I am going to represent a system of Mt as being mitochondrial cells in a fluid. I can say then that it is an array from a programming stand point. In terms of the body it can be associated with a specific cell number by an internal DNA sequence. This would be much like the IP address of an internet node. So I could send GCAT=address + XXXX ( a control sequence ) in the liquid environment and it would match a specific cell Mt[GCAT].
If it is cell free extract with Mt and it were arranged simply as content addressable memory then it would operate like this: I produce a string of DNA which is the query and it responds with a sequence of DNA which is the response.
Content addressable memory is much like having a crowd of people listening to a loud speaker and playing bingo. If I say "B9" , then the person who has all the elements yells bingo. In the case of wild card matches it is a bit more difficult. This is where a unique identifier plays a part. I can limit the response to a multiply valid query by also selecting an identifier range to limit the amount of "noise".
One of the key elements in this process is the interface to constructed DNA sequences, which serve as the transport layer for the protocol. In many ways it all functions like an internet. If elements are localized by position, then they can be "turned on and off" to form a particular shape.
So you see the idea of programming a model from genotype information to blender model and then to a physicality is a fairly simple process overall.
In the body, communications is done with sequences of amino acids and this becomes keyed on their shape. I personally think that the process would function much better with an RNA transport layer, however designs that originate by selection do not have the luxury of foresight.
So then it programs much like this: Mt[x][y][f]=z and as a result it either stores, recalls and transmits, or executes a method. Programs can be cascaded in such a way that sending command "X" would trigger a number of elements to start sending "threads" of instructions. It is definitely a different kind of programming, however I have done many device drivers that are far more tricky to use.
On another note, in the 1990s it was assumed that sequencing DNA would remain a long and tedious process. I knew that was not true. In fact, DNA sequencing should become as common as TCP/IP protocol. This leads me to the thing which I know, and it is the fact that singularity, as it is commonly called, is already here and the computer that goes beyond human intelligence is just waiting in the wings for its entrance on stage. It is not so much the technology that is important, it is how people respond to it. It is easily possible to have a chemical computer system that constructs physical objects, makes medicine, and even changes the genetic structure of individuals.
I think it is the idea that some may control others by limiting their access to resources or knowledge that must and will be forced to change. What happens when a free and open source AGI ( artificial general intelligence ) computer becomes the most famous singer? Who will own the copyright on that?
0 comments:
Post a Comment