While I mainly host my datasets on my Github repository, I have also cross-shared some datasets on data.world as the platform is integrated with quite a couple of other tools. And also, data.world is more user-friendly for users who might not want to dabble into Github, and allows us to bookmark datasets. In addition, datasets are more easily searchable on Google Dataset Search.
Reading data hosted on data.world into Python and R is possible, however it requires an API authentication token. Instructions can be found here for Python and here for R. If you find setting up a hassle (it's a one-time hassle!) or are in a rush, you can read data from Github into Python (tutorial covered in a previous blogpost), or download the file and read from local.