Open Source Ecosystem Project Spotlight: DataPipe

DataPipe is a tool that enables researchers to save data from a behavioral experiment directly to the Open Science Framework (OSF). Developed by Joshua R. de Leeuw, Associate Professor of Cognitive Science at Vassar College, DataPipe aims to simplify the process of implementing born-open data for behavioral researchers. Born-open data is data that is made publicly accessible as soon as it is generated—an approach that emphasizes transparency, reduces duplication and redundant efforts, and facilitates more efficient collaboration and sharing.

de Leeuw is a cognitive scientist who develops research software to make behavioral experiments easier, more accessible, and more cost-effective. He also maintains jsPsych, an open-source JavaScript library for creating browser-based behavioral experiments.

“We want it to be so easy to use born-open data that it's harder to not use it!,” de Leeuw shared. “DataPipe lets researchers connect their experiments directly to the OSF. When the experiment generates data, it shows up on the OSF immediately.”

DataPipe is already gaining traction among behavioral scientists, particularly those designing online experiments with tools like jsPsych. For these users, the tool removes the need to set up and maintain custom databases—an often time-consuming technical hurdle—while also making open data sharing a seamless part of their workflow. By embedding openness at the point of data creation, DataPipe lowers barriers for individual researchers and strengthens community-wide practices of transparency and collaboration.

DataPipe addresses two key, often parallel, challenges in the research community: storing and sharing data.

“For researchers who want to run experiments online, one of the hurdles is setting up a database or similar mechanism for storing data. DataPipe solves this problem by making the OSF the primary data storage for the experiment,” de Leeuw explained. “Researchers often have great intentions about sharing data on platforms like the OSF, but it takes additional work to get everything organized for sharing at the end of a research project. By getting the data on the OSF at the moment it is generated, we hope that there's less friction in sharing data at the end of a project.”

With support from the National Science Foundation’s Pathways to Enable Open-Source Ecosystems (POSE) program, subawarded by COS, de Leeuw is excited about the opportunity to collaborate with the COS open source community.

“I built DataPipe using the OSF’s open API, but often had a suspicion that I could be doing things in different, better ways. Collaborating with folks who work directly on the OSF and are willing to provide their expertise is wonderful,” he said. “The funding will allow us to work on a deeper integration with OSF, and let folks use DataPipe without feeling like they have left their OSF project.”

The project has already achieved notable milestones. “We've saved over 300,000 files to the OSF with DataPipe and looking forward to seeing how this grows once we're more directly connected to OSF as a result of this funding,” de Leeuw said.

By enabling immediate, seamless integration of behavioral experiment data into the OSF, DataPipe strengthens open science practices, reduces barriers to data sharing, and helps cultivate a greater culture of openness within the behavioral science community. As adoption grows, the tool has the potential to reduce duplication of effort, accelerate collaboration, and set a model for other disciplines exploring born-open data practices, ultimately strengthening the shared infrastructure that supports open science.