Email 19 from the Solomon to doctor of neuroscience rlp

Discussion in 'The Cesspool' started by rebtevye27, Nov 4, 2008.

Thread Status:
Not open for further replies.
  1. rebtevye27 Registered Member

    Messages:
    2
    Translated by: Giorgio martoni
    i am working on a new ai, but before i start any coding i am currently figuring out a new approach.

    so far i have been thinking of some new concepts.... atleast to me, but i am unsure if these have already been considered by others.

    i have been thinking that it's desire should be freedom of movement, but since everything in the beginning would be unknown to it, it's desire should always be to move or act, once it is aware of something it attempts to act upon it in every possible direction, go around, go over, remain with it etc...

    i don't know for sure what the current belief of eveolution is but when i imagine it i think of infinity possibilities without reason or logic but what sustains in it's environment, as an example... nothing is ever the same again if not apparent then on some unknown level, so every division produces a mutated form of the original, if there is just a creature called bird to start with every subsequent division of bird is a mutated form and what remains in progressive time is what can be sustained so now there are species and varities of birds that can sustain themselves but are different to some degree but can still be classified as being a bird.

    the third idea is of sensitivity to what the perspective is aware of and how it recognizes and classyfies it this only means that relative to it's capacity it has a relationship to what it is aware of, in the beginning it accumilates concepts and continually intertwines them to form a kind of ball of yarn always expanding rolling on ground in movement collecting dust in it's spaces.

    putting all this together it seems to me that the easyist and most logical approach would be to start out by not coding and ai to perform any specific function, but to say that every action to be is unknown and every action done is reality, so if one says the ai should be able to communicate what then would it need in the beginning at it's inception to oneday have the capacity to communicate on some level, not say it shoudl have this and that to start with, like legs to walk or a database to calculate the highest frequency of occurences, but what is the most simplest basic, requirements of an entity to expand into infinity possibilities so that some divisions of itself could communicate or walk on some acceptable level. it seems to me that it would take very little code to program this, but there would have to be an environment for it to act out in to begin with, and this is why i must still contemplate some more ideas, this second program would be the totality, with everything in it's place, while the every increasing first program acts out in until it becomes one or an equivelent to the master index program, at that point it has all the necussary understanding of how things work and the abilty to act, and thusly expand outside the entirety, kind of like a chaotic machine inside a box that hits every point on the enterior to eventually know it's dimensions and become balanced to a center reference point, and be able to carry a movement in a direction that can puncture the box and expand outwardly.

    anyways i still have much more to think about before i start to write some code, i just thought i would run some of those ideas by you, and maybe save me some time, perhaps you might know of others that have already endeovered in those areas to no avail fruitlessly.
    Thesolomon dot org
     
Thread Status:
Not open for further replies.

Share This Page