# Do locations “move” when shifting datum?

I hope that this is the right Stack Exchange for this question. I may be oversimplifying this, but I'm looking for verification in my understanding of the following:

If I have a spatial dataset in 2 different data [datums], the coordinates [absolute location] of the "same" point may differ between the two data. This magnitude will depend on the geometric difference between the datum ellipsoids at the location of my point.

Whereas...

If I have a spatial datasets with the same datum but in 2 different projections, the coordinates of the "same" point will be identical, but the distance/shape of path/direction between this and another point may vary between the two projections, due to distortions introduced during the projection process. The type and degree of this distortion depends on the projection used and where it lies tangent to the datum ellipsoid.

Is this an acceptable degree of simplification to constitute a "strong understanding" of the topic? Or am I way off and conceptualizing this incorrectly?

• Yes you are right with the first part but not necessarily with the second. Please see this very very informative wiki page for projections: gis.stackexchange.com/questions/664/…. Lastly the datum is dependent upon the base shape (spheroid) and different projections from that would yield different results, including the position. – fatih_dur Nov 24 '15 at 1:08
• Another note, "true" locations do not "move" but their representation. In overall, there is no such thing as absolute location, datum (and projections based on this) is just [as "near" as possible true] representation [roughly can be called as quantifable simplification] of the location. – fatih_dur Nov 24 '15 at 1:17