Open Science Framework (OSF) announced today that one cannot create new projects in November and then all existing projects will turn read-only in February next year. OSF will move away from sharing materials, but keep the functions such as preregisration and preprint servers.
Previously on this blog: Teaching, Mastodon
This announcement struck me as shocking at first. OSF has been my default for sharing Open Science materials for my research since I came across the Open Science movement in 2017. To me, the core functionality of OSF is always the sharing of Open Science materials such as datasets and R scripts. However, I also use the preregistration and preprint server at some extent. But admittedly, not as frequently.
Just to preserve the historical context: OSF provides some storage space for sharing Open Science materials. As of writing, one can share up to 50GB of files on OSF per (public) project. One nice feature of OSF is to put the Open Science materials in a private project but generate an anonymized link to the materials. This feature is super useful for anonymous peer review. As the pioneer in this space, OSF opens up science beyond papers. There are other things that should also be part of the publication and should be shared openly for scrutiny: Data, code, documentation, etc.
Open scholarship needs open infrastructure. However, maintaining such an open infrastructure as a semi-academic, semi-non-profit project is not sustainable. This story is similar to the last thing that I dedicated a Consolatio: Mastodon. When it gets too big, the maintenance becomes more complex. Also, as written by the founder Brian Nosek, there are abuses. Additionally, more than half of the projects on OSF did not make public and these private projects contradict the original purpose of OSF to promote Open Science. Funding for OSF is also not stable.
I am sadden by the lost of OSF as the default option for sharing Open Science materials. Luckily for now, there are other noncommercial options available, e.g., Zenodo by CERN. But the question remains: How should we maintain a digital common in this late stage capitalism? Before too late and if you can, please donate to the noncommercial digital common that you routinely use, be it software repository such as CRAN or website such as Wikipedia and arXiv. I don’t want to sound like a libertarian, but there ain’t no such thing as a free lunch.