I'm fine with technical definitions of words such as "information." It's just that we should be clear on the difference between those and the connotations of the natural language meaning of the word. In so far as information is natural and not artificially programmed, the "signal" is equivalent to an effect that indicates some properties of its cause. This is the sense in which smoke carries the natural meaning of fire because one causes the other. But if natural information is no more than the probability involved in causation, why set us up for being misled by the anthropocentric connotations of "information"? Why dress up causality in this way to make it seem as though natural systems were intelligently programmed?

Science is used to misleading itself in this way, as when Newton spoke in clockwork terms about natural "mechanisms," which supported his deism. Metaphors can interfere with technical scientific work, which is why scientists should steer clear of natural language concepts. I suspect there's a big difference here between the science of information theory itself and the popular descriptions of the theory.