What is Science?

This is a very interesting article on the definition of science. 

A little history: The first proto-scientist was the Greek intellectual Aristotle, who wrote many manuals of his observations of the natural world and who also was the first person to propose a systematic epistemology, i.e., a philosophy of what science is and how people should go about it. Aristotle's definition of science became famous in its Latin translation as: rerum cognoscere causas, or, "knowledge of the ultimate causes of things." For this, you can often see in manuals Aristotle described as the Father of Science.
The problem with that is that it's absolutely not true. Aristotelian "science" was a major setback for all of human civilization. For Aristotle, science started with empirical investigation and then used theoretical speculation to decide what things are caused by.
What we now know as the "scientific revolution" was a repudiation of Aristotle: science, not as knowledge of the ultimate causes of things but as the production of reliable predictive rules through controlled experimentation.

This is somewhat hand in hand with the history of 'Le Cid' in which Corneille got grief for going outside of Aristotle's rules.

Galileo disproved Aristotle's "demonstration" that heavier objects should fall faster than light ones by creating a subtle controlled experiment (contrary to legend, he did not simply drop two objects from the Tower of Pisa). What was so important about this Galileo Moment was not that Galileo was right and Aristotle wrong; what was so important was how Galileo proved Aristotle wrong: through experiment.
This method of doing science was then formalized by one of the greatest thinkers in history, Francis Bacon. What distinguishes modern science from other forms of knowledge such as philosophy is that it explicitly forsakes abstract reasoning about the ultimate causes of things and instead tests empirical theories through controlled investigation. Science is not the pursuit of capital-T Truth. It's a form of engineering — of trial by error. Scientific knowledge is not "true" knowledge, since it is knowledge about only specific empirical propositions — which is always, at least in theory, subject to further disproof by further experiment. Many people are surprised to hear this, but the founder of modern science says it. Bacon, who had a career in politics and was an experienced manager, actually wrote that scientists would have to be misled into thinking science is a pursuit of the truth, so that they will be dedicated to their work, even though it is not.

Even though I've quoted at length, the full article is worth reading, especially for the comparison to modern approaches.  As far as the critique of Aristotle goes, Bertrand Russell said something similar about the launch of the Italian renaissance.  The rediscovery of old texts allowed readers to choose between Aristotle's insights and those of Plato (and others).  This opened up intellectual space for challenges to settled dictum.  New things were discovered, new avenues were explored and things absolutely blossomed. 

I do wonder how well we do this today.  I'd argue that once a scientific idea gets politicized, it is very hard to keep it in the scientific realm.  Each camp looks for signals of agreement rather than testing for evidence.  The only way that this can change is by a revolt amongst scientists and I don't see that happening any time soon.
I also wonder about the art world.  Makers of art have never had more freedom of theory to pursue their ends.  In fact, the idea that there is a limit to what may be described as art is wildly unpopular.  I don't know that this has led to a better overall universe of art.  In the museums of the year 3000, will there be more art from 1700's or the 1900's? 
I've got my suspicions but if I'm right, what does that say about the usefulness of rules?