Reading · The Paper Pilot's Digital Garden

Discord has a bit of a monopoly on what it does, and there is no currently existing alternative I believe we could target a mass exodus to. But let's go over the alternatives that exist and how switching would look.

Alternatives

You'll notice a lot of the positive endorsements in this list are open source. That's really important for ensuring that users can switch to another host or another fork if the original main instance enshittifies or shuts down. RIP Guilded.

Stoat

Stoat, previously revolt chat, is my current top pick for discord alternatives. Its worth pointing out that they have not made it clear how they plan to be financially sustainable, though. The influx has users has made everything really slow though.

Fluxer

I see more people discussing this favorably than I expected, but I’m not thrilled about it only being vibe coded by a single person and the slightly suspicious squashed commit at the end of the git history. But besides that it’s similar to Stoat but has promised eventual federation.

PieFed

This is not a discord-like, but it actually makes some compelling arguments for why it should be considered. Depending on the type of community you have, this might fit your needs substantially better than discord did.

Zulip

A nice looking alternative that is self-hostable, but for casual users who just want to quickly create a community on the main instance, they might find having limits on message history and file uploads on the free plan unacceptable, even though it makes it a much more solid business model than discord and will hopefully delay it's enshittification. This is similar to slack as well.

Chatto

I do like the idea of a single client connecting to all servers directly, without needing to go through a home instance. Having to create accounts per-server is annoying though, and for OIDC the server owner has to set it up with each provider, which is a lot to do.

Matrix

Matrix has a lot of problems that are not showing any likelihood of being fixable. I like that it's federated, but everything else about it works against it.

Steam

Funnily enough, Steam chat is a hidden gem in the realm of discord alternatives. It supports group chats with channels, has global emojis and stickers via the points shop instead of a subscription, has various features like VC, screen sharing, voice messages, reactions, and (slightly more restricted) file uploads. It’s run by a libertarian who I’m not the biggest fan of, but would expect to push back more than discord did on stuff like age verification, it has a sustainable business model that doesn’t rely on enshittifying the chat experience, and a lot of your friends already use Steam anyways so the network effect won’t even get in the way! The biggest things, imo, that it’s missing are replies, threads, and robust permission systems. This can only replace small discord groups between trusted friends.

Root

I haven't tried this app, but based on the LLM smells on its website, the fact it's already got $9 million in investments that are going to be expected to have some form of return, and its closed source, I see no reason to consider it over stoat or other options here.

Signal

Signal is not discord-like, but it is notable for being essentially the only actually secure messenger app. That's certainly not a requirement, and discord itself was not secure, but as part of moving away from discord people are reevaluating what they need, and if e2ee is important for you, no other option is as secure as signal.

Prototypes

There are also several alternatives currently in development that look promising once they're further along. For example:

Polyphony is a discord-compatible server and will eventually include a client that can show chats from multiple servers, including discord itself. Being discord compatible means any tooling made for discord should be able to work with this software too, just by changing the base URL. It's part of an ambitious project called polyproto for federation and identity management. Also stoat has said they're open to implementing the polyproto API, which would allow revolt servers to also appear in the polyphony client.
Spacebar is a similar but smaller in scope project. It also aims to recreate the discord server and a client that can talk to multiple servers.
Roomy is a discord-like built on ATProto, which stores the messages in your PDS, ensuring you have control over your data.

Switching

Mass Exodus

Network effect makes switching hard, and historically it's only happened in cases where something new comes along that is so much better people are willing to try it without their friends, like musical.ly (TikTok), or when something really egregious happens at the status quo app, like happened with Twitter getting bought by Elon Musk triggering an exodus to mastodon and bluesky.

For the latter case, it must be pretty severe to actually trigger the exodus. Apps enshittify without triggering exodi all the time, discord included.

Additionally, how the app is structured can help or hinder the viability of a mass exodus. Reddit, for example, is similar to discord in that it is broken up into communities, and the leaders of those communities thus have a lot of power, and those leaders are the most likely to care a lot about the issues with the platform compared to casual users. However, during the reddit exodus after the 3rd party app changes, any community leaders who tried to shut down and migrate their communities had their moderator privileges stripped from them by reddit and new, sympathetic moderators, put in their place. I think this same approach would NOT work on discord - they couldn't simply replace server owners with other users, and if the server owner makes the server read-only (with only instructions on how to find the community on the new platform) there is no obvious place for people to coordinate an alternative server to switch to.

I don't think the trigger has happened yet, partially because there also isn't an obvious platform to coordinate moving to. One thing I think that will matter a lot for convincing owners of large servers of a potential new platform is robust tooling to match discords bot ecosystem. If existing tools can be expanded to support multiple platforms, or alternatives written for other platforms, that will lower the switching cost for those reliant on those bots today.

Small servers/groups

While mass exoduses would have to leverage large servers to be successful, a lot of people may suggest they actually can't leave because of their small, intimate servers and group DMs. However, these are the precise forms of communities that the network effect breaks down on. If instead of collectively convincing most of the entire user base to switch, you only have to focus on 5 people who know you personally, its a lot easier! I, and likely you as well, probably already have group DMs that migrated to discord, so there's no reason to believe another migration can't also happen.

Bridging

Bridging, where messages are automatically copied from one platform to another, can help during the transition by allowing people to move platforms while still communicating with the part of the community staying on the old platform. Bridges are allowed by Discord TOS since they use bot accounts and not user bots and there are already many bridging options available. I haven’t played around with these too much beyond a matrix bridge with my discord server, which I don’t recommend as I don’t recommend matrix anymore, but Lightning looks like a really promising bridging framework with support for a lot of services.

Discord Alternatives