[FEATURE] In the process of submitting data to the Vault, the possibility to add comments by the datamanager

[Jos-London]

Is your feature request related to a problem? Please describe.

Within Yoda there is a process for submitting data to the Vault. The researcher must then fill in a 'correct' metadata form for this purpose. The data manager can accept or reject the submit request based on the assessment of the metadata form.

If rejected, the researcher will receive a notification that the submit request has been rejected, but without further explanation. It would be useful if the data manager could immediately provide feedback (comments) on which fields were the reason for rejection.

Describe the solution you'd like

In the process of submitting data to the Vault, the possibility to add comments by the datamanager when the metadata is rejected.

This feature simplifies the submitting process in Yoda and prevents or reduces unnecessary email communication. There is also a registration of communication between researcher and data manager within Yoda.

Describe alternatives you've considered

...

Additional context

This feature is highly desirable and is discussed at almost every demo session for researchers.

[lwesterhof]

FYI the default metadata schema includes a remarks field which is used for remarks from the datamanager.

[Danny-dK]

My 2 cents. I have done data curation for several years in Yoda now for various scientific disciplines. My experience is that it is often better to send the user an email with reasons for the rejection and the improvements that can be made. This is easier for both the data manager (in terms of text and formatting) and the user (user does not need to empty remarks field and often there is a list of improvements to be made to file and folder structure, metadata, file and folder naming structure, data structure, documentation, metadata, etc). Because the improvement changes will often involve switching from views and folders within Yoda webportal or local client, it is often easier to have the email on a separate screen to check things that need changing. The mail can then also be easier shared with project members and up till now I have never encountered that there is no further communications between the user and the data manager (it very much goes back and forth; sometimes a lot). With an email the user would also not be able to (accidentally) delete recommendations and improvements, which for the data manager that reviews dozens to hundreds of submissions a month is very nice and allows a quick view of whether the previous recommendations have been applied.

A separate module like the data request module could be developed for review / curations purposes, but my gut feeling would be that it complicates the platform and process a bit too much?

[Danny-dK]

At Leuven University they apparently use a Dataverse instance. They created a data review dashboard. See here amongst others on page 9 IDCC24-LightningTalk-Curation of dataset publication is support, not just about control (KU Leuven).pdf

What I heard from someone (Ignacio Sancho Vargas; WUR Data Librarian) attending the presentation is that:

• you have a tab in which all requested reviews for datasets are gathered and where a data set can be assigned to seomeone for review. In the case of Yoda, this would be the Yoda data manager that gets a notification when a dataset is up fro review. When multiple Yoda data managers exist, they have to decide amongst themselves over email (or other communication method) who will perform which submittal. Such a dashboard might be a nice feature.
• after clicking on a dataset, you get to a tab where details of the data reside (the data package I would assume). This would allow review of the data package. Some thing that the Yoda data manager does as well by specifically opening the folder in the research area.
• a third tab allows for providing feedback of which part can be generated from a checklist of missing elements. The generate form checklist part I would assume to be generated for the researcher before submitting for the vault in Yoda or for publcation.
• Then you have the accept or reject tab.
  

Some are given (partially) within yoda as well, but this does look nice and structured. There are indications that used code will be made public on Git.

I think one of the involved is @okaradeniz (just mentioning here for future reference)