Sharing is Caring (about Research): Three Avenues for Sharing (Protected) Text Collections and the Need for Non-Consumptive Research

A decade ago, the computational turn in communication science was heralded by promises of unseen treasure troves of available digital text containing communication of politicians, journalists, social media users, and many other groups. As these treasures become more closely guarded today, the field needs to think of new strategies to continue to enable researchers who want to engage in computational communication research. One of these strategies is to make text data sharing a more common practice in the field -- which would also greatly enhance reproducibility of research. In this article, we outline three avenues to share as much of your data as possible, while still honouring ethical and legal restrictions. Given the relative lack of infrastructure for some of these avenues, we also highlight the capacities of the Amsterdam Content Analysis Toolkit (AmCAT) to enable non standard sharing strategies. We especially highlight the functions for *non-consumptive research* -- which means analyses methods that can be performed without access to the full data set.