Some Myths about Art, Science, and Software

One of the most destructive myths in the industry is that somehow science and art are mutually exclusive, that scientists and engineers cannot think creatively and that artists cannot or should not be educated to be competent in measurement or science because it would hurt their creativity.

As previously noted, feature richness is a major component of the perceived quality of a software product. A rich set of features requires a creative and innovative approach to design. This requirement has led to an infusion of artistic talent into the software development process. Typically, this artistic talent lacks foundation training in science or engineering. Even worse, the Internet opened the door for absolutely anyone to put up a Web site and claim to be a professional-anything.

Myths about Art

Art means to join or fit together. It is commonly defined as follows:

1. The human ability to make things; creativity of human beings as distinguished from the world of nature.

2. Skill; craftsmanship.

3. Any specific skill or its application, for example, the art of making friends.

4. Any craft, trade, or profession or its principles.

5. Creative work or its principles; such as making things in a manner that displays form, beauty, and unusual perception: art includes painting, sculpture, architecture, music, literature, drama, dance, etc.

6. The quality of being artful or cunning.

7. The quality of being sly or cunning; tricky; wily.

The infusion of artistic talent into software development has undoubtedly improved features. It has also caused serious quality degradation in the software development process. One of the most popular myths in the software development community is that only artists are creative and that artists cannot be creative if they are constrained to follow rules.

I spent 25 years in the performing arts, and I am an artist as well as an engineer. No one drills harder on fundamentals than an artist. Musicians, dancers, painters, singers, and writers all practice for hours each day for many years to build something called technique. The depth and quality of an artist's technique is the measure of that individual's mastery of his or her discipline. An artist cannot be competitive without great technique. Because of my own experience with art and science, I am very wary of the person who claims to be an artist but whines about following a disciplined approach. This person is not likely to be a competent artist.

Myths about Science

Scientific methods grow out of a body of factual knowledge, not out of myth and supposition. The term science refers to a body of knowledge that is a body of models and generalizations that organizes and correlates observed facts. The purpose of gathering these facts is to make predictions. These predictions are then tested by comparing them to actual observations or experiments.

An accepted scientific conceptual scheme is usually called a theory. A theory is never proved. A theory is considered to be a valid model of reality if it correlates well with a considerable body of facts and if no one disproves it by finding a fact that contradicts its predictions. There is always the chance that the best-established theory will be disproved by someone who can show its predictions to be in error. At this point the theory will have to be revised, corrected, refined, or abandoned.

Science starts by systematically recording facts. These facts should be accurate, well defined, and quantitative. The scientist tries to order and correlate these facts. The product of this process is usually a working hypothesis that is capable of prediction. A working hypothesis is provisional. It is meant to guide further investigation and is subject to constant change and refinement. A hypothesis is incorporated into scientific theory only when it has been empirically verified in many ways. Any hypothesis is abandoned or revised if its predictions are contradicted by observation.

When a scientist publishes a new idea, it is called an invention or innovation. Invent means to come upon, meet, or discover. This definition includes (1) to think up; devise or fabricate in the mind such as to invent excuses; and (2) to think out or produce, such as a new device, process, etc.; or to originate, as by experiment; or devise for the first time. According to Webster's New World Dictionary, innovate means to renew, or to alter, to introduce new methods or devices to bring in as an innovation. These definitions certainly sound like creativity.

Myths about Software

The popular myth is that the artist just throws a bunch of stuff together and comes up with a miracle of creation. This is not to say that the scientist cannot be creative, ingenious, or inventive, nor that the artist is undisciplined and cannot keep good records. But that is the myth. If you are inventing something new, you use a process called experimentation. In commercial software, we call this process developing a new product. How the experiment is conducted or the development process works is what separates the scientists from the artists.

While the artist is likely to rush right in and begin following her or his creative intuition, the scientist begins by performing background research and formulating an approach to conducting the experiment. The scientist keeps careful records of the steps in the process, the assumptions, the ingredients used, and their quantities and the outcome.

Pure art is good at one-of-a-kind things and limited editions. Creation is about taking risks to try new and different things. Success is capricious. Uniformity and reproducibility are not the hallmark of art. Most of an artist's career goes into trying to find the spotlight-fabulous painting, hit song, best-selling book, or whatever. In the commercial world we need reproducible miracles. The next release of the software should be even better than the current release; art makes no guarantees.

For example, let's consider cookies. Great software and great cookies have a lot in common. They both evolve through a process. A business needs not only to develop a good cookie but it also needs to be able to make as many of these cookies as the market demands. In addition, everyone who buys a cookie wants a good cookie, so the cookies must be consistently good, batch after batch. The maker needs to be able to make lots of good cookies time after time without unpleasant surprises. It is generally not worthwhile for a business to create one batch of fabulous cookies if it cannot ever reproduce them for sale in the future.

The artist may invent a truly marvelous cookie, but there is no guarantee that it can be reproduced with the same quality. The cookie invented by the scientist may not be so marvelous, but it will most probably be reproducible with consistent quality.

Additionally, artists tend to ignore requirements that do not agree with their artistic sensibilities. So, it is not unusual for the marvelous cookie that the artist created to be a poor fit for the cookie-buying market. While it is true that a scientist's creativity may be inhibited by his or her knowledge of the facts, the artist's creation can be crippled by ignorance of the facts.

Fact: 

What is required is a balance between creativity and method. This is where engineers come into the picture.