A Model of Scientific Evolution

---

Example of Scientific Evolution

Light (and Vision)

Pre-Paradigm Stage:  Historical Theories of Light and Vision
	Before 1700. There is no underlying principle by which to 
	judge any particular theory.
		- All objects somehow emit light themselves.
		- The eyes project light onto objects.
		- The "form" of an object is projected to the 
		  mind via the eyes.
		- ...
	After 1700: The emergence of two strongly supported paradigms 
	for the nature of light.  Both are validated through experimental 
	evidence.
		- The corpuscular theory.
		- The wave theory.
			Required that an ether exist throughout the 
             		universe.  The Michelson-Morley experiment 
             		proved that this ether did not exist.
	Post-Michelson-Morley
		- The wave-particle duality of light becomes the accepted
		  paradigm.  At this point, the articulation of this 
		  paradigm is "normal science."

---

Resistance to Change

The Paradoxical Nature of the Human Mind

Switched-Color Playing Cards Experiment

	Evidence of the mind's occasional inability to assimilate 
	changes in the commonplace.

	The behavior of the group of test subjects is analogous to 
	the behavior of the members of a community undergoing a 
	revolution - e.g., Einstein's reluctance to accept quantum
	mechanics.

Inversion Goggles Experiment

	Evidence that the mind can, with great regularity, 
	assimilate changes in the commonplace.




"Though psychological experiments are suggestive, they cannot, 
 in the nature of the case, be more than that.  They do display 
 characteristics of perception that could be central to
 scientific development, but they do not demonstrate that the 
 careful and controlled observation exercised by the research 
 scientist at all partakes of those characteristics."

---

Implications for Software Engineering

Is software engineering a science?

	Scientific communities are able to reach a "firm 
	consensus unattainable in other fields."

	Multiple programming paradigms:

		- Procedural
		- Functional
		- Object-Oriented
		- Logical

Does it matter whether or not software engineering ever becomes a 
science by Kuhn's definition?

	Some potential benefits of working from single paradigms in 
	terms of software engineering:

		- Consistency in all aspects of the design process.

			- Clients know what to expect.
			- Reusable code repository.
			- Reduce possibilities for confusion and error.
			- Enhanced productivity and quality.
			- Reduce production costs.
			- Elimination of system compatibility issues.














© 1997 by Slimy - What do you want to eat today?[TM]

SlimyLeapfrogHamburger®

Don't even think about using my trademark.[TM]