Nature's many attempts to evolve a Nostr

Perceived decentralization and oligarchy risk

• Several comments argue Nostr will converge on a small set of dominant relays, similar to email’s large providers or Mastodon’s big instances.
• Others counter that user-owned keys plus the ability to switch/broadcast to multiple relays makes it less “oligarchical” than homeserver-based federated models.
• There’s concern that if one relay becomes “really good,” network effects will recreate centralization despite technical flexibility.

Relays, outbox model, and censorship resistance

• Advocates emphasize the “outbox” model: clients post signed messages to multiple relays (their own, peers’ preferred relays, and some global relays).
• This is claimed to make censorship difficult: relays are interchangeable, and once two users connect, clients can query each other’s preferred relays directly.
• Critics respond that relay operators can still censor locally, governments can outlaw relays or block traffic, and users can’t reliably detect silent filtering.
• Disagreement over strong claims like “you cannot censor Nostr”; some push for more precise “harder to censor than X.”

Incentives, economics, and scalability

• Skeptics note no clear positive incentive to run high-traffic relays and significant bandwidth/abuse risks; fear a “race to the bottom.”
• Proponents point to “zaps” (Lightning-based payments), paid relays, and community motives as incentive models.
• Concerns remain about spam, DDoS, and whether PoW or fees will still work once attackers care about the network.

Moderation, “sewage,” and social dynamics

• One thread argues that a raw Nostr feed becomes “sewage” without strong defaults; filtering burden shifts unfairly to users.
• Others defend the design: clients already hide global feeds, focus on people-you-follow, web-of-trust, labeling (NIP-32, NIP-56), and community relays.
• A big sub-thread debates whether pushing moderation to users is realistic, especially under coordinated harassment or mass-spam.
• Some see Nostr’s goal as reducing the power of “activist moderators” (e.g., instance-level defederation) while still allowing relays to ban illegal content like CSAM.

Identity, keys, and “normie” adoption

• Many predict Nostr will stay fringe because most users won’t manage cryptographic keys or accept irreversible loss of identity if keys are lost.
• There’s debate whether key management is fundamentally beyond users or just a UX/abstraction problem (apps managing keys silently, remote signers, m-of-n schemes).
• Comparisons to house keys highlight the difference: losing house keys does not mean losing the house; losing a private key can mean losing identity with no recourse.
• State-managed PKI and eID cards are mentioned as partial analogues, but there is deep distrust of government-controlled keys.

Comparisons to other systems

• Several comments say Nostr is “reinventing email/Usenet but worse,” or akin to NNTP/freenet with signatures.
• Others contrast it with ATProto/Bluesky (different philosophy, labeler-based moderation), Matrix/ActivityPub, and older P2P systems (Groove, Mojo Nation).
• Some argue email is more decentralized in practice today; others counter with deliverability horror stories and spam-filter oligopoly.
• DHT-based and blockchain approaches are raised as alternative decentralization strategies; some claim true robustness needs BFT/blockchains, others push back on “blockchain determinism.”

Architecture, terminology, and clarity issues

• Multiple readers found the article’s explanation of “relays” incomplete or misleading: current relays act more like database servers than pure pipes, and do not forward to each other.
• Questions about how messages propagate without relay-to-relay links, and concerns about clients needing to “pummel” many relays with duplicates, are only partially addressed by references to the outbox model and relay sync mechanisms.
• Some call Nostr “centralization with extra steps” or a “statically peered superpeer” architecture that inevitably leads to a fragmented, partially connected network.

Overall sentiment

• Technically inclined commenters find Nostr’s simplicity, signed messages, and key-owned identity genuinely interesting.
• A roughly equal amount of skepticism targets its real-world viability: moderation, incentives, spam, UX, key loss, and the risk of repeating earlier, failed P2P experiments.