Discovery is another. "Differential dataflow" and "timely dataflow" don't quite convey clearly what problem they're solving or what machine characteristics they rely on or even where to expect more performance from them and how. Not saying that they aren't performant, but we need to say how and where pretty clearly.

For example, Spark makes it clear that its performance comes from exclusively in-memory compute across a cluster.

Spark may also be "good enough" from a performance standpoint for many use cases. New tools can get adoption among small players only if they are radically easier to use and deploy.