Real-world data analysis as a developer means combing through customer data in SQL, historic (activity) logs in ElasticSearch, exports from Salesforce, issues in JIRA, random additional Excel files from product and sales, and myriad other systems. Switching between tools to transform, join, group, filter, search and visualize is a pain. SQL IDE for SQL, Kibana for Elastic logs, Excel for visualizing, and Python when all else is too cumbersome.
I want to have access to all data sources in a single place and be able to share and script seamlessly between them. I want to end up with a visualization, a recurring report, or a CSV I can send along to someone else.
I am building DataStation to tackle this problem of data exploration.
The easiest way to play with the concept today is through the entirely in-memory online environment. Or you can watch a short demo to see a simple example in action. Since it runs with no server component, the online environment has a few natural restrictions: it cannot connect to SQL databases and it cannot make HTTP requests that don’t set CORS headers. In the next few weeks we’ll release the (also open-source) desktop edition that does not have these restrictions. Feel free to subscribe to hear about this availability and other upcoming features.
The core of DataStation will always be open-source. In the future if it makes sense we may offer a commercial licensed version with features needed for enterprise use.
Due to the offline-first design, DataStation will always be a tool you can bring into your work environment and use just like any other text editor. Don’t worrying about corporate security since it is not (today) a SaaS. We don’t store your data and use minimal analytics to inform product direction.
You can help out by using DataStation, reporting bugs (of which there will surely be many), checking out the source code, joining the community on Discord, and getting in touch about potential use cases.
I’m looking forward to hearing from you!