August 2020 Summaries

The official release of Apache Superset 0.37 marks a significant milestone in its development as an open-source business intelligence application, introducing a range of enhancements and fixes. Key improvements include a re-architected visualization code base that allows developers to create custom data visualization plugins using various JavaScript libraries without modifying the main Superset repository. This release also strengthens security with enhanced row-level security features, enabling multiple table row-level security filters, contributing to its enterprise-grade readiness. Additionally, the update emphasizes better code quality through stricter linting, full Mypy coverage, and the continued migration to TypeScript, ensuring long-term maintainability. Other notable enhancements include CSV upload support to AWS Athena, partition filter options to Presto, and new integrations like SQL-based email alerts and Slack reporting. The Superset community continues to grow, with new users such as Netflix joining, and contributors actively shaping the project's future with plans to share a public roadmap soon.

Nielsen Corporation, a global measurement and data company, has transitioned to using Apache Superset for its business intelligence needs, unifying various R&D teams under a single apps platform. Previously relying on a patchwork of proprietary solutions, Nielsen sought a customizable, open-source solution to address common needs across different groups, leading to the adoption of Superset. Amit Miran, Software Team Lead at Nielsen, highlights the advantages of Superset, including its robust support for Druid, ease of customization, and scalability, which are crucial for Nielsen's operations. The transition involved integrating Superset into Nielsen's platform, creating client-facing dashboards, and actively participating in the Superset community to develop new visualization plugins. Moving forward, Nielsen aims to continue contributing to the Superset community and expand its own visualization capabilities.

The blog post introduces a three-part series on building a StackOverflow dashboard using BigQuery and Superset, focusing on data visualization to assess community health and project growth. It highlights the use of public datasets hosted on Google BigQuery, a column-oriented database optimized for analytics, and details the cost structure associated with querying these datasets. The post guides readers through creating a BigQuery project, setting up a Service Account for external access, and configuring Apache Superset to connect to BigQuery using the pybigquery SQLAlchemy plugin. The process includes generating a private-public key pair and filling out necessary connection details in Superset to establish communication with BigQuery. The aim is to equip readers with the skills to create Superset dashboards from any BigQuery public dataset, with the promise of exploring data preparation and visualization in subsequent posts. Additionally, it mentions Preset Cloud as a managed solution offering BigQuery connectivity and directs readers to further documentation and community resources for troubleshooting.