Tod means Fox

Conforming data is an essential part of data integration. Conforming data takes on many shapes and may include updating data types and lengths, attribute names and conventions, data atomicity, universal conversions, etc. The idea is that, in all circumstances, the attribute means (and therefore acts) the same across all entities in the database.

For example, let’s look at temperature. Business requirements (and perhaps your region of operation) will dictate the format temperature should be stored in. When integrating data, chances are great that you’ll have different temperature formats (Celsius and Fahrenheit) coming from different sources. In order to have a conformed database, you create a rule that states that all temperatures are to be stored in X format. I like Celsius, personally, so I convert all Fahrenheit temperatures to Celsius:

nCelsius = (nFahrenheit  - 32) * (5/9)

If you disagree, then go the other way:

nFahrenheit  = (9/5) * nCelsius + 32

Like Kelvin? (You must be a scientist!)

nKelvin = nCelsius + 273.16

There are virtually hundreds of these types of conforming decisions to be made in a typical large integration project. Build yourself a toolbox of common conversions, and apply them to your data as needed.