It's not just that. Spark is smaller, easy to understand, easy to modify, and grabbing someone's modified version is unlikely to make your data invalid. Complex

It's not just that. Spark is smaller, easy to understand, easy to modify, and grabbing someone's modified version is unlikely to make your data invalid. Complex codebases with a lot of functionality are less approachable, and having to verify that each change does not compromise its functionality discourages the use of as many forks.