LG's latest offers a pretty reasonable example of how this will play out for many.
We find that the conventional Barnes-Hut MAC can introduce potentially unbounded errors unless θ < 1/3, and that this behavior while rare, is demonstrable in astrophysically reasonable examples.
This means that if someone prosecuted for either offence can bring some evidence to raise an issue that they have a reasonable excuse (for example if they work as a photographer, or can show some good reason for taking a photograph), then the burden of proof shifts and it is for the prosecution to prove, beyond reasonable doubt, that that reasonable excuse does not exist.
This is a reasonable description, for example, of a point source that moves on a spherical surface located at distance R from the target of a radiation treatment.
During that week, they position NRA-supporting politicians to stake out an extreme goal of the gun lobby and claim it as one end of a reasonable argument (for example, Gov. Robert F. McDonnell said Virginia should consider "arming teachers, principals and other school staff").
In some cases this is a reasonable inference since, for example, several tyrannosaurs are known with feathers, strongly implying that even Tyrannosaurus had them, but what about more distant relatives?
Trying to eliminate an infectious agent by eradicating its carrier is a tall order unless the insect population is actually only hanging on by its fingertips within an essentially hostile or variable climate, or if there is a reasonable infrastructure to, for example, keep marshes continuously drained.
In the religious sphere for example a reasonable pluralism might contain a reasonable Catholicism, a reasonable interpretation of Islam, a reasonable atheism, and so on.
Thus good results can be achieved only by starting with a "reasonable" initial HMM, for example assigning very low probabilities to certain transitions (such as determiner → determiner, determiner → verb, adjective → verb).
