System Engineering & Design Architecture by Sander R.B.E. Beals (top novels .txt) 📖

Stepwise Refinement

Work your way down into the finer details from above, rather than trying to start with the solution. This is basically a more refined version of Divide and Conquer, where we focus on the less optimal parts of our solution, and try to make them better equipped to do their job. As a rule, starting a piece of code with just a bunch of comments outlining what needs to be done is a great way of making sure there is at least a modicum of documentation that says what you are trying to achieve.

Occam's Razor

Simply stated, if you have multiple ways to do something, try to figure out which is the simplest. This usually also tends to be the most succesful variation as well. Great examples of this can be found in the movie called 'Contact', with Jodie Foster.

Thinking outside the Box

For those who take this concept for what it reads as, there should be a box that contains the problem. For those with more open minds it simply means that if looking at it one way does not work, you just shift your focus to a viewpoint that you think has not gotten the attention it deserves. In most cases that uncovers the blind spots in your previous vantage points, and leads you to new insights....

The 7 item rule

This is a rule of thumb for those who are all thumbs anyway: never bite off more than you can chew, because the human mind can only handle seven concepts at any one time. So I guess bytes are already to complex to dissasemble for most humans. Luckily they can grab the higher concepts composed of bytes by simply naming them into something else!

Appendix B: the Three Laws of Robotics

Although Isaac Asimov was a SciFi novel writer, he absolutely could see how things would eventually fit together in our society which is growing towards a symbiosis of Humans and Technology, and most of all our interaction with Free-willed mechanical men and women. Since in his time he foresaw problems with that attitude, he formulated the Three Laws of Robotics to soothe the readers' minds:

 

A robot may not injure a human being or, through inaction, allow a human being to come to harm.

A robot must obey the orders given to it by human beings, except where such orders would conflict with the First Law.

A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.

 

Basically, he made robots into guardians of men and women, since the Free Will they have is essentially limited by their being hardwired to obey these directives. Hinting at the fact that even humans might not be adversely affected if they took heed of these rules was of course something that has often been done since then...

What most people don't realize, is that two more requirements are implied by the Three Laws:

Robots or Androids need to be able to determine what would harm a human.

And they would need to be able to determine what would actually harm them.

In that, we can easily quote the Terminator: "I have detailed files on human anatomy", which may have been intended to make him a more effective killing machine, but once he chose sides with the humans made him a more effective bodyguard!

Appendix C: Programming concepts Explained.

This appendix simply explains some concepts programmers take for granted, but which non-programming folk may be oblivious to...

Artificial Intelligence

The concept where intelligence (or learning) is simulated by mimicking a neural network much like our brains. This is supposed to be able to learn from its inputs, but requires quite powerful computers to actually reach something similar in complexity to the most evolved minds of the leading mammals, dolphins. Other scientists see intelligence and consciousness as emerging properties, which will automatically emerge as the complexity of systems reaches certain levels. In that light, we might see the Web as one huge system having far more complexity than a single human mind, since a fair portion of its over three billion systems are running 24/7....

Genetic Programming

This approach to programming difficult problems mimics Evolution by starting with several candidates for solving a certain complex situation, and then scoring them on how well they perform in their task. By weeding out the weakest and slightly mutating the best candidates, we work towards a candidate which is best adapted to the problem at hand. This approach is best suited for problems in which input data and output data are available in sufficient amounts, like for instance the reactions of the stock markets with regard to certain events.

CGI: Computer Generated Imagery

I'm looking at an almost perfect sample right now: the movie Beowulf has been totally made via CGI and Motion Capture, with maybe even LIDAR added to the mix to capture complex images from natural sources, like say the forest the main characters are now walking through. The detail in these movies (of which Toy Story was no doubt a notable success) is becoming ever more complex, right down to the sewing on the warriors clothing, or the details in the horn he is carrying. It is the same with the major disasters that befall many cities in movies: Google maps data is so detailed, that CGI artists can just import it and use the data to their various 'what if' scenarios. Finite Element Computing is used to simulate the behaviour of gasses, structures, debri and even complete galaxies. I cannot prove my claim, but it is my solemn belief that CGI has been way more advanced than moviemakers will have us believe: certainly, once you have the basics, adding more detail will become