The table below maps each NGI0 project to the centralized-platform risks it most directly addresses. Projects overlap in scope, but each has a distinct primary focus:

Project	Primary risk addressed
ActivityPods 3.0	Surveillance, data lock-in, centralized custody
Bana	Mass profiling, harassment in open networks
Bonfire	Addictive design, opaque algorithms, moderation at scale
Castopod	Creator dependence, platform-mediated podcast distribution
Fediverse Test Framework	Interoperability failure, fragmented federation
ForgeFlux	Monopoly in developer collaboration infrastructure
Lemmy	Link-aggregator monopoly, opaque ranking, platform lock-in
PixelDroid	Mobile image-sharing surveillance, cross-platform lock-in
Spritely	Unusable security UX driving users back to centralized apps

Compared to proprietary platforms such as Facebook, none of these projects optimizes for ad revenue or engagement maximization. The trade-off is that users and communities take on more responsibility: choosing an instance, understanding federation, and sometimes running their own infrastructure.

Centralized social media concentrates risk. A single company's algorithm, policy, breach, or outage affects everyone on the platform simultaneously. The projects in this dossier distribute that risk across independently operated nodes connected by open protocols.

No single project replaces every function of Meta (Facebook, Instagram), Google, Reddit, Spotify, and GitHub combined. Instead, the projects in this dossier demonstrate that each layer of social infrastructure can be rebuilt without surveillance and lock-in as core design principles: personal networking (Bana), community platforms (Bonfire), image sharing (PixelDroid), news aggregation (Lemmy), podcasting (Castopod), developer collaboration (ForgeFlux), data custody (ActivityPods 3.0), open-network interoperability (Fediverse Test Framework), and usable privacy and security (Spritely).

The remaining risks of federated systems (instance-level moderation failures, uneven technical quality, smaller audiences) are visible and addressable because the software is open and exit is always possible. That transparency is itself a defence against the hidden harms of centralized platforms.

If the goal is to reduce harm rather than replicate every feature of proprietary social media, start by identifying which risks matter most to your community:

• Privacy and data ownership → ActivityPods 3.0, Bana
• Community governance without engagement algorithms → Bonfire, Lemmy
• Media independence → Castopod, PixelDroid
• Technical reliability of open networks → Fediverse Test Framework, ForgeFlux
• Making decentralization usable for non-experts → Spritely

Communities need not wait for a single perfect alternative. Running or joining federated infrastructure today, even alongside existing platforms, builds the skills, social connections, and technical fallback that make future migration possible when centralized systems become untenable.