our FR5 lathe interface has two meanings for ready

Yes, it can accept the request and wait. The integrator's current sketch treats that first response as though the wait is over. That's the part I want corrected before it reaches the operator sheet.

Then it is a sequence disagreement, not merely untidy names. Show the waiting state explicitly. The application needs to know a request exists without assuming the loading step can proceed.

Isaac, who writes and clears each proposed signal, because two names won't help if both programs think the other side clears them?

Put Gabriel's ownership question in a shared state table with the events. Include what makes each state become true and what ends it, not just arrows between boxes.

The authors are working through that table. Their first version covers request receipt and later availability separately, but clearing after a withdrawn request is still disputed. Safety functions remain separate from these ordinary job states.

Use a withdrawn request as the next example, then. What should each side know when the sender no longer wants the original job?

And a lost connection after receipt. Don't let reconnect turn a remembered request into permission for a new one. That needs agreement while both authors are still looking at the same table.

David, yes. The normal sequence is the easy bit. I'd rather leave recovery marked unfinished than squeeze it under a vague reset arrow so the drawing looks complete.

What does the draft screen show while the received request is still waiting, Isaac? That seems like a cheap check with an operator now.

It says request received, waiting for loading-area availability. Operator understood that distinction, but asked who to call when the wait continues. We've added the support role. Withdrawal and reconnect rules are still being agreed.

Thanks for reporting the operator's question. A technically correct waiting state can still leave someone standing there wondering whether anyone else knows about it.

Does the support role have the request identity on that screen? Otherwise the first thing they do is ask which of several waits this is.

Yes, the active request reference is in details with the selected sleeve job. I have not called the interface finished; the two unresolved recovery cases are visible in the handover draft.

Good place to leave the draft. Please post what they agree for withdrawal; that's often where a nice normal sequence starts telling two stories again.