Social Jailbreak

Architecture for Decentralized Social Media Aggregation

Revisiting a 2018 Proposal in the Context of 2026

Perig Gouanvic
Independent Researcher, Montréal
January 2026
Original Brief (2018) GitHub

Abstract

TLDR: Browser extension breaking social media monopolies. We forgot basic word processor features matter—formatting, structure, media. Worse now: AI-copiloted markdown has nowhere to publish. Extension captures rich expression, double-publishes to legacy platforms (degraded + link) and decentralized web (IPFS + ActivityPub). Simple need becomes trojan horse: once installed for formatting, you're building the free web—enabling mass migration from corporate prisons.

There is no solution in sight to the problem of social media monopolies. These monopolies have become social jails where users are captive to the retention mechanisms of the platforms. Worse, it has become nearly impossible to free our social existence on these platforms as the sharing mechanisms increasingly retain most of the experience within the original platform. Users are left with what appears to be advertisements for any other given social monopoly.

At the end of the 2010s, it was still conceivable to create a platform which would extract content from different platforms and put the user back at the center of the experience. Now restrictions make it impossible. European legislators have enforced interoperability of messaging systems but not of social media messages and publications.

On the not-for-profit side, Mastodon—taking advantage of recent crises of confidence regarding Twitter (renamed X by an oligarch) and Facebook—has garnered a sizable following and has proven its technical ability to aggregate and welcome various platforms, including for-profit platforms such as Meta's Threads. However, this is not bidirectional: Meta will not welcome the free world into its Threads, while Mastodon welcomes Threads users. Threads was rather a hiccup in the history of monopolies where Zuckerberg tried and failed to make a dent in the market dominated by the other tech oligarch, Musk.

Mastodon and other social media revolving around ActivityPub cannot succeed in retaining enough people. There is a well-known mechanism in social media monopolies where it is impossible to facilitate the migration of a critical mass of people sufficient for them to remain on the new platform. The reason is simply that migration is not a personal but a collective decision made by social networks of people.

Social Jailbreak is an app that was initially conceived as a way to allow masses of people to prepare for mass migration to the free world of social networks and to enjoy the freedom of publishing and consulting content in a platform-independent way.

Publishing Experience

From the user point of view, consulting and publishing can be described as follows.

Publishing involves using the very same social media monopoly (but to this we can now add AI interactions and the full variety of interactive or collective experiences on the web), but using a pop-up editor not unlike grammar and spelling editors (and this functionality would actually be retained in the app for obvious reasons).

In the editor, the user will be able to format their publication in a rich way, knowing that the abuseware practices of platforms such as Facebook—which limit even the mere introduction of an underline or basic HTML URL links—will not constrain them. (The term "abuseware" describes software designed to exploit rather than serve users; see Cory Doctorow's related concept of enshittification and cybersecurity researcher Inti De Ceukelaire's work on abuseware.) The editor allows the user to properly write something that is often difficult in platforms which do not allocate sufficient screen real estate to actually write as one would in a text processor.

On mobile, this enhanced editor would be called differently from the laptop experience, but the basic functionality is to detect that application XYZ is open and a text interface is available. The user only has to validate that they want to leverage enhanced text processing and media publishing.

Technical Architecture: Pre-Publication Storage

The second step is to publish, and this is where the Social Jailbreak infrastructure comes into play.

First, the publication is stripped and adapted to the abuseware practices of the social media monopoly—whether in terms of character limits, URL insertion, text formatting, image insertion, etc. Of course, a link to the original publication is made apparent (note that Medium, Substack, etc. are all platforms whose users' publications can be freed from by SJB).

Then, the original publication (note that original publications should retain as much as possible in terms of context of initial utterance, such as URL of the social jail involved and discussions leading to this comment, within the limits of the intellectual ownership laws imposed on social jail users) is sent to the SJB network, which is composed of:

  • IPFS storage for decentralized, content-addressed permanence
  • ActivityPub federation for interoperability with Mastodon and other free platforms
  • Local-first architecture ensuring user data sovereignty

Consulting Experience: User-Controlled Curation

Consulting the SJB network should be adapted to the preferences of users. Just like operating systems which have different flavors such as MATE, XFCE, etc. on Linux, or different desktop environments on macOS and Windows, or web browsers (Chrome, Safari, Firefox, etc.), SJB browsers could be within web browsers or autonomous apps and may be structured and algorithmically organized as the user decides, depending on what the developer community comes up with.

In other words, SJB doesn't provide a single UX experience and doesn't have a signature that would distinguish it from competitors—because it has already won the competition by giving users control.

Viral Adoption Strategy: Solving the Collective Migration Problem

We have described the difficulty of migrating away from a social media monopoly: it is not a personal but a collective decision, and those who have minimized this problem have caused their users to experience a different kind of captivity—the captivity to micro-communities dominated by paranoia and spite towards those who chose to remain in the other jails.

Social Jailbreak merely presents itself as an enhanced publishing device for any place on the web where you want to publish something publicly. The virality is apparent:

  • Who wouldn't want to have the basic functionality of a word processor?
  • Who doesn't experience the constraint of publishing on Facebook as if we were still in the 1980s—limited to plain text—when even basic word processors had already overcome these limitations in the decades that followed?

While Facebook is the most prominent example, all platforms enforce restrictions on users of this nature.

So if SJB is ultimately a failure, it will at least have been a way to express yourself without those limits. But the objective is that most users welcome the possibility of freeing their publications no matter where they are, before they actually click to publish.

This could be made fully automatic. Installing SJB could even be conceived as a simple way to contribute to the free web: you install it and you never think about it anymore, but you know that you're building the free web. This would be like those browsers and search engines which plant a tree or give to some charity each time you search with them—but it would be that each time you publish with SJB, you give to freedom, not (only) to the monopolies.

Implementation Status

Current Phase: Prototype planning and architecture design

Original Concept: Published in Journal of Brief Ideas, September 2018
View original brief

Technical Stack (Phase 1 MVP):

  • Backend: Python/FastAPI or Node.js
  • Database: PostgreSQL (migrate to IPFS Phase 2)
  • Federation: ActivityPub protocol integration
  • Platform APIs: Facebook Graph, Instagram, X/Twitter, Mastodon
  • Storage: IPFS + Filecoin for decentralized permanence

Seeking:

  • Technical collaborators (especially ActivityPub/IPFS expertise)
  • European funding alignment (Sovereign Tech Fund, DG Connect)
  • Beta testers from activist and academic communities

References

BibTeX

@misc{Gouanvic2026SocialJailbreakRevisited,
  title={Social Jailbreak: Architecture for Decentralized Social Media Aggregation - Revisiting a 2018 Proposal in the Context of 2026},
  author={Gouanvic, Perig},
  year={2026},
  month={January},
  howpublished={\url{https://periggouanvic.github.io/social-jailbreak/}},
  note={Independent Research, Montréal. Prototype planning phase.}
}

Original Brief (2018)

@article{Gouanvic2018SocialJailbreak,
  title={Social Jailbreak: a social media aggregator designed to circumvent social media monopolies},
  author={Gouanvic, Perig},
  journal={Journal of Brief Ideas},
  year={2018},
  month={September},
  url={https://beta.briefideas.org/ideas/8e322a486ac6a4c98fd742580169e2d1.html}
}