the difference between projecting data and changing a dataset's spatial reference

[TedRakel]

Is there a difference between

1) projecting the data in a feature dataset using arc toolbox->data management tools->Projections and Transformations->Batch Project
and
2) changing the spatial reference on a feature dataset

Also, does anyone know a way to reproject data using something other that arc toolbox or the python arcpy.project_management method?  Perhaps something in c#.

[markdenil · Accepted solution]

There is all the difference in the world between
projecting the data in a feature dataset and
changing the spatial reference on a feature dataset

projecting modifies the coordinate space and
calculates the feature coordinates to what they become in that new space.

changing the spatial reference just changes the description of the space
but leaves the feature coordinates unchanged.

It is the difference between the princess kissing the frog (and changing him into a prince)
and the princess just calling the frog a prince, and letting it go at that.

Which option she picks depends on what she wants in the end.

You can modify coordinates by changing the spatial reference on a geometry object, for example while you are inserting it into a feature table with an Insert Cursor. This tends to be very slow.

... and of course many things are possible using ArcObjects (in for example, C#)...

[TedRakel]

Thanks.  That answers my question.