Easier for you maybe. More than half of my Design Patterns use a Thread::sleep. I bet the majority of the users out there are currently using Thread::sleep. It is less work for the maintainers but massively more work (in aggregate) for the users.
Also, eliminating it from the docs is not sufficient. I’m not sure it is even in the current docs. It needs to be added to the docs in big bold letters:
Do Not Use Thread::sleep() in Rules.
or some other appropriately worded statement.
A breaking change announcement also needs to be made to inform the users of this change.
Finally, it would need to be added to the change log for 2.2 when it gets released.
Ideally, the Rules DSL itself would be modified to not allow Thread::sleep(), though that might be a major headache if it is even possible.
See the first example rule in the Working with Groups in Rules Design Pattern for an example.
If you are not on a super fast machine persistence takes a little bit of time to catch up so that lastUpdate returns the actual most recent update time. The amount of time required depends on the system OH is running on. Your system may be fast enough that persistence is able to keep up without a sleep.