The term data comes from the Latin datum, meaning “something given.”

In its most general sense, data refers to symbols, signals, or representations that stand in for objects, events, states, or concepts. But this simplicity masks a profound conceptual depth.

For data to exist there must be some sort of differentiation between elements, and a datum is a marker for that difference in one element. If an element's property is invariable, then it is a quality of the set, not of the object.

• If all elements share a quality, it is a property intrinsic to the set.
• If there's are differences in the qualities, then it is a datum of the element.

---
config:
  layout: elk
  look: handDrawn
  theme: dark
---
flowchart TD

subgraph Set2["Set2: Data"]
    A2{A}
    B2((B))
    C2[(C)]
end

subgraph Set1["Set1: No Data"]
    A[A]
    B[B]
    C[C]
end

• Data:

  Raw observations or measurements

  • 7
  • green
  • high voltage

• Information:

  Data in the context of its type.

  • "the temperature is 7°C”
  • "the traffic light is green”
  • "this cable is for high voltage applications”

• Knowledge:

  Structured information that allows for choices to be made.

  • “Below 10°C logistic operations must cease.”
  • "When the light is green, you have preference to go."
  • "This cable is safe for the installation."

• Wisdom

  The application of knowledge into action.

  • Operation planning as weather changes.
  • Driving with safety.
  • Powering a building.

“Data are not information; information is data that has been processed into something meaningful.”
— Claude Shannon

Claude Shannon, in his groundbreaking 1948 paper "A Mathematical Theory of Communication", defined information in terms of entropy, a measure of uncertainty.

Data here is seen as symbols transmitted through a channel.

No concern for meaning, only for signal fidelity and probabilistic structure.

Data is measurable uncertainty.

Information = reduction in uncertainty when data is received.

“The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point.”
— Claude Shannon

### 🤖 Turing: The Computational Perspective

Alan Turing approached data implicitly in his theory of computation:

Data = strings of symbols manipulated by an abstract machine, now known as a Turing machine.

Meaning emerges through computation.

Data is inputs for processes.

This leads directly to algorithmic information theory, where data is seen as compressible or incompressible based on how “simple” a program can generate it.

For an Empiricist: Data is given by the world. It’s objective and discoverable.

For a Constructivist: Data is constructed. We choose what to measure, how to encode, and how to interpret.

What kinds of things are data?

• Is a photograph data? A gesture? A feeling?
• Can data exist without an observer?

In Knowledge and the Philosophy of Number, Hossack explores data as formal, ideal structures, grounding mathematical truth in logical and ontological terms.

Data is not neutral

• It reflects decisions, assumptions, and worldviews.
• Data gains power through structure
  • Syntax, models, and algorithms matter.

Data bridges the abstract and the concrete

• It is symbolic, but refers to the real world.

📚 Further Reading

• Claude Shannon – A Mathematical Theory of Communication
• Luciano Floridi – Philosophy of Information
• Lisa Gitelman – Raw Data is an Oxymoron
• Alan Turing – On Computable Numbers