2025-06-21
It has been the de facto situation for years that kink communities, along with some adjacent groups like furries, have centered on Telegram. I think that’s understandable: Telegram has been around for a long time and provides a very good user experience. However, Telegram is not without its problems: the company has an iffy track record on security and privacy, and in practice Telegram rarely provides E2E which is increasingly viewed as the best practice for even casual messaging.
I’m not here to litigate the use of Telegram, but I do dislike Telegram for my own reasons. These mostly come down to my concerns about corporate control of messaging platforms and data sovereignty. These lead me to favor open-source options.
In this article, I will review the landscape of instant messaging platforms with a particular eye towards their suitability for kink communities. In other words, I’m going to try to answer this question: if not telegram, then what? Unfortunately, you’ll see that I don’t have any enthusiastic recommendations—so perhaps you should read this, in part, as a call for the community to pay more attention to several nascent open-source efforts.
First, who am I to have opinions on the subject, besides some asshole on the internet? This is an odd question to answer because of how I keep this identity separate from my “professional” one—a common trait of kinksters that is one of the complications of many existing IM services.
I have over a decade of experience in system administration and DevSecOps with a focus on security and regulatory compliance. I have a graduate degree in security. Perhaps more to the point, I have been fallen somewhere between “enthusiast” and “community leader” on perhaps a dozen open-source messaging solutions spanning back to IRC and the very late days of Usenet. I can even claim some involvement in the original Cypherpunk movement, although that gets into my embarrassing teenage years.
Over time, I have come to have many opinions, all of which I like to think are correct—but you might disagree. So to start, let’s agree on some requirements.
There are a few things that I view as important in instant messaging that are somewhat distinctive to the kink community. There are other things that I think are less important. Let’s quickly cover some of these, because I am going to discuss the current landscape in the context of these requirements.
I think it’s useful to discuss the major commercial options, because they’re what most people currently use. Do read this with the understanding that I am specifically trying to get away from options that cannot be “self-hosted,” at least optionally, so I’m not going to repeat the same arguments about corporate control, uncertainty of their future handling of NSFW content, etc. for each one.
Telegram is the current leader in this space. Telegram’s main advantages are a very good user experience, with solid clients available across all platforms and a relatively high level of polish. Telegram is also quite reliable, even for large media files. I’m not surprised that Telegram has mostly won the kink world, I think they’ve nailed the IM user experience. Telegram does have a paid premium offering but I don’t feel like they gate anything that important behind it.
There are also quite a few downsides to Telegram. E2EE is opt-in even for person to person messages and the UX gets worse when it’s enabled (e.g. the desktop client no longer works). Moreover, it’s kind of a sketchy product from a sketchy company run by a sketchy guy with reprehensible opinions. Even more so than usual in the software industry, which is a high bar. Telegram’s record on compliance with law enforcement and other government investigations is also rather… complex, so while Telegram seems cautious about what information they can and will release, it’s hard to have great confidence in their policies and practices.
Telegram fails on the multiple identities front by requiring a unique phone number be verified for each account. Yes, there are plenty of ways of dealing with this, but they’re annoying and I prefer platforms that don’t make you go through the extra work. It does provide user switching in the app, though, which is a good thing.
Discord is probably the other most dominant platform. Discord also has a very good UX, although it’s more complex than Telegram and the conceptual layout of “servers” and “channels” can become annoying (especially considering the cultural norm of popular servers having sixteen thousand different channels). Whether you prefer Telegram or Discord on a UX basis is a matter of opinion. Discord has good clients available for all platforms (“Must be your lucky day!” Linux update experience aside), and handles media well besides strict sending file size limitations depending on how much you pay them. Discord also has good built-in audio/video rooms, a big feat.
Discord’s cons include a heavyweight client and user experience, some annoyances related to its concept of multiple “servers,” and how aggressively it steers you towards paying. Moreover, Discord is a commercial service with commercial motives. In practical terms, this manifests as limited privacy protections, the risk of a future policy against adult content to make the product more investor and advertiser friendly, and in some broader sense the ceding of our kink communities to corporate control.
Discord doesn’t try to stop you from registering multiple accounts, but the mobile app doesn’t have an account switcher. The desktop app does, so that’s an odd limitation that will hopefully change in the future, but it sure is annoying.
Stickers: yes, managed by server admins
Signal has a very strong reputation for security, with a very security-first development history and a well-respected team behind it. The app is also pretty well-implemented. I’m rather critical of Signal in this context, though, for several reasons. First, its group support is very basic and not scalable to large groups. Second, accounts require telephone verification with more aggressive blacklisting of temporary/wholesale VoIP phone numbers than Telegram, making it more difficult (but still not impossible) to create multiple separate accounts. Third, Signal’s E2EE design fundamentally ties your Signal identity to the phone client, leaving fundamental limitations in the desktop client and message history. This is especially painful when combined with the lack of account switching in the mobile app.
Signal has recently mitigated the multi-account problems a bit by providing more of a sense of a “profile” and making it clearer when your phone number will and will not be exposed to another user, so the situation is not as bad as it used to be, but I still have a hard time recommending Signal in a kink context because of how closely tied accounts are to a phone number. Most people simply do not have an “AD” phone.
It should be said that Signal is nonprofit and mostly open-source, but I’m still including it in the commercial category here because parts of the server implementation are closed, Signal does not support the use of servers other than their own (and doing so is practically complicated), and Signal mostly competes with the commercial offerings. That’s sort of a compliment to them, they’ve achieved a level of success that few other open-source options have.
Acknowledging that there are ways to work around them, it is my opinion that Signal’s active efforts to prevent users having multiple accounts makes it unsuitable for kink communities. The fact that you can bypass these restrictions doesn’t eliminate them as a problem; it means that “AD” accounts are unwelcome on Signal and the feasibility of creating and using them will change over time as Signal continues their anti-abuse efforts.
Stickers: very basic
All the fun of hornyposting on LinkedIn. Actually, that joke would have been better for MS Teams.
Let’s take a look at the open-source options, whether federated or not. I’ll preface this by saying that I do, conceptually, prefer federated systems. However, when you consider issues around security, moderation, and abuse, federated instant messaging is a remarkably complex problem. Presumably for that reason, the federated options tend to be behind the centralized options in usability and reliability. In practice, I think federated systems do have the substantial benefit of account centralization. Non-federated options require people to register a new account, which is enough friction that it probably makes people less likely to adopt them.
One of the reasons I have been paying attention to this space is that I would like to help move things off of Telegram, and I have the resources to foot the time and financial burden of hosting something open-source. I don’t want to launch into an obligation like that without feeling pretty good about the adopted platform, though, so I’ve periodically been making test deployments and evaluating the options.
Matrix is currently the leading federated instant messaging system. I have been a heavy Matrix user since nearly the genesis of the project. Unfortunately, Matrix is encountering huge headwinds in the form of abuse, funding, and community issues. The quality level of the “primary” server and client implementations from Element is pretty poor, and the server (Synapse) is particularly resource intensive and difficult to administer when large groups are involved. As a matter of opinion, I am skeptical of Matrix’s future. I think that the quality problems and slow development velocity in Synapse and Element have accumulated to a degree that they will be difficult to overcome. Community-driven efforts like Conduwuit, FluffyChat, etc. may surpass Synapse/Element to become the leading implementations, but I do have my own concerns about the sustainability of those projects.
I want to believe that Matrix is the answer for kink communities, but I just can’t say that with a straight face. My expectation is that in the next couple of years there will have to be a significant change, either Element figures things out and becomes more sustainable, Element collapses but the community sustains other implementations, or Element collapses and the community collapses and Matrix is over. If one of the first two happens, maybe Matrix will be the answer then. Considering the technical complexity of Matrix and the bad blood that has become cemented in the community, I fear that the latter is the most likely.
Matrix has a conceptually strong E2EE implementation but there are some practical issues (mostly around, big surprise, software quality) that create reliability problems and some security concerns that are “not great but not terrible.” Matrix doesn’t have any theoretical problems around media, but there are practical reliability issues. Abuse has been a widespread problem. The spam, CSAM brigading, etc. can be better mitigated within the context of a closed community, but still creates all kinds of problems around user experience and future sustainability.
Stickers: barely working
Look, if I had my way, we’d all be on IRC and we’d just UUCP nudes to each other. I don’t think it’s going to happen.
Stickers: ASCII art
XMPP is kind of like IRC if it grew up and got a job at Cisco. I actually really, really like XMPP and use it daily, but I don’t think it’s a serious contender in this space. For example, consider the clients: there’s a really excellent client for Android (Conversations), but on desktop (and I think iOS as well although I’m not an iOS user) you kind of have to settle for “good enough.” Many of the relevant capabilities of XMPP like large groups (XMPP MUC) and media sharing involve extensions to the protocol that mostly work fine, but can still have problems around UX and occasionally clients that don’t support them. Like IRC, XMPP MUC does not try to keep message history. E2EE is available for XMPP but has multiple distinct implementations, and they’re mostly pretty basic. In short, I don’t think people used to “modern” IM products would be happy with XMPP.
Stickers: in an experimental XEP not implemented by many clients ### Mattermost, Rocket.chat Mattermost and RocketChat are two centralized open-core options that I am grouping together because they are really quite similar. Amusingly, both are rebranding themselves as IM for tactical operators, because in practice the main commercial market for self-hosted IM is defense and intelligence. I have administered both at various points and use Mattermost on an ongoing basis. They’re really pretty good, in terms of software quality, with a good web-based client experience. Neither are federated and both have some limitations in their free or “community” editions. Completely honestly, I think Mattermost might be one of the best open-source options from a usability perspective. However, the fact that it’s non-federated creates an inherent limitation: people will have to register an account which creates friction to adoption, and users of multiple Mattermost servers will have to deal with the server switcher in the app which is inconvenient.
Stickers: nope
Delta Chat is a decentralized and E2EE chat application that makes heavy reuse of existing server-side and encryption implementations for email. It is fair to call it IM over SMTP with the Autocrypt email encryption implementation, although for performance reasons Delta Chat encourages use of servers that support its Chatmail protocol (the Chatmail daemon can currently be deployed alongside Postfix and Dovecot; the Delta Chat project is also supporting an effort to develop an integrated server stack in Go). The technical design of Delta Chat is interesting and certainly eccentric, which makes it all the more surprising that it’s actually pretty good.
In limited testing I’ve found the apps to be well-implemented and easy to use; Delta Chat is clearly designing their apps as a Telegram clone which, considering Telegram’s good and popular UX, feels like the right decision. Delta Chat is not without limitations: it’s a fairly early stage project (around five years old but without wide adoption), and its completely E2EE-first design leads to discovery issues. You are expected to start Delta Chat conversations with people by exchanging keys via QR codes; you can of course exchange keys by another method (like another IM platform) but it’s safe to say that Delta Chat does not provide an easy user discovery experience. I have not tested Delta Chat with large groups but I assume that it will run into the usual scalability problems with E2EE, plus perhaps performance implications due to its reliance on mail servers.
Delta Chat is federated, and since it is interoperable with email for the backend, it’s federated in an interesting way… conceptually you can use Delta Chat as a basic email client, and anyone can connect the Delta Chat client to the email server of their choice. If Delta Chat is the client on both ends, E2EE is set up automatically. Performance over “normal” mail servers is poor due to the slowness of using SMTP for each message; Chatmail support on the server makes it significantly faster (competitive with Telegram in my experience). Delta Chat expects the server to treat messages as ephemeral and remove them after a period of time, so chat history is not preserved except for in client backups (similar to Signal).
Note that Delta Chat is unrelated to the Chatmail commercial secure messaging product. That’s just an unfortunate and confusing name collision.
Stickers: under development
Revolt is an open-source product aiming squarely at a Discord clone, with similar semantics around servers and channels. It is not currently federated, although the project is open to implementing federation in the future. E2EE is also not yet available. I would say that Revolt is currently promising but not mature enough to be a serious contender for a community that isn’t made up of open-source software enthusiasts. The clients are very basic, and the lack of E2EE limits its advantages over commercial offerings. I do think that Revolt is a promising start and may become a major contender in the future.
Stickers: not yet at least
SimpleX is an open-source, E2EE, distributed IM platform with a novel design that probably provides the strongest privacy/security guarantees available. SimpleX is a fairly new project and is somewhat feature-limited. In keeping with the inherent limitations of E2EE and group chats, SimpleX has only limited support for group chats which is not expected to scale well to large groups. Some work is underway to develop a new approach for large groups, but at present, the group situation excludes it from consideration. SimpleX is also a VC-funded software company, creating some sustainability questions (e.g. around future features gated behind payments).
Stickers: very basic
Tinode is an open-source IM platform with a fairly traditional architecture, positioning itself as a replacement/modernization of XMPP. There are a lot of things that I like about Tinode, but it’s in a pretty early stage of development and shares some of the limitations of XMPP (like E2EE only via layered-on-top solutions like OTR). The clients are very basic. I do not consider an option due to its immaturity, but I hope that Tinode continues to improve.
Stickers: very basic
There are a lot of early-stage open-source IM projects, and some that have been around for quite a while that I haven’t tried out. I’m sure I’m leaving out some options, but I don’t think that I’m leaving out any options that are credible contenders for kink communities. Please tell me if I’m wrong!
I wish I had some definite conclusion that we should all be adopting whatever open-source distributed messaging system, but as I said up front, I’m not comfortable recommending anything as a clear winner right now. Depending on the specific set of compromises you are willing to make, my opinion is that the best options are right now are probably Matrix or Mattermost (this mostly revolves around how important federation and E2EE are to you).
And, of course, we have to get to a core question: is there really that much appetite to get off of Telegram? In other words, if you set up something else for kink communities, would anyone use it? I think the friction of adopting something else would have to be very low, and I’m not sure anything else really offers that level of ease right now.
All that said, I feel like this should have some kind of call to action, because one of my goals is to bring like-minded people together to see if we can find better options and perhaps support existing projects that may be able to meet this need. To that end, and for maximum irony, consider joining this telegram room to discuss.