Home Blogs & News

Now Available: FAIR 2.0

Version 1.0 of FAIR Data Services was released in 2020, a major new component of the Aridhia Digital Research Environment designed to put the FAIR data principles into practice.

That is to make data:

• Findable
• Accessible
• Interoperable
• Reusable

FAIR version one achieved this by allowing data owners to create datasets, with a DCAT compliant catalogue and associated metadata dictionaries. Users could easily browse these metadata records and request access to the underlying datasets. In addition to these user features it gave administrators a secure, fully audited platform through which they could track and control access to sensitive data.

Since that first release, the FAIR team has continued to improve the product, and a number of substantial new features have been introduced in the past 3 years.

FAIR 2.0

FAIR has always offered users a degree of customisation within their own hub, and our Landing Page is now completely customisable. Users can add their own banner image and text, provide custom links, and custom content.

Custom Catalogues, allow data owners to create custom catalogue templates for their own datasets, rather than using the standard template. The standard FAIR catalogue template is built on DCAT standards, but we understand that data owners may have additional information they want to hold in their catalogue entry that is not captured by the standard template. To enable this, custom catalogue templates support a variety of field types, and FAIR can support multiple catalogue templates simultaneously, allowing different datasets within the same FAIR hub to have different catalogues.

Our improved Search service allows users to search across FAIR catalogue entries and filter on their results. In addition, following the introduction of custom catalogues, it is now fully configurable, and can support searching across multiple catalogue templates, giving users a consistent search experience when there is more than one catalogue type in the FAIR hub.

In addition to search, users can now also discover datasets of interest via FAIR Collections. This feature allows data owners to curate their datasets into branded collections, which they can add their own logo and description to.

All datasets within a collection contain a link to it in their metadata catalogue, making it even easier for researchers to discover data of interest. Users can also choose to browse a full list of available collections directly.

While metadata is useful, ideally users want to know more about the characteristics of the underlying data before submitting a data access request. Cohort Builder is a powerful new tool that allows users to do just that.

Cohort Builder does not give users direct access to the data, but when enabled on a dataset, it allows users to visualise the data using a variety of charts, and build queries that will help them determine if there is data in the dataset that meets their particular requirements.

Data Owners retain full control of Cohort Builder, and can choose to enable or disable it on their own datasets, or apply fine grain controls: removing particular dictionaries or fields from the query builder, or disabling visualisation where a cohort has a small number of records within it.

Cohort Builder also allows users to create a Data Access Request (DAR) that only contains records of interest and not the full dataset. The DAR process itself is now fully configurable. Data owners can create custom DAR forms for requesters to complete and custom approval workflows, which can accommodate complex approval processes for data access. Data Owners can also track the number of requests, approvals and transfers of their data via Dataset Metrics, which are available on all datasets.

Data Owners can also use our new Data Validation feature, which allows them to identify and resolve any mismatches between dataset dictionaries and the associated data:

A full history of validation reports is available to Data Owners via the FAIR UI.

Our next release introduces a significant new feature: Dataset Conditions. This will allow data owners to apply conditions to their dataset which will be programmatically applied when the data is transferred to a DRE workspace. For example, a data owner could stipulate that their data can only be transferred to a workspace hosted in the EU, and FAIR can enforce this, only permitting transfers to workspaces hosted in that region.

Taken together we believe these features represent a significant upgrade to FAIR, therefore this release will be FAIR v2.0. But development doesn’t stop at version two, and in the first part of 2024 we’ll be working on the following features:
• User Groups – allowing administrators to create and manage groups of users for data sharing
• Transfer Types – supporting different transfer types e.g. manual where the data is not held online, or token transfers where the user is authenticated to access data remotely
• Dataset Creation Wizard – making the dataset creation process as user friendly as possible
• Metadata Copy – allowing data owners to duplicate the metadata of their existing datasets

If you have any feedback on these new features, or would like to know more about FAIR Data Services please contact us.