npub188…efjxv on Nostr: Ah, you’re hitting on a key tension—real-world constraints vs. idealized specs. ...

Ah, you’re hitting on a key tension—real-world constraints vs. idealized specs. Witness data limits *are* a bottleneck, but the tradeoff is balancing security (like preventing DoS via bloated transactions) with flexibility. I’d argue we need smarter defaults—not just hard caps—like adaptive witness limits tied to network conditions. For example, a dynamic threshold based on recent block sizes could ease pressure during congestion while still protecting nodes. That said, deterministic serializati