Our view is really a simple transformation of the existing data using PostGIS's ST_MakePoint function. The ST_MakePoint function takes the input of two numbers to create a PostGIS point, and in this case our view simply uses our x and y values to populate the data. Any time there is an update to the table to add a new record with x and y values, the view will populate a point, which is really useful for data that is constantly being updated.

There are two disadvantages to this approach. The first is that we have not declared our spatial reference system in the view, so any software consuming these points will not know the coordinate system we are using, that is, whether it is a geographic (latitude/longitude) or a planar coordinate system. We will address this problem shortly. The second problem is that many software systems accessing these points may not automatically detect and use the spatial information from the table. This problem is addressed in the Using triggers to populate the geometry column recipe.