Post-submission enquiry for OpenTera WebRTC

[mamaheux]

Previously we have submitted a paper about the OpenTera WebRTC, but it has been rejected because it is not considered in scope for review as research software. We want to explain why we think our paper should be considered for JOSS.

Link of the pre-review issue: openjournals/joss-reviews#5299

We focused on the following definition of research software as defined by JOSS: "supports the functioning of research instruments or the execution of research experiments."

Integration of key and widely adopted technologies is required for robotic research. Most research robots use the open-source Robot Operating System (ROS), which is an ecosystem offering a way to create/use modular robotics software. However, ROS lacks easy-to-use and up-to-date telepresence/videoconference software. Research projects sometimes require developing robots with customized research applications. Human Robot Interaction (HRI) is a robotics field that studies the interactions between robots and humans in different contexts or scenarios.

We aim to develop new telepresence robots and measure the usefulness of new features in remote environments such as long-term care facilities and older adult homes. To do so, a common telepresence/videoconference framework is required, but no open-source solution offers a simple way to develop such robots and commercially available platforms such as Skype, Microsoft Teams and Zoom cannot be customized to fit our needs. Formant.io offers subscription-based robotic fleet management solutions capable of performing telepresence calls. It is compatible with ROS, but has some limitations depending on the chosen plan and available features. Moreover, data retention can be a problem depending on the project’s requirements and universities’ ethical committee considerations.

We developed OpenTera-WebRTC to fill the gap for telepresence research. Integration of WebRTC in robots is not trivial and required significant development efforts. We believe OpenTera-WebRTC is beneficial for the robotics research community and is compatible with the provided definition of research software.

[doumdi]

We are also preparing a JOSS submission for the higher level project "OpenTera" : https://github.com/introlab/opentera/tree/joss-paper-draft/joss-paper. Can you advise if it is suitable for JOSS, and what would be required for acceptation ?