I am trying to choose between these different generalist repositories:

  1. Zenodo
  2. Open Science Framework
  3. Figshare (or Dyrad??)
  4. Harvard dataverse
  5. General cloud services
  6. ...

Figshare has a nice comparison of them ( https://doi.org/10.6084/m9.figshare.751546.v1), but I was looking for some guidelines to help me to choose the best one that I might have ignored. For instance, many European Projects require to use Zenodo so maybe is better to get acquainted with it, however, Open Science Framework seems to offer a better way to organize the projects but it seems there is no doi to the files uploaded. Is there any guideline or anything I should be aware of before choosing to put my dataset in a generalist repository?

Content I would like to share

I would like to use a generalist repository because I would like to share all my outputs. This might consist of code, dataset but also figures, drawings and some times I share also some piece of hardware design (e.g. step files for 3D printing the adapters, sensor net, some PCB design).

  • You might want to specify what kind of content you would be using the repository for. – Tommi Apr 30 '19 at 7:12
  • Consider what license you want to publish your outputs under. – Anyon Apr 30 '19 at 11:14
  • @Anyon yes, and regarding this are there any differences between these generalists repository? – G M Apr 30 '19 at 11:39
  • @GM Last time I checked Zenodo offered a wider range of licenses than Figshare, but if the one you want is covered it might not matter much. I don't know how your other options compare in this regard. – Anyon Apr 30 '19 at 12:05

I haven't used any of these resources myself but hopefully the following points can help you in your decision:

This kind of repository and the open science trend in general are pretty recent. This is why it's a bit difficult to evaluate and compare the different options, there's not enough hindsight yet. This is also why, in my opinion, the main risk is that the platform you choose would become obsolete or disappear in the future. So to be safe you might want to choose a platform which allows you to export your stuff easily.

Another important point you should consider is the policy of your institution: the outcomes produced as part of your academic job can be subject to IP restrictions. So it might be a good idea to check with your local IP or legal office what kind of license they allow/recommend and whether there is any IP disclosure process in place.

It's also worth mentioning that there are a few more generic options for sharing outputs: for instance Github, Google Drive, and of course the good old personal webpage (most universities let you create one under their website).

  • Actually, I am using Google Drive and I wanted to put it in the list it has some feature that are missing the others, I'll edit the question now – G M May 1 '19 at 17:18

You can find repositories that assign DOIs to datasets in the Repository Finder Tool.

You can filters those that actually meet the criteria of the Enabling FAIR Data.

edit: Full disclosure, I work at DataCite as a developer, and participated in the development of the repository finder.

Not the answer you're looking for? Browse other questions tagged or ask your own question.