMS and how we use our phones to send messages in general.

Zuckerberg laid out the basics of the plan in Facebook’s new “Interoperability” principle: “People should be able to use any of our apps to reach their friends, and they should be able to communicate across networks easily and securely.” He didn’t give a lot of details on how all this would work, and there’s a lot we just don’t know — but the basic idea was a single catchall system for all the messages you get on Facebook, Instagram, and WhatsApp. And since Facebook controls the backend for all three apps, it can store more complex metadata, like which app will send you a notification when a WhatsFaceGram message comes through. As Zuckerberg described it:

We plan to start by making it possible for you to send messages to your contacts using any of our services, and then to extend that interoperability to SMS too … You can imagine many simple experiences — a person discovers a business on Instagram and easily transitions to their preferred messaging app for secure payments and customer support; another person wants to catch up with a friend and can send them a message that goes to their preferred app without having to think about where that person prefers to be reached; or you simply post a story from your day across both Facebook and Instagram and can get all the replies from your friends in one place.

Of course, things get more complicated when you try to include protocols like SMS, but there are a couple of clear precedents Facebook could follow. The best example is how SMS works on Android: users can designate a preferred app like Signal, which will use its own protocol with other Signal users and default to SMS when it encounters a number that isn’t on the system. iMessage is a walled-garden version of that, splitting between SMS in green bubbles and data-based encrypted messaging in blue bubbles, with the split built directly into the operating system. Facebook is proposing a similar system on a broader scale, applying blue-bubble-style features where possible across its entire suite of apps.

If you’re of a certain generation of techie, this all might sound a bit silly. There are lots of truly interoperable messaging standards, whether it’s SMTP (aka email) or XMPP (aka Jabber), and if you don’t like how they work, you can always start a new one. Facebook isn’t proposing anything like that. Facebook is proposing a new system that it controls, which would aim to crowd out existing open standards as well as competing services from Google, Apple, and everyone else. It’s very much a power play, and if it works, it would put Facebook at the center of one of the most important things we do with our phones.

It would also come with its own business opportunities, depending on how the system is constructed. Robust encryption would mean the company can’t read the text of messages, but if Facebook kept the metadata, it would reveal who you’re texting, which could be a powerful tool for targeted advertising and help Facebook build out its graph of who knows who. It wasn’t mentioned in Zuckerberg’s post, but Facebook is also reportedly working on a blockchain payment system that would let users send money through messaging apps, which would make the proposed system even more lucrative. And as regulators around the world start to think about splitting off WhatsApp and Instagram, bringing all three networks onto a single messaging system could be a crucial political protection.

There are a few more ambitious gestures in Zuckerberg’s statement, but they’re more dreams than concrete plans. It’s clear Facebook would like it if iOS would follow Android’s lead and let third-party apps take over SMS duties (something Apple is unlikely to do). He also suggests it would be nice to extend features like encryption to standards-based messaging, although it’s not clear if Facebook is actually putting its weight behind a new carrier-adopted messaging standard or if Zuckerberg is just flagging a problem so we don’t get mad when it pops up later.

The biggest problem for the proposed scheme is something Zuckerberg only touches on briefly. We’re in the middle of a massive shift in how texting works, as the SMS protocol is replaced by a new protocol called RCS, or Rich Communication Services. Like SMS, RCS is basically controlled by carriers, updating the standard with new features like read receipts and higher quality images. (This is a good explainer, if you want more detail.) Carriers and operating systems are taking a while to get on the same page with RCS, but it’s coming, and eventually, it’s expected to replace SMS entirely.

The problem is that, as it stands now, RCS isn’t supported inside of third-party messaging apps, which means you won’t be able to use WhatsApp to send RCS texts. So even as Zuckerberg tries to build bridges with open standards, the latest standards are cutting him off. RCS is still in flux and it’s not impossible that third-party support might arrive.

But even then, there are other issues: Apple hasn’t said a word about whether the iPhone will ever support RCS and RCS is not an end-to-end encrypted protocol. Without a mobile operating system, Facebook won’t have much say in how it turns out, which makes this a very interesting time to launch a new push for interoperable messaging.

It’s an ambitious project, and a risky one, too. Google has spent years in messaging purgatory, trying to reconcile different apps and teams into a single unified product — until it finally gave up and ceded the whole thing to carriers with RCS. Facebook will have less goodwill to draw on, and the interoperability project cuts against some of Facebook’s core products and principles. It’s easy to imagine the product getting stranded, or simply being abandoned halfway through.

But if it succeeds — even slightly — Facebook will have given us a new approach to one of the most fundamental tech products there is and it will have a shot at extending its social networking monopoly into the messaging space. Faced with mounting unpopularity and dwindling options, Zuckerberg seems to have decided it’s a risk worth taking.