Losing Your Religion: Interoperability with AutoCAD Map 3D and ESRI - Part 1

I’ve been speaking at conferences for several years about CAD and GIS interoperability. It’s one of those topics where there’s a lot of interest and a lot of misinformation. Or at least, it seems to be much more difficult than it really is. I’ve been moving data between both systems for years, with very few real challenges. What I’ve found, is that the real issue is not the technical aspect of moving data back and forth, but the differences in how the software is generally used. AutoCAD (and other CAD systems) are primarily used for doing design work, and GIS (ostensibly ESRI, but it could be any GIS system) is primarily used for managing as-built facilities and systems. The real challenges are working between the design and as-built management processes. In other words, the issue isn’t CAD to GIS, the issue is Design to As-Built.

I’m going to make several blogs here in a series of the issues and some methods to make the process easier. This first post, I’m going to discuss the a bit about the similarities in the technologies. Following that, I’ll be posting on the differences, barriers, myths, and other issues involved.

Both AutoCAD and ESRI are built on basic primitive elements that are combined to create representations of real objects. Both systems include:

Points - a representation of a single location. It could represent a physical object such as a pole, manhole or brass cap in the ground, or it could be a non-physical point, such as a crime scene location or the corner of a property line. In any case, the systems both record a coordinate consisting of an X and a Y and possible a Z (if elevations are being included). The X and Y value could represent any projection or coordinate system, such as degrees of latitude and longitude or northings and eastings from a state plane.



Lines – a representation of of a connected set of coordinate pairs. Every line is going to have a start point defined by X, Y and/or Z, and and end point defined by an X, Y and/or Z. It could represent the centerline of a road, the edge of a building, or a buried pipe.



The line may be defined the system by coordinate pairs, such as point A and point B, or it could have the actual coordinate values in the line definition, such as this example from AutoCAD (a listing of a line – the start point is the set of parenthesis with the 10 X Y Z, and the end point is 11 X Y Z):



In some cases, for example, ESRI, the actual coordinates of the line are stored within an object “envelope”, which is a rectagle enclosing the object.



Polygons – a representation of an area. It could be a representation of a parcel, a building footprint, or an animal migratory zone. It is defined by lines and so by a series of bounding coordinates. Generally, in vector systems (save that discussion for another time), polygons are defined by their boundaries. They could be defined by groups of lines, or it could have the coordinate values built into the definition, simlar to the lines (as shown above).



In ESRI, the coordinate pairs are contained in an envelope bounding the entire object:



Attributes - data associated with an object. Associated data could be an identification number, a name, a description of the object, the color, size, diameter, etc. This is what turns a simple point, line or polygon into a representation of a fire hydrant, electric line or county. Attributes may be stored and linked to the object in a myriad of methods. It could be based on a common identifier stored in the object definition and the attribute list, as in a primary-foreign key relationship, or the definition of the object may be created to include certain atribute sets intrinsically. In some cases there may be a mixture of methods. For example, in AutoCAD, objects have intrinsic attributes (such as blocks attributes), extended entity data (attribute values associted to an individual object), or object data (data tables stored internally in the drawing and linked to objects). Additionally, both systems include methods to link objects to externally associated databases to extens the attributes of an object.



Understanding these similarities is key to understanding how to integrate these two systems. The next post, I’ll discuss the primary differences between the two.