The scientific method is based on the idea that physical objects or events will always behave in the same way under the same conditions. Thus, a scientific law is universal. H20 boils at 100 C without exception but only if the vapor pressure equals the atmospheric pressure and only if there are no minerals or other impurities in the water that affect its boiling point.
Thus in order to observe the operation of a scientific law, the conditions under which it operates must be repeatable and that means identical in all relevant respects. This is, of course, the challenge in experimental design. To clearly observe universal causal relationships between two entities or events that are to be analyzed, all variables that might interfere in that relationship must eliminated or controlled and accounted for. The unambiguous observation of a scientific law depends on isolating it from its “real world” conditions.
And in order to guarantee the repeatability of experimental conditions, all these causal factors must be discreet and clearly distinguishable so they must be represented as precise, measurable quantities. The conditions under which an experiment takes place must be mathematically determinable so that we can judge whether those conditions under repetition are precisely the same.
Thus, although scientific laws are universal, they are also hypothetical. They apply only if the circumstances under which the law can operate have been repeated. In reality, even small differences can disrupt the operation of a natural law.
The development of technology, although it makes use of scientific discoveries, works in the opposing direction. For technological advances, newly discovered scientific laws must be made to operate in the real world and that means they will be deployed in contexts where conditions might differ significantly from those in the laboratory and cannot be comprehensively controlled. While science depends on eliminating accidents, technology depends on anticipating them and building in tolerances and flexibility to allow for varying conditions of use.
Our bridges stay up not only because we’ve discovered the relevant scientific laws that govern bridges but because engineers are good at dealing with the complexities and uncertainties of the real world.
There is a tendency in the modern world to model our concepts of morality and practical reason, as well as our social inquiries, on natural science. (Kant was quite explicitly doing this with his notion of the moral law)
This is a bad idea. In the real world in which we must act, conditions are never exactly repeatable and we can never isolate all the variables that might influence a situation. Acting with too much regularity leaves us inflexible and unable to respond to change. The analogy with science which persists in moral philosophy has made moral theory largely irrelevant.