Why does Kaspa need a block byte-size limit on top of its mass system?

Kaspa's compute mass and storage mass control resource consumption within blocks, but neither one directly caps how large a block can be in raw bytes. Even a block that stays within both mass limits could still be very large in byte terms, driving up both the storage space each node must keep on disk and the bandwidth consumed every time a block is broadcast across the network. A separate byte-size limit gives Kaspa developers finer control over these worst-case costs, ensuring that nodes running on modest hardware are not silently priced out of the network.