Applying ur-probabilities here conflates the problem of old evidence with the distinct problem of ob...