Often people want to share data, choosing the right method can be complex.

This page hopes to help you choose the right solution.

Options:

Situations:

  • one to few people
  • one to many people
  • share with public
  • co-authoring
  • moving datasets between institutions

GDPR (Dutch: AVG)

Whenever you are sharing documents or files, be aware that restrictions apply when sharing personal information and all involved parties must explicitly accept responsibility for keeping within the legal boundaries.

Warning

Do NOT use we-transfer, it is entirely unclear where the data is stored.

One to few people

With small data (up to a few megabytes) , e-mail attachments are often the easiest, but sometimes e-mail filters will get in the way, for example zip files, executables, or even some document formats are routinely blocked to prevent spreading of malware. A good way around that is surf filesender or surfdrive or Microsoft Teams. Within the faculty, you can use shared storage to share documents within a group.

A usb-stick in a physical envelope (with encrypted data on it) can even be used, but this is often too much work.

one to many people

If you want to send a file or dataset to a lot of people, but not make it public, there’s no clear solution that stands out. The solution also depends on how private the data is within the group, if it contains data that must remain within the group, encryption is probably a requirement. Systems where the group is know (unix groups, teams, etc.) which allow storing files are usually the best place to use for sharing within the group.

make data publicly available

For data or files that you want to share with the public, it should be downloadable using a web-link (https://....), we have several ways to make data available like this, either by making a website or adding a web sharing tool to a shared volume. For publications there the Radboud Data Repository (RDR).

co-authoring documents

Another concept of sharing is shared editing, where two or more people can edit documents at the same time or at least have edit access to the documents.

A document on a shared filesystem can be sufficient, but this is not suitable for parallel editing, because the file sharing has no knowledge about the tool being used to edit.

Shared editing exists on many platforms, like Microsoft Teams, google docs, overleaf (for LaTeX documents), all of these tools require a login and/or a license and may or may not be acceptable within your organisation.

moving large datasets

Sometimes, when a lot of data must be copied or moved from one physical location to another, it can take a long time to transfer over a network. It may be faster to move a physical disk, or server to the other location. If the only copy is being physically moved, this may be undesirable.

If time is not critical, a multi-terabyte transfer can be done over the network, but then the rsync tool is recommended, since interruptions can be efficiently dealt with.

Note about Surf Filesender

We have installed the commandline tool for surf filesender on lilo and clusternodes, all you need to do is download your configuration and set it up in your homedirectory. For this you need to login to Surf Filesender and download the configuration from there to your homedirectory/.filesender/ and then use the command filesender.py with the correct options to easily send a file or directory from the linux commandline.

setting up data sharing

If you have questions about how to do it or if you need our help, feel free to ask.