Separating Debezium from Kafka

Is Debezium mean to be fully functional without Kafka? I suspect the answer is yes. If so, are there any plans to extricate Debezium from its dependency on Kafka? I can think of three main dependencies (there might be more):

1. The data model (Kafka connect Schema/Struct)
2. The deployment model (embedded vs. Kafka Connect)
3. The state (schema history topic, schema change topic, , etc.)

I am mostly interested in (1). I've also had some discussion with gunnar.morling, who has mentioned wanting to find an equivalent Kafka Connect Struct/Schema that isn't tied to Kafka connect. I've had some discussion with shivangshah15, who is interested in all three. I think (2) is probably the most separated (embedded mode does exist), but (1) and (3) are concerning to me if DBZ is intended to exist outside of Kafka.

The intent of this post is really to get an understanding of the vision–is DBZ intended to be separate from Kafka? If so, do the project leads view the current dependencies as problematic? If so, what should be done to address these dependencies?