precision domain principle - LeFreq/Singularity GitHub Wiki

This is a tentative term to mean:

• a given datum must be a single dimension. If have a measure z of a hypoteneuse, you must break down that variable into it's counterparts: (x, y). However, there may be a different reference point-of-view, like polar coordinates, where this datum is in it's native form, in which case you make a conversion function. Color is another example. If you have a single color X, you must know what goes into the representation or creation of that color. There are at least three dimensions to color, so these must be separated into R,G,B or something equivalent. The tuple (R,G,B) is a single-precision datum to represent color, a new single-precision element comprised of lessor single-domain parts.
• If you have a chunk of data, you must be able to know the different ways or dimensions in which that data is viewed. Or even could be viewed. To know this latter, you have to be a master of data science, not just a computer programmer. A list of heterogeneous data might be a sloppy way to hold a disorganized mess. Find the data science to categorize the data, so that it can be placed in an ontology.