TALAMANA · THE AI LITERACY MAP FOR ARCHITECTURE AND DESIGN · Foundations · AGE 15—17 · FACTUAL · DURABLE
Reading data
Every dataset answers a question someone chose to ask, and nothing else.
The idea
Before any AI there is data: site readings, surveys, costs, climate numbers. Reading data means asking three questions. What does it actually say? What was measured, by whom, when? And what is quietly missing? Every dataset answers a question someone chose to ask, and says nothing about the rest.
Why it matters
Every AI tool on this map takes in data and gives back a confident answer. That answer is only as good as your reading of what went in.
See it in the studio
A footfall survey counted weekday mornings only. The market runs on Sunday. The data is accurate, and wrong for your question. No machine will warn you.
Watch for this
A chart with no source, no date, no count. That is decoration dressed up as evidence.
Try it
Take any chart from this week's news. Answer in writing: who measured, when, how many, and what is missing. Two of the four are usually hard to find. That difficulty is the lesson.
Prove it
Take one dataset from your own work and state what it says, how it was gathered, and one important thing it leaves out.
How it works
Statisticians call the missing-things problem sampling bias. There is a ladder from raw numbers to decisions: data → information → knowledge → judgment. Machines climb the first rung fast. The top rungs are still yours.
What this idea builds on
- A starting idea.
What this idea opens up
Sources
Open this idea on the map · The complete map · Logika · RBDS AI Lab, India · revised every edition.