The motivation behind building Splitgraph is simple: advancements made in the last decade in software engineering and operations, as well as existing best practices, are mostly yet to make their way to data science and data engineering.

There are currently tens of thousands of open-source off-the-shelf services and libraries available, most of them well documented and with clear installation instructions. With Docker, those components can be packaged into self-contained images that can be dropped into any stack with little extra work required.

Instead of server fleets being configured manually, the rise of infrastructure-as-code has meant that whole clusters are now defined in version-controlled definition files and can be provisioned, scaled and reconfigured automatically.

Versioning and revision control is something that every software engineer knows about. Every change to code is tracked. Software builds are performed on isolated CI machines: there's a clear build process that avoids the "works on my machine" class of problems.

Finally, good tools that implement all of this enhance existing abstractions without breaking them. Any compiler, IDE or editor can benefit from Git, Mercurial or SVN without having to be aware of them. Any Unix application can be Dockerized and deployed without being recompiled to use Docker version of system calls.

When we started building Splitgraph in 2018, there was nothing around that hit all these points. Things are beginning to change and a lot of interesting tools are getting released to help data work. However, we still don't feel like any of them completely satisfy our vision. We list applications and services that are most similar to Splitgraph on our Frequently Asked Questions page.