All projects and organizations are made up of people, and people are political enti- ties. There are egos based on subjective views of one’s personal worth, qualifica- tions, and status within a project. There are territorial disputes between equals, managers, and systems engineers, all of whom establish technical and political territories, and then often succeed in acquiring the territory of a neighboring
subsystem. This happens all the time, and it is part of life, but it also sends costs through the roof. It demoralizes intelligent young people who don’t understand the fundamental stupidity and selfishness on part of the ‘‘techno-warlords,’’ ‘‘feather merchants,’’ and charlatans who have nothing to offer but political infighting and trouble.
This is an area that is fraught with problems. Often, to see what is inhibiting progress, the manager must makes his rounds, listening and watching as he goes about the offices of his team. He should not get anyone to complain and point fingers, because the subordinate managers are warlords and will hold a grudge.
Besides, many warlords are outstanding technically and you need them to get the job done; everybody has flaws, after all. Such unresolved issues only serve to alienate a young person with little experience from the team leader and the team.
What the manager does is establish and re-establish policy; he must pay careful attention that this is being complied with. If that doesn’t work, you must resort to removing the obstacles to getting the job done.
RESOLVING PROBLEMS AND IMPASSES 135
11 The Development of System Architectures
One of the most interesting discoveries I have made during my career working in computer science is the role played by philosophy in architecture. Specifically, there are philosophical aspects to the ways in which people think, perceive, listen, decide, and work together toward a common goal. Architecture is more than a mere physical representation of the structure of a system; it is a projection of the archi- tect’s vision of the subject, the hypothesis of what he perceives the architecture to be. It is a subjective impression on the architect’s mind, and far from what the objective design will be in the end.
The architecture of a building or a classical temple is simple compared to that of software. The form, proportion, and size of a structure are limited by the materials available to carry the load. Since a model can easily be constructed, the vision of the architect can quickly be implemented to scale and shared with others. The customer can make comments and ask for refinements in a process that ‘‘adds substance to the object,’’ which also helps the architect gain a deeper understanding of what he is try- ing to accomplish. Sometimes the vision is incomplete. It is only the initial incarna- tion of an idea, after all. When a physical model is actually built, it may surprise the architect, for it may look far different from what he had imagined.
I can only ask myself how others think and envision what they build. I have in reality not the foggiest idea of how to do so unless I can visualize the product of their minds. Architecture is a thing of harmony and beauty, and it is a product of the mind, like a picture or a painting. A photograph represents an event in time, as seen through a lens. It may have the photographer’s imprint in a combination of his abilities in composition, light, contrast, and subject. However, it still is an event captured. One can imagine Ansel Adams waiting for a month or two to capture the exact sunset that was on his mind for years. A painting is different; philosophi- cally it represents the artist’s interpretation of what he sees and feels, if he possesses the talent to combine them into a picture.
As a young boy, I often wondered why Pablo Picasso made that terrible painting of Guernica. I hated cubism. Then, as time went on, and I read and reread the great classic philosophers, I began to understand a few things about this school of art that
The Cognitive Dynamics of Computer Science: Cost-Effective Large Scale Software Development, by Szabolcs Michael de Gyurky
Copyright#2006 by John Wiley & Sons, Inc.
136
Pablo had founded. After my maturing in Vietnam, and adding to that my childhood experiences during World War II, I began to understand what he was saying with that painting. He was unable in the classical style available to him to capture all of the event of Guernica: the scene, the terror, the individual pain, and above all, the scream of agony of a town under attack. He painted not from the visual rubble, ruin, and dead bodies, but from a part of his mind calledinner sensingin Kantian philo- sophy. This is where the impact of the external sensing function is converted into mathematical format for processing and sent to a function I think of asabstraction.
Here, in the abstraction, the mathematics is reconstructed by combining the algebra, geometry, sound, smell, motion, mass, and so on for further processing, analysis, and storage. I am not saying that I am completely correct, but it is because of my understanding that I place high value on architectural design and how it works, and understand why it often doesn’t work. Obviously, when it doesn’t work, it has a great impact on cost.
This phenomenon influences not only our understanding of how we visualize, but how we perceive and interpret an object. Not being able to understand some- thing does not mean that it is not correct; all it means is that our ability to under- stand another’s vision is limited. You could say that it is a limitation imposed by IQ, but that is unfair. IQ is a flaky subject; the results of IQ tests can be influenced by culture, language, and age. An idiot savant may be able to do sheer magic with the- oretical mathematics, but have no other talents whatsoever. I have known several individuals like that. Dismissing what a subordinate engineer or scientist is saying because you do not understand him puts the entire architecture in jeopardy, schedule- wise and budget-wise.