You’ll see how to create a new file geodatabase. In this file geodatabase you'll see how to build a topology to control the spatial relationships between some of the features and feature classes. Feature classes that participate in a topology must have the same spatial reference, so they must be in the same feature dataset. Since this geodatabase does not contain a feature dataset, you will see how to create one.

The spatial reference for this feature dataset will be imported from an existing coverage. Importing the coordinate system information from an existing coverage or feature class is one way to set the spatial reference for a feature dataset. You can also select one of many pre-defined coordinate systems, or define your own.

The next step is to set the XY Tolerance for the dataset. The XY Tolerance is used to resolve the location of coincident vertices during overlay of features and topology clustering. All of the features in a dataset share the same XY tolerance.

The default XY Tolerance is usually acceptable. In this case it is 0.001 meters. The best data that will be stored in the dataset is accurate to a few meters. You might use a smaller XY Tolerance if you planed to add other data to the dataset later that could be accurate to a couple of centimeters. A millimeter XY tolerance will preserve centimeter level accuracy, so you’ll accept the default value.

You’ve seen how to set the coordinate system and define the XY Tolerance for the dataset so it can contain the feature classes from the coverages and maintain the shared geometry of the features without degrading their accuracy, or the accuracy of higher resolution data that you might add in the future. If you had data from shapefiles, or less accurate data, you might have set the XY Tolerance to a higher value (perhaps a few meters) in order to better integrate linework and automatically resolve overshoots, undershoots, dangles, and slivers during topological and overlay operations.