But we must never forget, despite all of its staggering usefulness, the computer is still only a human engineered accomplishment. And when viewed from this perspective, the fundamental principles of engineering apply to it just as they apply to all of the other man-made products. And one of the most important truths in engineering is that, in the end, you eventually have to
'throw away your first-generation designs and start again'.
There are no exceptions to this. All long-term products must pass through this often painful process; that is, of course, if you wish to continue to improve them into the indefinite future.
But there are two ways that you can live in accordance with this engineering truth: you can voluntarily throw away your first designs after you have had a successful first-generation product, and 'start again'. This, for most people, including most engineers, is the 'hard-path'.
Then there is the 'easy-path'. You can try to expand upon and grow your product on first generation designs. But in the end, the short-comings that are always inherent in first-generation designs will lead to more and more problems in the 'product' that eventually 'force' you to change if you wish to continue forwards. And in most situations, this occurs when a 'crisis' is reached.
Unfortunately, to date, the vast majority of computers in use are, at both the hardware and software levels, based on 'first-generation' designs; designs that should have been retired years ago. The only large-scale exception to this has been with chip fabricators; those individuals that make the physical circuits in which computers operate. These individuals long ago threw out their first-generation designs and started-over, several times. That is why they are now making electronic components that are as little as 5 atoms thick.
It is time for the rest of the 'computer world' to follow the fabricators lead; to put away 'first-generation' designs and move on to 'next-generation' products.
The question is: how to do this?
At first consideration, it seems that fundamental change to the computer world is near un-to-impossible. The overall short-term 'cost' of replacing all of the computers is to staggeringly high.
It is true that computers have become so wide-spread that overnight change is not possible. However, one must never forget that the turnover in this industry is extremely high ( most computers are gone in four to ten years after creation) . And with fast turn over can come, when done correctly, fast change.
But even with fast-turnover, throwing away of all of the first-generation designed computer-products at both the hardware and software level cannot occur simultaneously. There are just too many of them. Rather, next-generation products will need, at first, to be introduced one at a time. Yet if this introduction process is done correctly, the first round of second-generation products will lead to an even easier introduction of the next set of advanced products which will, in their turn, will accelerate the next round, as so on.
In this way, the chain of next-generation products could be introduced and made wide-spread in a very much shorter time than what most people might think.
The question is: what is the optimal chain of product introductions? The next post will answer this question.
