The Sky is High
So the general marketing plan would be for the Flexible System Architecture to standardize the home computer market by gradually taking over the sub-processing within it. The reader may assume that the final step will be to put the whole Intel architecture into the custom section. That is certainly a possibility, but the overall aim would be even more engulfing than that: the creation of a Hardware-based Operating System Standard (HOSS) utilizing the FSA.
It can be said that all operating systems are hardware based, since they consist of programs which ultimately run on a physical computer, be it micro, mini, or mainframe. But they are not closely coupled to the underlying hardware, and that will be the difference in future computers running the FSA HOSS. This isn't the place to get into a heavily technical discussion, so all I will say here is that in the implementation of this future HOSS, this architecture will, in a very real sense, BE the operating system.
This doesn't mean attacking UNIX, Windows, or any other operating system head on. This HOSS would be an "operating system for operating systems," just as the FSA is the "architecture for architectures." Many of the more powerful workstations offer the user a choice of what OS to run, and the HOSS will lend itself to this approach by being in top-level control of the resources of a computing system.
A more technically minded reader might now be thinking, "He's talking about a scheduler. Any old processor can act as a scheduler." Maybe in its earliest forms, the HOSS will serve in such a capacity, but the FSA has power enough not to be limited to this role. It provides a path to true reconfigurable computing.
Any chip implementing the architecture, whether standing alone or having extra circuits in its custom section, would retain the standard section for fast, seamless communication within a larger system. A later, more mature HOSS would be able to use this structuring to aid in gathering together the necessary combinations of sub-processors to perform each new task thrown at it quickly and flexibly.
Another strong advantage of close coupling to hardware is that of better computer security. Nothing can guarantee complete protection from hackers and spoofers and viruses, but I believe a huge improvement can be obtained if certain features of this architecture are intelligently applied. Once again, this is not the place to go into technical detail; all I will add here is that security is a very marketable commodity - and therefore rather useful in a prospective industry standard.
There's really no predicting what applications may be developed for the future HCs of the world, and no predicting how complex may be their interactions. Power stations are not designed to handle the average use of electricity, but the peak load. Similarly, for the peak loads computers may have to face in the future, it would be best to have a powerful infrastructure already in place: the FSA HOSS. Perhaps, since it may ultimately control so much, it would be better to call this technology a Hardware Network Control System, rather than a hardware OS, but its name is not as important as its capabilities.
Eventually the industry should be able to do away with plug-in boards altogether and move to simple add-in devices, just as today we have add- in memories. All in all, a more uniform, cost-effective approach, one which should result in systems made up of both fewer parts and fewer different parts. Such modules will be quite cheap if they are all kept as much alike as possible, based on a single overall design.
I could go on, but this is getting into ideas I would rather discuss in confidence. I will reiterate that I envision that the ultimate physical basis of computing will consist of pluggable modules. Users will eventually add processing power as easily as they upgrade memory today. Whether these modules will be based on the Flexible System Architecture or some other is moot, but I do not see the current Intel design as being suited to the role.
On the system software side, I, like many others, think the various Microsoft Windows offerings are an overloaded barge sitting low in the water. I cannot predict if it will tip over and sink, but overly complex systems do eventually get replaced. I am not a wild eyed anti-"Wintel" fanatic; the point I'm trying to make is that changes are always happening in the computer field, and whether my predictions come true or not, I want to see this adaptable microengine both technically and financially positioned to move on major opportunities as they do occur.
I freely admit to being very ambitious concerning this architecture. My personal mission is to see it become an industry standard, rather than merely a player in this or that market segment.
As I mentioned at the outset, much of the engineering effort to date has been applied towards the creation of a software simulation of the FSA; a precursor of a development system, which can also be offered for sale. Just as importantly, such a simulation is an invaluable tool for support of continuing design of any computer architecture, and I wanted to bring this one to a level of sophistication whereby a team of engineers could start using it immediately to that end.
As for the architecture itself, It currently has "patent pending" status by way of a 124 page technical description submitted as Provisional Patent Application, and which lays out the basic concepts and the guidelines for future development. The groundwork is thus finished, but there is still much to do, which is good; a project at this stage should inspire the best engineers to do their best work by offering them chances to be truly creative.
As for me, had I not taken this sidetrack into entrepreneurship, I'm sure I would have worked my way into engineering management by now. Design is still satisfying, but I continue to feel a growing desire to build and create through a human organization; I want to actively help spearhead the FSA's growth to its ultimate destiny.