4 ms·

"If someone tells you a fact you already know, they’ve essentially told you nothing at all." I think it's interesting to consider how this is/is not true in a

• phreeza 4y ago
  > "If someone tells you a fact you already know, they’ve essentially told you nothing at all." The reality of zero information transfer is more subtle than this quote makes it sound, and covers the point you raise. In reality, you must not just know the thing you are being told, but you also have to know in advance that it is the thing you are going to be told. So if I tell you "the sun is shining" when you already know it is, there is still information being transferred if I could also have said "yesterday it was raining". Only if "the sun is shining" is absolutely the only thing I could have said is it a zero information statement.
  • pyinstallwoes 4y ago
    There is information in the fact a signal was sent and received from the observer of the receiver.
    • phreeza 4y ago
      If the receiver didn't assign a prior probability of 1 to the sender sending the signal then yes.
      • asah 4y ago
        It is impossible to assign a prior probability of 1 to just about anything in this world. Perhaps death and taxes, but definitely not telecommunication !
  • Frost1x 4y ago
    Intent still exists here. We must also consider the communication median. People typically use speech to communicate which may tell me something new about the state of the speakers emotion. It may tell me if you have a cold or sore throat in the process. It's lossy and may even tell me something about the median it passes through. Zero information transfer is hard. I've never considered it before now but I wonder if it's even possible. There's even a time component and space component involved at a fundamental level it seems. Saying something at one time and another could differentiate things. Space clearly does. I pretty much need know the state of everything involved in the information transfer to not have information transferred. In that case what is "knowing?" It's not so binary when you think about it. Human memory is lossy, filled with gaps and fictions mixed with reality. Is the information captured and stored truly accurate? Is it mutated by my conscious thought and how I transform, process, and then store it?
    • picture 4y ago
      I think linguists have a specific name for the things that people say, where it doesn't carry literal meaning per se but rather in the act of saying it. "Good morning!" What does that mean? Is it a good morning, this? Am I now compelled to have a good morning? Are you telling me about your good morning? Maybe, but it also conveys friendliness or politeness and or other things.
• lr1970 4y ago
  > If someone tells you a fact you already know, then it could be communicating several different things. You are alluding to the "meaning of communication" while information and communication theory is strictly about bits that are stored somewhere or/and send over the channel. If you know with 100% certainty that the next bit will be zero then receiving this bit gives you no information beyond what you already knew. Of course, if your certainty is less than 100% then the bit delivered you some fractional quantity of information. Mathematically, communication channel couples together random process X at the transmitter with random process Y at the receiver. The game is to describe X while observing only Y. Mutual information is a quantitative measure of how much Y "knows" about X. There is no human meaning attached to it whatsoever.
  • telchar 4y ago
    You could say the first X bits of signal from the start of the communication to the point where you can realize what is being said and can predict the rest with certainty represent the informative part of the message; that is, the less you expected them to say the thing they’re saying in that context, the more information is transmitted.