User Tag List

Page 1 of 15 1234511 ... LastLast
Results 1 to 12 of 170

Thread: AI the Charlatan way!

  1. #1
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)

    AI the Charlatan way!

    If we could teach computers language instead of numbers and symbols, they would interpret things quicker. To teach it language we would need to group symbols together to be coded into one 'word', all definitions of it should be input or grouped into that 'word', and that should be hardiwred to the mother board, so, instead of counting ones and zero's, they will be using thir 'vocabulary' to execute tasks. Then we could converse with the computer, having it spew out words from it's logic, until it comprehends the langauge being used by the user.

    Of course the mother board will look quite different, but the selection of the group will no doubt make the exectutions quicker, and, make the language quicker. Each node thing on the mother board could group similar words into it's group and then make a comprehensible set out of what it recieves, teaching it that some things are more likely to mean this or that, and then through attrition it could make up the answer of the task set out for it.

    How about giving it a conscious? A conscious comes about by aggression of the desire to leanr or know, so, then it would be important to make them want to know. This could be set up as a question applied every so often as if it were a task, and the computer could try to answer it using what it has been given to work with. You could set it up each night and let it run through while you sleep. Then in the morning you come back to the computer and tell it if it is correct or not, maybe even learning thnigs along the way for mankind!

    This could be done with 'nodes' instead of on off swithces and then adding a erasable memory that it stores all of it's conclusions in.

    In time the common answers that are right can be added to the nodes that replace the on off switches then it can evolve, until such a time as it can think for itself, always setting the task as to ask questions about itself, reason out it's memories with it's nodes, and answer queations about itself, ask it if it is finnshed, then tel if it is right or not and write it to memory, taking note of the paths used, which should also be in language. It might sounds long and ardous at the moment, but progress should be swift.

    !! Going to my destruction !!

  2. #2
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)

    Running multiple programs

    It would be better if the computer could run lots of programs at once. This though requires ram and lots of memeory to accomplish, so, how do we do it another way?

    We could also make the programs run off of each other, using the same memory as the other programs, lowering the ram required and unused disk space by partitioning the ram so that - seeing as how it doesn't run off of on off switches in the hardware anymore - It will use the same hardware multiple times. So the 'node' on the motherboard that replaces the on off switches with words or terms could be used multiple times, as it is drawing from the mother board, reading from it, not turning it on or off. This means that it could switch from term to term multiple times, reading it from the mother board and using it. There is no limit to how many times you can use a symbol in maths, so these are more like mini programs that are built into the motherboard.

    Then there could be simple groups of words put into terms for use as programs, using programs on the motherboard means that they will not require memory to run, but rather just go straight to the cpu where they will be processed. This requires less hardware, and space for the 'pc', so a conventional pc will be able to take up less space and be packing more into a smaller package.

    Then there is cybernetics. Seeing as how it is so small, and communicates wih the brain and nervous system, it may be displayed in the minds eye, not physically on the eye, but rather inside the mind in their imagination they will be able to access all sorts of things for their use.

    The command line should read the same as in the simplest form of programming, and efforts should be taken to expand the simplest form of programming so that it remains easy to process, using less and less characters.

    !! Going to my destruction !!

  3. #3
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)

    Interaction in cybernetics

    Interacting with the cpu should be simple too, hopefully. For a terminal you should use a keyboard, but, for cybernetics you would need something new.

    Maybe if the cybernetics were based on impulses from your nervous system or whatever, you could go to the progrmas you want to without actually voicing it - you just activate it subconsciously as if you were raising your hand to block a attack from another, or as if you were pulling your hand out of a fire, spontaneous. Then that will guide you through the programs that are running and you can read what you are doing in your minds eye.

    This interaction means that you will not be able to lie to yourself. For example if you want to play games or something like that it will automatically take you there, so laziness will have to be overcome with willpower, forcing yourself to run the program continously.

    This may pose a problem when it comes to having other things on your mind. You could be in the middle of a program and wonder what the weather is, then it would automatically take you there.

    Internet on this medium could be accessed by having an aerial inside your cybernetics, then you could pick things up, and of course radio could be picked up aswell, like through a three g today.

    So we need to be able to force programs into our mind's eye. This means that simply willing ourselves to load a program, and willing a 'lock', which can be consciously released would solve the problem.

    Then there comes the issue of interaction. Interaction could be accomplished by impulses from your brain instructing your pc what to do, like willing it like a muscle, as attached to your nervous system that is basically what it is.

    But how do we get images to your brain? If it is visualised in your mind's eye, how does that happen? Basically it would be all subconsious, so you won't actually see anything, unless you use your imgination to imagine as if you were seeing something. Then i would be that words just enter your mind, so it will be completely 'oral', thus far, but we could work on images in the future.

    !! Going to my destruction !!

  4. #4
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)

    Components

    To go forward we need to stop using transistors, diodes, resistors, and capacitors.

    To go forward we need to replace all of these with 'one word transistors' tat communicate with the motherboard and cpu, and the cpu needs to be replaced by a helmet for the brain that works off of input for the 'transistor' on the mother board and vice versa, so we will no longer need cpus.

    The new transistors will be words or groups of words - if there are a million transistors in a motherboard, then there are less than that in a dicionary, and some get used more than once per sentence. The sentence would be like a command line, so the motherboard will be more like a program in fact, one that works off of the brain directly for cybernetics and helmets.

    The parts of the computer would all be programs running off of the brain, stored in the transistors as one word or groups of words, whichever is easier. I can see that the joining of words can be done by a similar transistor and be based as a program I have mentioned, relevenace and trial and error, and, having a 'personality' of progress, peace, war, problem solving, and a unique 'personality' to interact with the person because of history, as that is how a persoanlity is formed.

    For example if the person asks if the computer is doing well today, the computer may decide that this is good, and ask the person how they are each time they log on, or, there could be a personality circuit installed based on a variety of personas, like one that is dedicated to helping people for people in medicine for example. This personality could be based on what the computer is told is relevant to it's output, and then the computer can have a goal.

    To talk freely witht he person they should learn things in their memory circuits that they remember for future, deleting things it feels the person is no longer interested it, but keeping the primary operating system, or, it could send signals to the brain for storage and then reload the stored information. This will take up matter in the brain, so people should be able to store whole libraries of information insdie their brains, using it to it's potential. Of course they may wich to delete stored things, like in the case of rape for example, or a harrowing experience of some other kind like losing a loved one could be deleted.

    !! Going to my destruction !!

  5. #5
    BANNED TheColdTruth's Avatar
    Join Date
    Feb 2004
    Location
    Ohio, USA
    Posts
    332
    Threads
    5
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Reading your posts are almost always painful and disappointing.

    You immediately miss the problem with AI: the ability to instill in a computer the ability to be taught. That and the fact that letters are symbols for more complex symbols, just like numbers or well, symbols. It's just a different language, aside from that, they're the same.

    The rest is just so much expansion on an idea that wasn't grasped to begin with.

    I really have to wonder.. are you high when you post here? That would certainly explain a few things.

    You just don't understand how so many things work... yet you postulate on such complex things! I just don't get what you could possibly gain from making posts like this.


  6. #6
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)
    Quote Quote by: TheColdTruth View Post
    Reading your posts are almost always painful and disappointing.

    You immediately miss the problem with AI: the ability to instill in a computer the ability to be taught. That and the fact that letters are symbols for more complex symbols, just like numbers or well, symbols. It's just a different language, aside from that, they're the same.

    The rest is just so much expansion on an idea that wasn't grasped to begin with.

    I really have to wonder.. are you high when you post here? That would certainly explain a few things.

    You just don't understand how so many things work... yet you postulate on such complex things! I just don't get what you could possibly gain from making posts like this.
    Actually on off switches are longer to convrt than a hard wired word. Porcessing words at a time instead of on off switches saves a lot of time.

    The AI comes in where the computer remembers in it's memory what is the right combination of 'words' or 'phrases' and then speculates for new ones, that we tell it work or don't work.

    Ok I can see where you are coming from - the computer needs to know what is right by itself. This can be achived by giving it a task, and, finding alternative methods to gain this goal - by specualtion based on trial and error - it will learn itself how to get to the goal it was given.

    For example, it could be told that it needs to kill the player in a game. It will eventually, or immediately, cheat, as that is the fastest way to kill the player. Now we want the computer to cheat with other things, making our lives easier. Wel tell the computer what we want done and it finds the fastest way to achieve that, often using methods we might not have thought of ourselves, thereby teaching us.

    Now hardiwring this to ourselves is called cybernetics I think. If we were to make it small we could store it in our bodies, on our skulls, shoulders... wherever. That will interact with our nervous systems and we will recieve these signals as if they were impulses in our own bodies.

    Making it personal we could also see it preferring some methods to others, as the task that has been laid out for it means that it will prefer the fastest method, allowing it to idle. Telling it that idling is good between tasks means that it will take that as a goal and find the fastest way to complete the task and idle.

    If it were an accounting package it may store things in it's memory that are used a lot and apply them to the formula it is using, or, add other things to the equation that make the same answer and are quicker because idling is 'good'. This will teach us and it how to learn.

    Giving it a personality is not out of the question, but is needless at this point.

    !! Going to my destruction !!

  7. #7
    BANNED TheColdTruth's Avatar
    Join Date
    Feb 2004
    Location
    Ohio, USA
    Posts
    332
    Threads
    5
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    In a binary language, 1=yes, 0=no.

    Just because you don't understand programming language doesn't mean it isn't made up of words.

    Giving a program at this point is impossible by definition. It can assume nothing personal; it does what it is programmed to do.

    Your point is so convoluted I just don't even get at what you're trying to say.

    I showed this thread to a few of my friends just to make sure I wasn't missing a good point you were making.. and the general consensus was that this thread makes absolutely no sense.

    What are you trying to say? Why elaborate on things you have no understanding of?


  8. #8
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)
    Well I showed this thred to a few programmers myself, and they said it was possible. But don't bog me down with ad populum please.

    What I am trying to do is make it faster. Instead of using on of swithces I want to use transistors that make up words. We don not use on off switches, I think in words so can a computer.

    When it comes to using the 'transistors' they wil be interpreted by the operating system and programs.

    !! Going to my destruction !!

  9. #9
    BANNED TheColdTruth's Avatar
    Join Date
    Feb 2004
    Location
    Ohio, USA
    Posts
    332
    Threads
    5
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Jesus christ.

    Clearly you DIDN'T talk to any COMPETENT programmers.

    For you to not understand programming language and written language this much really takes .... something.

    Google "binary converter" if you still don't believe me, ffs.


  10. #10
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)
    Ok, I am going to try this really slowly. Instead of converting things into binary, we do away with binary, so it works on text only.

    The problem is the computer doesn't understand text yet, but placing the text into it's transistors, instead of on off switches will solve this.

    If it uses an on off switch, then it will use a word instead, or letters, or both so it can learn new words and numbers.

    It will understand because it will only work with them, not binary. Binary complicates things and takes longer, and depends on the cpu anyway. To program the cpu to understand these things we instead of using binary we use words and numbers.

    This is not a program, it is hardware that precedes a program. Do you understand what I am talking about now? I know full well what binary is.

    !! Going to my destruction !!

  11. #11
    BANNED TheColdTruth's Avatar
    Join Date
    Feb 2004
    Location
    Ohio, USA
    Posts
    332
    Threads
    5
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    My god, you really are just that stupid, aren't you?

    Go buy a book on how to program visual basic, learn a bit about it, then come back here and post.

    Maybe then you won't sound so utterly and completely idiotic.


  12. #12
    Amateur stripper Charlatan's Avatar
    Join Date
    Apr 2007
    Location
    Cape Town South Africa
    Posts
    4,196
    Threads
    652
    Post Thanks / Like
    Blog Entries
    67
    Mentioned
    16 Post(s)
    Tagged
    0 Thread(s)
    I don't know why I entertain you any further... in fact I won't! It is my job to solve problems and come up with new ideas, so you are calling my bosses stupid? I would take action against you, but you are not worth it. In fact I think you are replaceable in whatever you do, and know I am not. So go follow up on something else.

    !! Going to my destruction !!

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •