diff --git a/smocore/marionette/lifetime.cpp b/smocore/marionette/lifetime.cpp
index 87b4460..7c36708 100644
--- a/smocore/marionette/lifetime.cpp
+++ b/smocore/marionette/lifetime.cpp
@@ -76,6 +76,19 @@ public:
 				+ ": Must be executed on Marionette thread");
 		}
 
+		/**	FIXME:
+		 * It may be necessary to add a delay here to ensure that all in-flight
+		 * timer timeouts have finished executing? Or some other mechanism.
+		 *
+		 * We need some way to ensure that in-flight timeouts don't get fired
+		 * during the finalize sequence. This is because they may depend on
+		 * state that is being finalized or has been finalized at the point
+		 * when they timeout.
+		 *
+		 * This seems to be actually happening with the delayed calls to
+		 * AttachDeviceReq::attachDeviceReq2() inside of livoxGen1.cpp.
+		 */
+
 		device::DeviceManager::getInstance().finalizeDeviceReattacher();
 
 		smo::mind::globalMind->finalizeReq({context, std::bind(
diff --git a/todo b/todo
index 75d508f..95e8fdd 100644
--- a/todo
+++ b/todo
@@ -25,3 +25,7 @@
   after they expire just in case boost doesn't clean up the internal
   callable that we passed it. Or else we'll have circular sh_ptr
   references in our continuations.
+* Think of a unified mechanism or a design pattern which will ensure that
+  timeouts will always either expire or be canceled before program
+  finalization.
+  Or more generically, before their lifetime ends.