Skip to content
Where We're At

Open questions

Our (still young) ecosystem has open questions with no consensus answer. Agree or disagree as you see it, because there are no right answers here.

Decentralization

  • Strongly disagree
  • Disagree
  • Neither
  • Agree
  • Strongly agree

  1. The PLC directory being one service is a serious problem.

    Mean 3.71171 responses

  2. Moving the PLC directory to an independent organization will meaningfully reduce that risk.

    Mean 3.90171 responses

  3. The network would survive if Bluesky PBC shut down tomorrow.

    Mean 2.89174 responses

  4. Bluesky PBC should do more to ensure the network will survive without them.

    Mean 4.13172 responses

  5. ATproto is decentralized enough today.

    Mean 2.59173 responses

  • Strongly disagree
  • Disagree
  • Neither
  • Agree
  • Strongly agree
Agreement on a 5-point scale, centred on neutral. Each row shows its mean from 1 to 5 and its number of responses.

Sustainability

  • Strongly disagree
  • Disagree
  • Neither
  • Agree
  • Strongly agree

  1. Running core ATproto infrastructure is affordable for an individual.

    Mean 3.02162 responses

  2. I can keep doing my ATproto work at the level of funding I have now.

    Mean 3.27153 responses

  3. Money pressure has made me consider stopping my ATproto work.

    Mean 2.67152 responses

  • Strongly disagree
  • Disagree
  • Neither
  • Agree
  • Strongly agree
Agreement on a 5-point scale, centred on neutral. Each row shows its mean from 1 to 5 and its number of responses.

Who should pay for core ATproto infrastructure?

164 responses

  1. Grants and public funding80.5%, 132 people
  2. A cooperative of operators75.6%, 124 people
  3. An independent nonprofit75%, 123 people
  4. The people using it65.9%, 108 people
  5. Bluesky PBC62.2%, 102 people
  6. Anyone2.4%, 4 people
  7. A small protocol-level fee1.8%, 3 people
  8. Advertising0.6%, 1 person
  9. Corporations0.6%, 1 person
  10. Donations or open collective0.6%, 1 person
  11. Freemium or paid tiers0.6%, 1 person
  12. Other big apps0.6%, 1 person
  13. Operators funding themselves0.6%, 1 person
  14. Taxes0.6%, 1 person
Share of the 164 people who answered. Several answers each, so shares can add up past 100%.

Maturity

  • Strongly disagree
  • Disagree
  • Neither
  • Agree
  • Strongly agree

  1. Bluesky PBC should do more to actively teach people about the network.

    Mean 4.10172 responses

  2. The ecosystem needs a process for resolving lexicon naming conflicts.

    Mean 2.96160 responses

  3. The protocol is stable enough to build a business on.

    Mean 3.58166 responses

  4. ATproto needs permissioned data before it can support most of the apps people want to build.

    Mean 4.01170 responses

  5. Spam and abuse on the network are under control.

    Mean 2.85171 responses

  • Strongly disagree
  • Disagree
  • Neither
  • Agree
  • Strongly agree
Agreement on a 5-point scale, centred on neutral. Each row shows its mean from 1 to 5 and its number of responses.

What is hardest about building on ATproto right now?

83 responses

  1. User aquisition25.3%, 21 people
  2. Funding18.1%, 15 people
  3. User education16.9%, 14 people
  4. Private data13.3%, 11 people
  5. Too dependent on Bluesky PBC7.2%, 6 people
  6. Documentation gaps6%, 5 people
  7. Marketing6%, 5 people
  8. Spam and moderation burden6%, 5 people
  9. OAuth is too complex4.8%, 4 people
  10. AI use3.6%, 3 people
  11. Bluesky Staff Behavior3.6%, 3 people
  12. Developing shared lexicons3.6%, 3 people
Show 34 more
  1. Non-atmosphere user flow3.6%, 3 people
  2. Time3.6%, 3 people
  3. AI hate2.4%, 2 people
  4. Consistent sign-in experience2.4%, 2 people
  5. Focus of app builders2.4%, 2 people
  6. Legal questions2.4%, 2 people
  7. Protocol instability2.4%, 2 people
  8. Anti-Patterns used by Bluesky1.2%, 1 person
  9. app interleaving patterns1.2%, 1 person
  10. AppView-specific data1.2%, 1 person
  11. ATproto being associated with social media only1.2%, 1 person
  12. Branding being based on hosting a PDS1.2%, 1 person
  13. Bus factor for OSS1.2%, 1 person
  14. Developer experience1.2%, 1 person
  15. Differences between frontend and backend oauth1.2%, 1 person
  16. Distribution1.2%, 1 person
  17. Extending lexicons1.2%, 1 person
  18. Figuring out the correct architecture1.2%, 1 person
  19. Finding apps that need to be built1.2%, 1 person
  20. Fine-grained OAuth permissions1.2%, 1 person
  21. Group primitives1.2%, 1 person
  22. Hosting1.2%, 1 person
  23. Interop with other apps1.2%, 1 person
  24. Lack of libraries1.2%, 1 person
  25. Lexicon migration1.2%, 1 person
  26. Missing lexicons1.2%, 1 person
  27. Motivation1.2%, 1 person
  28. No atmospheric profiles1.2%, 1 person
  29. No backlink notifications from PDSs1.2%, 1 person
  30. No cross-atmosphere follows1.2%, 1 person
  31. Overall complexity1.2%, 1 person
  32. Promotion of copied apps1.2%, 1 person
  33. Records acting as backup1.2%, 1 person
  34. Signature verification1.2%, 1 person
Free-text answers grouped into topics. Share of the 83 people who answered.

Helpful resources