Open access license

Revision as of 22:53, 23 June 2019 by P.petrelli (talk | contribs)
Template:Stub This is a stub page and needs expansion

  If you are sharing freely your data or code you might be wondering why you should be using a license, but making your data freely available to others doesn't mean that you shouldn't establish in which way it can be used. This is were open access licenses come in handy and these is why you can have different flavours of the same open access license. Here we summarise the ones we commonly use, for a much more detailed view of these licenses ANDS (the Australian National Data Services) has a comprehensive guide which covers also copyright. The important thing to udenrstand about copyright is that it's usually hold by your own institution, veen if you're publishing with CLEx, this is because CLEx is not a legal entity and your institution has the Intellactual Property of your work. When you apply a license to your data you are doing so on behalf of your institution. Unless there are particular circumstances around your data, for example if it is already covered by an agreement or is of a sensitive nature, most institutions have an open access policy and would be fine with a open access license. If you are in dopubt check with your institution if they are fine with the license you want to use.
  Before looking at the different flavours of licenses, we need to understand what are they covering.   First of all these are all open access licenses so by applying one of them you are usually given free access to your data, which means that you are not charging for your data use, you can regulate though how a potential user can access and use your data. Usually a license will cover the following points*:

  • Commercial use and/or research use: you could limit the use of your data to research only or exclude any commercial use and or redistribution of your data.
  • Attribution:  
    • a user is required to cite and/or acknowledge your data, if you choose this option make sure you are also sharing the information on how to do so.
    • if a user is sharing data with others they should also share the license it comes with
  • ShareAlike: a user can modify, augment or transform your data to create a new datasets but has to share it under the same license, so they cannot restrict the use or their derived product.
  • No-Derivatives: if a user modify, augment or transform the data in any way they cannot redistribute it. This can be chnaged of course if they contact you and you allow them an exception.

As you can see there are a lots of different flavours, usually any license will cover "attribution", which is also usually the main reason to apply a license: to get your work recognised!

Licenses are really useful for both the "data" creator and the user, it helps a potential user to work out quickly if the data is usitable for their intended use and helps them citing and using the data in the way the creator wants them too.   While we consider both datasets and code as a form of data, from a licensing point of view they are treated differently. For datasets we suggest the Creative Commons licenses. As well as the international version there is also an Australian Creative Commons. These licenses are simple to use, they offer 6 different flavours, which cover most use cases, and an online tool to help you choose between them. They have a human readable version as well as the legal text and finally the International version was created to cover your data independently of the potential user country of origin.   For software you can refer to the Open Source initiative, we usually apply the Apache 2.0 license to the code we produce. Creative Commons are not suitable for licensing software. * I am listing this using the creative commons terminology but the same concepts might be indicated differently in other licenses.