Libervia

An honest look at what you actually get when you self-host one of the most principled — and most demanding — open-source communication platforms.

TL;DR

• What it is: An XMPP-based communication suite (AGPL-3.0) covering messaging, blogging, calendar, file sharing, and federated social networking — with an ActivityPub bridge for Fediverse interoperability [1][2].
• Who it’s for: Privacy-maximalists and open-source idealists who want full data sovereignty and are willing to invest significant technical effort. Not for non-technical founders looking for a quick Slack alternative.
• Cost savings: Data not available — Libervia has no commercial SaaS tier and no pricing page. The comparison isn’t Libervia vs. a paid plan; it’s Libervia vs. Matrix/Element or Mattermost if you’re willing to do the setup work yourself.
• Key strength: Ideological purity. AGPL-3.0, no ads ever, no commercial tier, full user data ownership, federated by design via XMPP, and a Social Contract that reads more like a founding manifesto than a terms of service [2].
• Key weakness: Development is extremely slow (roadmap items planned for Q4 2023 were still listed as “In Progress” as of this review [1]), there are almost no third-party reviews, the website was inaccessible during research — blocked by its own anti-bot protection — and installation is not beginner-friendly.

---

What is Libervia

Libervia is the web frontend for SàT (Salut à Toi), an XMPP-based communication platform developed primarily by Jérôme Poisson (known online as “goffi”). The project positions itself as a complete communication suite rather than just a chat client — it aims to cover messaging, blogging, file sharing, calendar, group collaboration, and federated social networking all within a single backend that speaks XMPP [2].

The project lives on goffi’s self-hosted Gitea instance at https://repos.goffi.org — not on GitHub — which is itself a statement about the project’s values. The website was blocked by Anubis anti-bot protection during research for this review. This is an irony worth noting on the record: a project protecting itself from AI scrapers made it difficult for a human researcher to reach its own documentation [website scrape].

The philosophical core is laid out in a Social Contract [2] that makes most open-source licenses look timid by comparison. It commits, explicitly and permanently, to: no advertisements, no sale of user data, full libre software licensing, decentralization through XMPP, anti-censorship support, and “opposition to any form of absolute authority” in project governance. This isn’t boilerplate. It reads like something written by someone who thought hard about what communication software should be.

XMPP (Extensible Messaging and Presence Protocol) is the federated messaging standard that predates most modern messaging platforms — it’s what Google Talk and early Facebook Chat were built on before they were shut down or made proprietary. Using XMPP means Libervia federates with any other XMPP server out of the box, giving it interoperability with a large installed base of servers and clients worldwide. The ActivityPub bridge extends this further: if it stabilizes, Libervia instances could communicate with Mastodon, Pixelfed, and other Fediverse platforms [1].

---

Why people choose it

There are almost no third-party reviews of Libervia. The project does not use GitHub, so there is no star count to reference. The name recognition outside of XMPP enthusiast circles is minimal.

The people who choose Libervia are not making a cost-benefit decision in the typical sense. They’re making an ideological one. The Social Contract [2] is the pitch: this is software built by someone who genuinely believes that communication infrastructure should be free, decentralized, and user-controlled — not as a marketing position, but as a condition of participation in the project. “The information regarding the user belong to her, and we will never have the pretension — and indecency! — to consider the content that she produces or relays via Libervia as our property” [2]. That kind of language doesn’t appear in Slack’s terms of service.

The ActivityPub bridge is the one feature that might attract users from outside the committed XMPP community. If you want a single self-hosted instance that can talk to both XMPP users and the Fediverse, Libervia is one of the few projects attempting that [1]. Whether the bridge is stable enough for production use is genuinely unclear — it was still listed as undergoing “stabilization” on the roadmap as of Q4 2023 [1].

The all-in-one pitch also stands out. Libervia tries to replace not just Slack or Signal but the entire communication stack: chat, blog, calendar, file sharing, social timeline [1]. The closest self-hosted parallel is Nextcloud with the Talk plugin, but Libervia’s approach is more integrated by design — everything speaks XMPP internally — rather than assembled from separate modules.

---

Features

Based on the project roadmap [1] and Social Contract [2]:

Core communication:

• One-to-one messaging via XMPP
• Web-based chat interface (Release 0.9, in progress as of Q4 2023 [1])
• One-on-one audio/video calls and desktop sharing (Release 0.9, in progress [1])
• Group audio/video calls (Release 0.10, planned [1])
• Web push notifications for messages and mentions (Release 0.9, planned [1])
• Desktop notification integration (Release 0.9, planned [1])

Social and publishing:

• Blogging and microblogging via XMPP PubSub
• ActivityPub ⬌ XMPP gateway for Fediverse interoperability (Release 0.9, stabilization in progress [1])
• List creation and customization (Release 0.9, planned [1])

Productivity:

• Personal calendar events and agenda view (Release 0.9, in progress [1])
• File sharing

Development and extensibility:

• Plugin/bridge API stabilization (Release 0.9, in progress [1])
• Multiple frontends: web (Libervia), desktop, command-line
• Code forge features — ticket system, merge requests, code browser, repository cloning — planned for a future release [1]

What’s not here: No SSO, no LDAP, no RBAC, no commercial licensing tier, no managed cloud option. The software is AGPL-3.0 with no upsell path [2].

The feature set is ambitious for what appears to be a project with a very small core team. Audio/video calls, the ActivityPub bridge, the chat interface, and the package manager integration were all still marked “In Progress” against a Q4 2023 target [1]. That tells you something real about the development pace.

---

Pricing: SaaS vs self-hosted math

There is no SaaS version of Libervia. There is no commercial tier. The project runs on AGPL-3.0 exclusively [2], which means the software is free, the source code is available, and you are expected to run it yourself.

Self-hosted cost:

• Software: $0 (AGPL-3.0)
• VPS to run it: $5–15/month depending on provider and user count
• Your time to set it up and maintain it: significant (see Deployment section)
• Optional domain: ~$12/year

What you’re replacing:

• Slack Standard: $7.25/user/month
• Teams Essentials: $4/user/month
• Mattermost Professional: ~$10/user/month

For a 10-person team otherwise paying Slack Standard, self-hosting a working Libervia instance would save roughly $870/year in software costs. Whether that saving materializes depends entirely on whether you can get the tool working reliably — which is a legitimate open question given the maturity signals above.

Pricing data for a managed Libervia cloud option: not available. There is none.

---

Deployment reality check

The project’s homepage at https://repos.goffi.org/libervia-web was inaccessible during research — blocked by an Anubis proof-of-work anti-bot challenge [website scrape]. This is a meaningful signal: a tool whose own documentation site requires JavaScript proof-of-work to load is going to be harder to research, onboard, and troubleshoot than one with a standard web presence.

The roadmap lists “Easy Installation Image Updates” — Flatpak and Docker — as a Release 0.9 item that was still planned or in progress as of Q4 2023 [1]. That suggests a reliable Docker quick-start did not yet exist as of that date. A demo environment was also planned but not yet deployed [1].

What deployment likely requires:

• A Linux server with root access
• Python environment (SàT backend is Python-based)
• An XMPP server (Prosody or ejabberd) as the communication layer
• PostgreSQL or SQLite for data persistence
• A reverse proxy (nginx or Caddy) for HTTPS
• DNS SRV record configuration for XMPP federation
• Separate setup for the ActivityPub bridge if you want Fediverse connectivity [1]

XMPP server configuration requires DNS-level work that trips up even experienced sysadmins. This is not a two-hour setup for anyone.

What can go sideways:

• XMPP SRV record configuration is finicky and errors are silent
• The ActivityPub bridge is explicitly described as not yet stable [1]
• The project appears to be maintained primarily by one developer — issues may wait
• Very limited community documentation compared to Matrix/Element or Rocket.Chat
• The self-resetting demo planned for isolated testing was still listed as “Planned” for Q4 2023 [1], meaning you can’t try before you install

Realistic time estimate for an experienced Linux administrator: half a day to a full day to get a basic instance running correctly, including federation tests. For a non-technical founder: this is not a realistic DIY project without dedicated technical help.

---

Pros and cons

Pros

• Uncompromising free software commitment. AGPL-3.0, Social Contract, no ads, no data sales, no commercial tier, no tracking. If ideological purity matters, this project delivers it more fully than nearly anything else in the self-hosted communication space [2].
• Federated by design. XMPP federation means your instance interoperates with the entire XMPP network from day one — no island effect, no vendor dependency [2].
• ActivityPub bridge. If it stabilizes, bridging XMPP and Fediverse in a single self-hosted instance is genuinely rare [1].
• All-in-one scope. Messaging, blogging, calendar, file sharing, social timeline under one backend — more integrated than Nextcloud’s plugin stack [1].
• No company risk. There is no VC-backed entity to go bankrupt, pivot, or raise prices. The Social Contract forbids the commercial moves that burn users of other tools [2].
• Self-hosted code forge. The project does not depend on GitHub. It practices what it preaches [project infrastructure].

Cons

• Development pace is slow. Multiple features with Q4 2023 targets were still marked “In Progress” [1]. Relying on a project with this velocity for production communication carries real risk.
• Effectively no third-party coverage. No reviews on Trustpilot, G2, Capterra, Reddit threads, or tech media found during research. It’s not possible to assess real-world reliability from external sources.
• Website inaccessible. The official documentation site was blocked by anti-bot protection during research [website scrape]. This is an ongoing obstacle for users trying to troubleshoot.
• No Docker quick-start. Easy installation was still an open roadmap item as of late 2023 [1]. Expect manual setup.
• Single-developer risk. The governance model explicitly rejects concentrated authority [2], which is philosophically consistent, but the practical bus-factor is high.
• Key features not yet stable. Audio/video calls in progress, ActivityPub bridge being stabilized, web chat interface in development [1]. The roadmap reads like a project perpetually approaching — but not quite reaching — production readiness.
• No managed cloud fallback. If installation becomes a blocker, there is no official hosted option.

---

Who should use this / who shouldn’t

Use Libervia if:

• The Social Contract [2] reflects values you hold independently — libre software, decentralization, anti-surveillance — and you’d choose it even if the alternatives were easier.
• You’re already running an XMPP server and want a richer web frontend and social layer on top.
• You have a dedicated technical person who can own installation and ongoing maintenance.
• You want a single backend bridging XMPP and Fediverse, and you can accept that this feature is not yet fully stable [1].

Skip it (use Matrix/Element instead) if:

• You want a self-hosted Slack alternative that a non-technical team can actually adopt within a week.
• You need stable, documented deployment with an active support community.
• You want video conferencing that works reliably today.
• You’re evaluating tools under a deadline.

Skip it (use Mattermost or Rocket.Chat instead) if:

• You’re migrating a team off Slack and need familiar UX and SSO from day one.
• You need a commercial support option as a fallback for production incidents.
• You want a tool with a realistic number of integrations out of the box.

Skip it (use Zulip instead) if:

• Your priority is async team communication with strong threading — Zulip’s Apache 2.0 license and self-host documentation are both in better shape.

---

Alternatives worth considering

• Matrix/Element — the most mature self-hosted messaging alternative. Larger community, more active development, better documentation, strong E2E encryption. More realistic for non-technical teams.
• Mattermost — the closest open-source Slack clone. MIT-licensed core, enterprise features (SSO, compliance exports, guest accounts), active community, commercial support available.
• Rocket.Chat — strongest UI similarity to Slack. Free community edition with paid cloud option. More enterprise-ready than either Matrix or Libervia.
• Movim — another XMPP-based web frontend, philosophically close to Libervia, with a somewhat better first-run experience and more consistent release history.
• Nextcloud Talk — if you’re already self-hosting Nextcloud, Talk gives you messaging and video without a separate deployment or protocol stack.
• Zulip — Apache 2.0, excellent async communication model, well-documented self-host path, has been around long enough to have real production reliability data.
• Prosody + Conversations — if you’re an XMPP purist who mainly needs mobile messaging, running a Prosody server with the Conversations Android client is a simpler path to the same federation benefits.

For a non-technical founder who wants to stop paying for Slack, the realistic shortlist is Mattermost vs. Rocket.Chat. Libervia enters the conversation only if free software principles are a non-negotiable starting point and technical resources are available.

---

Bottom line

Libervia is a project built by someone with genuine convictions about what communication software should be, and those convictions show in every design decision — the AGPL-3.0 license, the Social Contract, the self-hosted code forge, the rejection of advertising and data monetization [2]. For the right person, that matters more than any feature list.

The problem is that conviction doesn’t substitute for completion. Roadmap items with Q4 2023 targets were still pending as of this review [1]. The official website was unreachable during research [website scrape]. There is essentially no independent review coverage to validate real-world stability. The installation requires XMPP expertise that most teams don’t have and no Docker shortcut was yet available.

If your goal is to escape a SaaS communication bill, Mattermost or Rocket.Chat will get you there faster and more reliably. If your goal is to run communication infrastructure that you can fully inspect, modify, and trust — infrastructure that cannot be turned against you by a corporate owner — Libervia is the most ideologically complete option in this space. Just go in knowing that “ideologically complete” and “production-ready” are not yet synonymous here.

---

Sources

1. Libervia Roadmap — libervia.org. https://libervia.org/roadmap
2. Libervia Social Contract — libervia.org. https://libervia.org/social_contract

Primary sources attempted:

• Official repository: https://repos.goffi.org/libervia-web (inaccessible during research — Anubis anti-bot challenge blocked access)