Thanks for the kind words! I hadn't seen qri.io before, but from a read of their website I think it's broadly similar to Dolt (https://www.liquidata.co/ https:/

Thanks for the kind words! I hadn't seen qri.io before, but from a read of their website I think it's broadly similar to Dolt (https://www.liquidata.co/ https://www.liquidata.co/) which is git for datasets. Kaggle has a similar data hub at https://www.kaggle.com/datasets https://www.kaggle.com/datasets that's not open source but is in the same space. Our approach is to scan the datasets wherever they currently live in production rather than being a new way to store the data. The industry seems to have settled on FTP and S3 for now, and we think it's important that we connect to the same exact thing a customer would access. That lets us keep the dictionary up to date automatically without the data providers needing to change their storage infrastructure.