Mitochondrial Linux

I have been experimenting with blender, vrml, mmorpg, and scripts as emulations of material systems, physics and AI. I also have been studying ROM Linux in the form of a system core in the BIOS of a computer. I have also been studying self-replicating systems that are open source. I have decided to combine them with my knowledge of genetics and chemistry to produce a self replicating chemical system based upon DNA and RNA and proteins. I realized 17 years ago that it was possible to modify organisms to produce oil ( or any other chemical ) as a product in the same way that Genentech creates human insulin by inserting a plasmid or sequence in the main genome of a single celled organism. Brewers, bakers and many others have been using biological organisms in this fashion for thousands of years. It has only been recently that the degree of product and structure control has reached such a fine detail. The techniques are well established and actually very easy. They require very little equipment and many readily available components as they are derived from naturally occurring materials such as restriction enzymes which are produced by high temperature bacteria (Taq) .

I have decided to make a processor which has a POSIX core implemented in a self replicating 3D parallel processor composed of a modified mitochondrion in a prokaryote. I am just making the choices of which methods to use, but it seems a much wiser choice for a core mechanism ( than entropic semiconductors ) as it allows for molecular level memory and a content addressable format which is self powering. I would implement some proprietary or unique structure, but the amount of functionality which could be included immediately from the standard core would allow for a rapid scale up. There are some obvious problems with the implementation of some interfaces and mass storage,temporary memory, or computation is not a problem, but some applications rely upon techniques and process that may not be possible to emulate. I think it is a good valid step toward the open source concept of existence as it allows for the self assembly of a wide variety of chemicals and structures without a specialized or entropic controlled system. I think it is the most natural progression of technology and allows for data retention which conforms to life itself and would operate anywhere life existed. It is possible to create a great number of useful structures that could not have evolved due to the very nature of the continuity of the biological system itself.

The biggest advantage of this approach is that information transferred between developers could become a physical reality and the need for continuous upgrade would be unnecessary.

I was worried that I might not be able to implement floating point, but I think I can introduce it as an analog process. I have noticed some very organic like responses from my FPU recently anyway. The FPU seems to have some form of indeterminance anyway.

You might think that such a machine would be slow, but the degree of parallelism and the incorporation of quantum level analog comparison and retrieval compensates for this in scale. It is not necessary to use the same ionic shift mechanisms of K and Na to transport electrons. In fact it could easily be done at near light speeds with a single molecule and even transport from light to electron in the same way as photopsins. The bioluminescence that is inherent in organisms and prevalent in nature could serve in parallel to represent information at the molecular level. I have incorporated this gene in the lab before. The skin of an octupus can not only represent any complex image, but actually perceive it in parallel with it's presentation.