Scientific data does not fit on most public platforms. GitHub in particular has tight limits on file size, push size (100 MB), bandwidth, and storage ($100 / T

Scientific data does not fit on most public platforms. GitHub in particular has tight limits on file size, push size (100 MB), bandwidth, and storage ($100 / TB / month). Which isn't that surprising; git is designed for code, not data. Even if funders gave large sums of money dedicated to data publication, if recurring billing is involved it will eventually break as attention wanes. Data archives need to be managed by an institution or purchased with a single up-front fee, otherwise they won't stick around. There's also the aspect that, even if you as an individual take it upon yourself to publish your data without institutional support, anyone who reads your paper will most likely ignore your dataset. Which is somewhat demotivating. https://docs.github.com/en/github/managing-large-files/conditions-for-large-files#blocked-pushes-for-large-files https://docs.github.com/en/github/managing-large-files/condi...