Page 248 - Modern Spatiotemporal Geostatistics

1 2

POPULAR METHODS IN THE LIGHT
OF MODERN SPATIOTEMPORAL
GEOSTATISTICS

"In science one must search for ideas. If there are no ideas, there is
no science. A knowledge of facts is only valuable in so far as facts
conceal ideas: facts without ideas are just the sweepings of the

brain and the memory." V.G . Belinski i

The Generalization Power of BME

In the detective story "Adventure of the Sussex Vampire," Sherlock Holmes
makes the following remark:
One forms provisional theories and waits for time and fuller
knowledge to explore them. A bad habit, Mr. Ferguson, but human
nature is weak.
In terms of the BME epistemic paradigm, the first part of Holmes' remark,
"One forms provisional theories...," clearly refers to general knowledge at the
prior stage, while the second part, "...waits for time and fuller knowledge
to explore them," refers to specificatory knowledge at the integration (pos-
terior) stage. Building on such a logical distinction between general and
specificatory knowledge, one of the basic BME postulates is that two basic
concepts—informativeness and probabiliorism—are both desirable features
of space/time analysis, which refer, though, to different stages of the epistemic
process and are associated with different goals of scientific investigation. While
informativeness is associated with the goal of building a model that expresses
the available general knowledge base Q at the prior stage and is maximally
informative (in a well-defined mathematical sense), probabiliorism is associ-
ated with the aim of producing at the integration (posterior) stage space/time

229