Diseconomies of scale in fraud, spam, support, and moderation
Large online platforms claim they need massive scale to handle moderation, fraud and spam, yet many users report worse support, more scams and opaque automated bans compared to smaller services. Commenters argue that size creates rich targets and moral hazards: big companies externalize the costs of abuse, struggle to provide human support, and sometimes have weak incentives to crack down on fraud that still drives “engagement.” Proposals range from encouraging a more diverse ecosystem of mid-sized platforms and community-based moderation to changing liability and KYC rules so marketplaces like Amazon bear more responsibility for counterfeit and scam activity.
Market structure, competition, and “monocultures”
- Many argue that concentrated platforms create “monocultures” that are attractive to attackers and reduce user leverage.
- Preference for many medium-sized firms over trillion‑dollar giants; some say regulation should prevent monopolies, not just abuses.
- Others counter that even 90%+ market share can erode without regulators, and that centralization is also driven by IP law, government wartime/strategic choices, and network effects.
Moderation, governance, and politics
- Disagreement over whether large platforms are better at moderation:
- One side: scale is needed to satisfy governments’ censorship/“content regulation” demands and coordinate with states.
- Other side: small/federated communities (forums, Mastodon instances, subreddits, Discourse installs, Wikipedia) show that “the only thing that scales with a community is the community,” though drama and edge cases remain.
- Some suggest social platforms now resemble quasi-states and should experiment with more democratic governance; others reject the “platform as country” analogy.
- Large platforms often optimize for protecting the platform or advertisers, not necessarily community standards.
Fraud, spam, and customer support
- Repeated examples of poor anti‑fraud and support at scale: fake storage devices on marketplaces, return scams, ride‑share and taxi issues, gaming and ads fraud.
- One camp: large firms are technically far more competent but are huge, lucrative targets; fraud is a “market” with a nonzero equilibrium level.
- Opposing view: big platforms underinvest deliberately; they could afford robust human review but accept high abuse because it sustains user numbers and engagement metrics.
- Several note that small sites often see less fraud mainly because they’re not worth attacking; others stress that when small sites are targeted, manual, empathetic support still works better.
Email, federation, and centralization
- Running your own mail server: inbound spam is solvable, but outbound deliverability (especially to major providers) is hard, effectively centralizing control.
- This is cited as a cautionary example for new federated social protocols: open standards alone do not prevent de facto centralization.
Law, liability, and KYC
- Proposals: make platforms legally liable for marketplace fraud; require KYC or verifiable identities for sellers/advertisers.
- Critics warn this would encourage discrimination, create more bureaucracy, entrench incumbents, and generate huge sensitive data troves.
- There is frustration that large firms can settle cases to avoid precedent and that cross‑border fraud remains weakly enforced.