Wikipedia has an article on biological computing which describes some methods which have been implemented using DNA and other chemicals and enzymes. Link here. There are so many ways to achieve the goal that it is difficult to say what is the best approach. Like computer languages, electronic hardware implementation, and organisms with their associated advantages and disadvantages.
I would say that scalability, speed, energy, reliability, and ease of use would be the basic dimensions of the problem. It seems to come down to parasitic Ek that is generated in the process of making a decision. The computation can be limited and though the intent is to be able to compute, it would seem wiser to approach the problem from the goals instead of the abilities. It seems the goals are to make a computing machine that is monstrously fast and complex so that it can solve problems. That is a two part question and the ultimate goal is to solve problems. So it seems that just ramping capability without some sense of what is computed is the real problem. It is all well and good to have a machine that does 10n operations per second, but what is to be computed by the system? If the dependency relationship of the goal included a pinch point that could not be passed without a massive surge in computing power then it would be appropriate. At some point it must be considered if the goose chase leads to a real goose.
The basic principles of logic are simple and anything can be constructed once a specific agent such as an electron can cause cascade changes in a set of electrons in the structure. The series combination is AND, and the parallel combination is OR so that any relationship can be achieved once the principle of solid and reliable amplification of self-same elements is identified.
There are sea stars that operate on a form of pneumatic logic IIRC. A gating system can be achieved by valves controlled by fluid flow to amplify an effect. It is not even necessary to have a 1:1 correspondence between the agent of amplification and itself, it only requires that
( A -> B -> C -> A++ ).
In the overall scheme of things there are relationships that follow a common form and the nature of the interaction of factorial combined elements leads to situations that are not possible to be resolved by computation. Like the search for a bigger and more effective weapon, perhaps somebody lost sight of the fact that it made sense in the small, but fails miserably in the large. Perhaps it is not a bigger gun that is needed, but a different approach that considers the intent.
0 comments:
Post a Comment