As Twitter experiments with some metrics to improve the health of conversations, privacy issues have delayed partnerships with the outside researchers
Kurt Wagner / Recode : Tweets: @kurtwagner8 , @chrisdanforth , and @alessuube Tweets: Kurt Wagner / @kurtwagner8 : Here's an inside look at Twitter's ambitious plan to change the way we tweet. As with many things Twitter does, it's taking longer than expected. http://www.recode.net/... Chris Danforth / @chrisdanforth : Measuring & incentivizing “healthy conversation” on @Twitter will require 100x more investment than a pair of small underfunded academic projects http://www.recode.net/... Alessandra Suuberg / @alessuube : 1/ “[Researchers] would receive access to user data and a monetary grant from Twitter; in exchange, these researchers would create new metrics intended to measure the health of interactions on the service.” http://twitter.com/...
Context & Ripple Effects
Twitter's health-of-conversation push began with the 2018 review of proposals that selected two academic teams, and an earlier experiment testing whether showing users the Twitter rules changes behavior. This Recode report explains why the follow-through has lagged: privacy vetting around giving outside researchers user data and grants has slowed the partnerships themselves.
The friction matters because the academics involved, including Chris Danforth, argue that measuring and incentivizing healthy conversation requires far more investment than small underfunded projects can muster — meaning Twitter's own timeline depends on data access it has been slow to grant.
First-order effects
- The two academic teams chosen in 2018 are working without the full data access and funding originally framed for them, delaying any validated 'health' metric Twitter could ship.
- Twitter's product roadmap for healthier conversations slips, since metrics meant to inform features cannot be built until the researcher partnerships clear privacy review.
Second-order effects
- The delays push Twitter toward formalizing researcher access instead of ad hoc deals — a path it later took with the Developer Policy rewrite for academic research and the expanded Moderation Research Consortium.
- Academic teams become dependent on the platform's own gating of user data, a dependency that resurfaces when researchers warn an API paywall would jeopardize studies of influence campaigns.
Third-order effects
- If the pattern holds, independent study of platform health converges on tiered, platform-controlled access — the company decides who measures it, under what terms, at what price.
- That structure leaves civil-society and under-resourced researchers structurally disadvantaged, turning 'healthy conversation' research into a privilege granted by the platforms being studied.
The trend: Platform health research is shifting from open academic partnerships toward tiered, platform-gated data access that the platforms themselves control.