Let's learn about F.A.I.R data - Episode #6

In the last article, we covered the Interoperability principles and this is all about the Reusability principles.

R1.1: (Meta)data are released with a clear and accessible data usage license

When data is findable, accessible and interoperable, are we allowed to use it for granted ? The answer could be yes and no. The classic example would be the Creative Commons which is a non-profit organisation that helps public to get access to creative work in a legal way. In a different note, if we are using a copyrighted material for our own purpose, then we are bound to face a legal implication on that. This forms the basis for the right to use/re-use any data or metadata and all FAIR enabled data should be accompanied with a proper usage license. The complexity of licensing arises whenever the data is expected to be exchanged at inter-organisation or inter-country level.

So, without proper license, a FAI data will never become FAIR !!

R1.2: (Meta)data are associated with detailed provenance

A week back, while cleaning the loft I found a bottle which was unlabelled but it was filled with some liquid and the bottle was not tampered. I did opened and smelled it to understand what could be it, but I wasn't sure when it bought ? who bought it ? what it is ? Do we have an expiry date to it ? What can we do with it ? With no hesitation, I throwed this away in the bin.

Just relating the same with the above principle, a data asset should have proper metadata to understand the provenance to answer some critical questions to enable data reusability. Who has generated it ? from where and when ? What is the purpose of this data asset ? How often it will be updated ? Who will maintain it ?

A FAIR enabled data needs a clear provenance and associated information to support and facilitate proper usage of the data asset.

R1.3: (Meta)data meet domain-relevant community standards

The data and of course the metadata, needs to be stored in a structured, organized and standardized way to enable the impact of reusability. The formats of the data is one aspect, but it doesn't stop there, using common vocabulary, metadata stored in a standard format will also form the other dimensions of making data - a reusable asset.

Clubbing 100 excel spreadsheets is not the same as clubbing 50 excel spreadsheet, 10 access database object, 10 JSON files and 30 text files. Joining two datasets using a country attribute will fail if one uses alpha-2 code and the other on alpha-3 and finally, a extracted datetime reported on two dataset extraction differs but they have exported at the same time by two users in the system, this possibly could be due to time zone changes but unless it is referred to the data point, it causes confusion which emphasizes the need of enriched metadata information.

All the above mentioned aspect will help a FAI data becoming a FAIR one. With this article, the introduction and principles of FAIR has been discussed. The next episodes will be around specific topics on FAIR implementation and assessments.

Happy reading ! ????