I've had multiple team-members run away from Azure Datalake because the only way to "deal" with JSON data is to use "some half-baked example samples from some blog post".
I've been advocating for using ADLS and USQL but without the feeling of "native" JSON tools. I recommend legitimizing these objects by keeping them in this repository as a top level folder with a folder structure that enables contribution of additional formats by the open source community and getting rid of "samples" from the namespace.
brandonwittwer
commented
6 years ago
I've had multiple team-members run away from Azure Datalake because the only way to "deal" with JSON data is to use "some half-baked example samples from some blog post".
I've been advocating for using ADLS and USQL but without the feeling of "native" JSON tools. I recommend legitimizing these objects by keeping them in this repository as a top level folder with a folder structure that enables contribution of additional formats by the open source community and getting rid of "samples" from the namespace.