Open Athens
Search OpenAthens Open Athens

The moderator contrasted Peter’s description of tracking users to link resource usage with learning outcomes with the position many US librarians take in believing user privacy trumps tracking learning outcomes. Peter said he was familiar with this difference.

I think this is a valid contrast to make. An example of the US position in public libraries in particular can be seen in this article:  https://www.edsurge.com/news/2019-11-27-linkedin-pauses-changes-to-lynda-com-after-libraries-raise-privacy-concerns.

We made the switch to OpenAthens relatively recently and have needed from time to time to use the ability to see a user’s name (or email) in the OpenAthens admin as part of troubleshooting access issues. Assuming a move to a pseudonymous identifier removes that ability? Or is Peter talking about just what is passed to the content provider?

Thankfully not – the move to using a pseudonymous identifier I was describing was only about the user information that is released to the content provider or publisher.

Within the IdP, as an authorized admin it’s most likely you’ll still be able to see a clearer user identifier for the person you’re troubleshooting for within the console.

When there’s a legitimate need to identify an individual, for example with troubleshooting or the much rarer case of access or downloading exceeding the license terms, then the console should provide the ability to ‘de-pseudonymise’ the data.

I’d be interested to hear Peter’s thoughts on the use of R&S bundle. https://refeds.org/category/research-and-scholarshipThe R&S attribute bundle consists (abstractly) of the following required data elements:

  • shared user identifier
  • person name
  • email address
  • and one optional data element: affiliation

I was aware of the specifics of this only vaguely at the time of the debate, but thank you for sharing the details in the chat. Reading into this a little more, it looks like it’s outside of the immediate discussion we were having, since it’s not recommended for commercially-licensed products

‘This Entity Category (i.e. bundle) should not be used for access to licensed content such as e-journals.’

I think for the purposes it’s meant, which I think is for collaboration using lightweight resources set up by research collaborations between institutions such as wikis or datasets, then this standardized bundle would be very helpful. If set up correctly then use of the bundle would speed up the secure transmission of research data and information. Federations would have to quickly validate the resources, and also the participants in the research projects – those setting up the SPs or resources – would also have to know a bit about SAML and work well with their Identity Provider if that is hosted externally or administrated by their IT department.

Again, this is a use of IdPs and an area of Identity and Access Management where I believe librarians could help and facilitate, as they have done traditionally with print resources.