We propose to build an experimental news recommender community infrastructure to support research in personalization and recommender systems, AI and machine learning, natural language process, HCI, CSCW, and other fields that would benefit from the ability to carry out online field experiments with long-term users of a system. The cloud-based software infrastructure includes a recommender in which researchers can deploy custom algorithms and interfaces, a feed of news articles starting with those obtained through a partnership with the Associated Press, experiment-support modules including consent, payment, and surveying of subjects, and support for two news interfaces—first a news digest and then a progressive web news browser. In addition to the software systems, the infrastructure will maintain a set of long-term consented users, provide extensive support to researchers including master IRB protocols, training, sample experiments, datasets and metrics, and live support through a researcher support team. The infrastructure will be governed by a community advisory board drawn from the researcher community with representatives of the content providers and end-users and charged with allocating experiment slots and steering the development and management of the infrastructure.

Intellectual Merit
While the early years of recommender systems research had extensive experimentation with live users, the increased complexity of maintaining high-quality live systems has led to substantially reduced experimental research in recent years. The consequences are clear—research, and academic research in particular, is biased towards questions that are easy to answer using offline analysis on datasets. As a result, many of the hard but important questions are unexplored, at least in the published research. This infrastructure will enable researchers to carry out one-shot and longer-term experiments with real users without having to develop their own user base or system. It will enable study of important questions about how people react to different forms of recommendation (including diversity, balance of viewpoint, explanations, etc.). The infrastructure is also intellectually interesting as an example of infrastructure that includes a user community. It is our goal to set an example to support other future infrastructures of a similar nature.