What data management issues does this case raise? (Maddie Quistgaard)
GitHub could help address these issues with its collaborative interface. If one person does all the research and imports the data, the master of the branch would have to approve the pull request and would look at the data imported. This would then allow the head researcher to verify that the data is not falsified and from there he/she could accept or not accept the changes.
Why didn’t the peer review system identify the problems? (Macy Neblett)
The solution to this problem is slightly more difficult to solve with GitHub. They could have idenified the problems more easily had they used GitHub as a point of collaboration to view and edit their project. Instead of them just blindly accepting the data because of what it would prove, they could have looked at the data for what it was.
How do you think LaCour was able to publish falsified data in such a prestigious journal as Science? (Mine)
If the researchers had used GitHub to publish their work publically, Science Magazine could have gone onto their repository and looked at the data themselves to see where it came from and how it was generated. They could look at the experimental design as well as the data itself. Maybe if they had gone through GitHub, Science could have prevented themselves from publishing an forged and false research.
What are the responsibilities of individuals who co-author papers? What can or should a student (graduate or undergraduate) do when co-author is suspected of falsifying data? (Mine)
If individuals co-authoring papers used GitHub, it would be so much easier to help edit eachother’s work and help one another. For example, not just one person would have to do data collection. You could have multiple people collecting data on different aspects of the experiment. GitHub could also help co-authors find faulty data or bad experimental design. If someone notices that data has been falsified, they can easily ask other collaborators on the research to look at the data as well to see if they also think the data is faulty.