About this page.
The CSA CAD Guide for Archaeologists and Architectural Historians

Important notice: The CSA web site was re-designed in August of 2010. Some documents then available were considered complete and static; so they were not included in the re-design and were not updated. This is one of those documents. Information about dates of posting and revision remains here, but there will be no revision of any kind after August, 2010.


A Glossary of Terms for Use with
The CSA CAD Guide for Archaeologists and Architectural Historians

Please note that words that are underlined in the text have been linked to appropriate entries in this glossary. When a link from the text is activated, this glossary will open in a separate window, with the term in question at the top of the page. Underlining is the only indication that a term is linked to the glossary; the normal color change to indicate a link has been disabled to minimize distractions. If you are using a browser that has been set not to underline links -- a possibility in the preferences settings -- you will be able to identify the linked terms only by the change in cursor shape as the cursor passes over the linked term.

axonometric drawing - a three-dimensional drawing of an object without foreshortening and with all verticals remaining vertical. A scale along each axis applies to all dimensions along that axis, and all parallel lines remain parallel in an axonometric view. (See isometric drawing.)

boundary representation (b-rep) - one of the systems for modeling solid objects. Boundaries of an object are defined, and the software constructs a solid within those boundaries. (Compare constructive solid geometry.)

CAD - an acronym for computer-aided (assisted) drafting or computer-aided (assisted) design. CAD software is used to design or document physical structures or objects.

Cartesian grid - a 3D grid system, named for the French philosopher Descartes, who defined the underlying idea of a 2D grid. The 2D grid (using x- and y-axes) is commonly used in algebra and geometry to map lines derived from equations and to deal with geometric shapes. The 3D grid required the addition of a third axis, running through the 0,0 point -- the x-y intersection -- and being perpendicular to both existing axes. With the addition of the third axis, any point in space can be defined by x-, y-, and z-coordinates (positive or negative), provided that one knows where the 0,0,0 point is and what the orientation of the grid is. In CAD systems, the grid system is a given, and all object are oriented within the grid. (Positive x is normally assumed to be east, positive y to be north, and positive z to be up.)

close-range photogrammetry - photogrammetry practiced with the aid of computers, permitting non-stereo photographs to be used to calculate fully 3D survey data. Images may be used from uncalibrated cameras, but there is a loss of precision when using such images. Close-range photogrammetry is sometimes called desktop photogrammetry. Digital photographs may be used with some close-range photogrammetry programs. (See photogrammetry.)

constructive solid geometry (csg) - one of the systems for modeling solid objects. Regular primitive shapes (blocks, pyramids, cones, etc.) are added to and subtracted from one another to create the desired object. Compare boundary representation.

coordinate system - a specific instance of a Cartesian grid. In CAD systems a default Cartesian grid is assumed, and models are normally constructed within that grid system. However, an alternate Cartesian grid -- an alternate coordinate system, which may have a different 0,0,0 point and/or a different orientation -- can be established with ease and used along with the default coordinate system. For instance, the oblique face of a structure can provide the x-y plane for an alternate coordinate system so that simple dimensions on that plane -- treating that plane as a simple x-y grid -- can be entered. There are many reasons for employing alternate coordinate system when making a CAD model; so it is both an important concept and an important feature of CAD systems.

database link - a connection between a CAD drawing entity and a record or row in a database. The term may also refer to a general connection between a CAD file and a database file.

desktop photogrammetry - see close-range photogrammetry.

digitize - to enter data with a digitizer or to translate information (such as a paper drawing) into a digital format.

digitizer - a device for entering data into a computer. A digitizer is similar to a mouse in the sense that the screen cursor is positioned by the user and a physical device attached to the computer, but the user employs a pen or other moveable tracking device on an electronic tablet to move the cursor. Since the tablet presents a defined surface (unlike the unbounded and undefined surface on which a mouse can move), it is possible to relate position on the tablet to position in a coordinate system and even to scale the tablet surface for tracing drawings.

digitizing tablet - see digitizer

DWG - AutoCAD®'s native file format for CAD models. AutoCAD's dominance in the CAD market has meant that this format can be read and written by many other CAD programs. (See file format.)

DXF - drawing exchange format. A file format for the exchange of CAD information. Although the DXF format is public, it was defined by Autodesk (the company that produces AutoCAD), and changes are made to the specifications by Autodesk without consultation. As a result, the format, though widely used, is not considered a public standard. (See file format.)

element - See entity.

elevation - a drawing of the vertical face of a building, wall, or other object. Such a drawing has no three-dimensionality and is simply a head-on view of a vertical face.

elevation angle - the angular deviation from the horizontal to which a surveying instrument's telescope has been raised to sight a target.

engineering drawing - a drawing consisting of at least three views (front, top, and right side) of an object. Certain conventions are normally used to indicate hidden parts of the object; the point is to specify the object fully with simple plan views of the visible surfaces. Views of other surfaces may be necessary, as may cross-sectional views. Dimensions are normally provided.

entity - a CAD model consists of lines, circles, points, surfaces, and so on. All may be called drawing entities, providing a convenient term to refer to any piece of a model without specifying its character. Sometimes called an element.

file format - the specific way information is recorded in a computer data file. Specifications of a format permit the file to be written according to a standard, opened for use or alteration, and written back to a storage medium for later access. Proprietary file formats are those created and maintained by individuals or corporations and not made public.

GIS - an acronym for geographic information systems. Programs for connecting maps to tabular information about places/areas/objects located on those maps. Important analytical capabilities are provided. See Mark Gillings and Alicia Wise, eds., AHDS GIS Guide to Good Practice (http://ads.ahds.ac.uk/project/goodguides/gis/) for an overview of the use of GIS in archaeology.

GPS - global positioning system. This refers to the satellite-and-receiver system originally used by the military and now widely used throughout the world to find one's position in latitude and longitude. A number of satellites in geosynchronous earth orbits send signals constantly. Receivers in position to receive at least three of those signals can use the signals -- including information about the positions of the satellites and calculations of the distance from the satellites to the receivers -- to determine the positions of the receivers. Depending on cost and technological sophistication, the receivers provide differing levels of precision; the highest precision requires a ground station as a fixed location and relatively long signal acquisition times. Otherwise, precision may be quite limited, though changing technology means that even unaided GPS receivers can provide good data for ground surveying today. GPS locations are more accurate for latitude and longitude than for elevation.

hidden-line drawing - a hidden-line drawing shows a scene with appropriate objects or parts of objects hidden by intervening surfaces -- as in the real world and as opposed to a wire-frame drawing in which nothing is hidden. A hidden-line view is the on-screen version of a hidden-line drawing.

isometric drawing - an axonometric drawing with the three principle planes at the same apparent angle to the plane of the drawing surface.

layer - drawing segments. CAD files normally have their contents divided into segments that may be displayed/suppressed or drawn/omitted on command, making it possible to see only specific parts of a model at any given moment. These data segments are called layers. Layers need not have physical boundaries but may be entirely conceptual. Layers also provide some added security, since layers may be individually protected from alteration.

level - See layer. Level seems even more likely than layer to imply positional juxtaposition.

migrate - to change data from one file format to another.

model - in CAD use, a representation of some particular physical reality or realities (or planned objects) in computer form. The term is used in place of the simpler term drawing, because many drawings could be created from a CAD model. A 3D model is obviously too complex to be considered a drawing, since changing the point of view changes the drawing. A complex object drawn with multiple layers is also too complex to be considered a single drawing, even if it is a simple, 2D representation; changing the visible layers may change the drawing produced but not the model.

parametric program - in CAD use, a program that saves and provides access to the design sequence for later modification. A parametric approach to CAD modeling provides the model maker with the step-by-step processes used in the design process. Any step can be modified at any time, making the model always the result of the design steps, not a simple collection of geometric figures. The advantage to model makers is the ability to modify the design steps at any time in order to change the model.

perspective drawing - a 3D view with foreshortening and vanishing points. Many perspective drawings use only two vanishing points, permitting the verticals to remain vertical. A three-point perspective uses three vanishing points so that even verticals converge.

photo rectification - the process of altering a photograph in the copying process to change the apparent angle of view. An oblique-angled photograph is rectified when it is altered so that the apparent angle is head-on, and the correct geometric relationships between and among portions of the whole are preserved. (See plane transformation.)

photogrammetry - a general term for using photographs to derive measurements or point locations. Single-photo photogrammetry is equivalent to a plane transformation (see below) and can only yield information about points lying within a single plane. Photogrammetry as first used for 3D point determinations relied on the use of stereo pairs -- two photographs taken at the same time and with a fixed distance between the focal points of the two cameras. The two images could then be viewed with a stereo viewer or analyzed with standard trigonometric processes to determine the relative positions of any points visible in both photographs. If some points in the photos could be surveyed independently, all points could then be located in the same grid system. The system relies on very precise calibration of the cameras so that the calculations can yield reasonable precision; as a result stereo photogrammetry requires calibrated -- often called metric -- cameras, which are quite expensive. (See close-range photogrammetry for non-stereo-pair application of the principles of photogrammetry.)

pin-bar drafting - drafting on multiple sheets of paper (some of which are transparent) with the use of registration pins for alignment of the individual sheets. Each drawing sheet contains a different portion of the whole, and the pins allow all to be aligned correctly. Thus, when combined, the sheets show the full complexity of the object, but they show individual aspects when viewed separately

pixel - one individual, discrete point on a computer screen, the smallest element of the image made on a screen. (See raster image.)

plane transformation - the mathematical process of translating positions from one plane in space to another, differently-oriented but otherwise identical plane by knowing that certain points in each of the two planes can be matched to one another, making it possible to map all the remaining points on the original plane to equivalent points on the secondary plane. (See photo rectification.)

plotter - a device for making a large paper drawing. Several varieties of plotters are now available. Older models use pens on an armature and paper-feed rollers to allow paper and pens to move cooperatively and create continuous lines. Newer ones use electrostatic charges or ink-jet technology to put ink on paper as it is rolled through the printing mechanism.

random access memory (RAM) - the computer's volatile memory. When the power is off, the memory is empty. This is also the memory that can be randomly accessed by the computer's central processing unit (CPU); so information to be used actively must reside in RAM. (The hard drive is used for non-volatile storage, and information is transferred to RAM for actual use. Specific information on the hard disk cannot be located for random access by the CPU.) When RAM is insufficient for a specific purpose, some of the information will be placed in a special portion of the hard drive for retrieval -- often called the swap disk, and there will be a swapping of information between RAM and the hard drive so that the actively-used information is in RAM. That is a marginally effective process on most PCs; so using large files effectively requires more RAM than business-oriented machines normally need. (Virtual memory serves the same purpose of substituting space on a hard disk for RAM, and, although not identical in function, it suffers from the same speed problems.)

raster image - an image composed of dots or points. Images on cathode-ray-tube devices (televisions or computer monitors) are created by closely-spaced scanning lines that consist themselves of closely-spaced dots. Raster refers to the scanning pattern and is used in general to indicate an image made of individual dots rather than lines, circles, arcs, and so on. (See pixel; compare vector.)

refresh - to write a data file onto new media in order to counter the effects of magnetic decay.

render - to make a realistic, life-like image of an object or scene. Nearly photorealistic renderings can be produced with some CAD programs, though the best results generally require programs created specifically for this purpose. Such programs often use CAD files as their starting points. Thus, it is possible to create extremely realistic views of items modeled in a CAD environment.

scan - to create an electronic image of a paper document or object with a device attached to a computer (a scanner).

scanner - a device to create an electronic image from paper or film originals. The scanner uses sensors to determine the shade (or color) of the document at set intervals. The sizes of the intervals are determined by the scanner itself (generally stated in terms of dots-per-inch or dots-per-centimeter resolution) and by adjustments that can be made at the time of the scanning. The resulting electronic image is a raster image, with one pixel in the image for each sensing position of the scanner. The size of the image (measured in pixels) depends on the size of the original and the resolution of the scan. Three-dimensional scanners are similar in that they scan objects according to a pre-set resolution system, but they are able to record 3D locations of the scanned points, not simply the color or tone or a series of points in a grid.

solid model - a CAD model that treats solid objects as explicit solids in the model, not simply collected surfaces. Solid models permit realistic views, since surfaces are implied; they also permit calculations of object weight, center of gravity, tensile strength, and the like. Cross-sections can be made from solid models.

surface model - a model that includes specified surfaces. Surfaces must be defined as such; they are not implied (to the computer) by lines that meet to enclose space. Explicit surfaces are required to make a realistic view of a model, since surface specifications are needed for the program to determine which parts of the model would be hidden from a given vantage point. Surfaces are also required for renderings or VR production, since they provide the basic items for texture, color, and brightness.

surface normal - the exposed, visible side of a surface. Any defined surface is effectively two surfaces, one faced in one direction and the other faced in the opposite direction, 180 degrees away. The face of a block, for instance, may be the only part of that block visible -- and available for survey -- but the computer must have a way to know which side of the surface is visible. The surface normal is the visible face of a surface. (Surfaces can be defined by a series of points defining their corners. Viewed from one side, the points will have been entered in a clockwise direction; from the other side they will have been entered in an anti- or counter-clockwise direction. The side showing those points to have been entered in anti- or counter-clockwise direction is considered the surface normal.)

swing angle - the angular deviation from north for the telescope of a surveying instrument when a target has been sighted.

total station - a surveying device consisting of an electronic theodolite and a coupled electronic distance measuring device (EDM). The total station may also include a data recorder (almost always external to the surveying instrument itself so that it can be disconnected and connected to a computer) for retaining individual measurements -- swing angle (deviation from north), angle of inclination (from the horizontal), and distance to target. The total station itself will generally be able to calculate point positions, as will the data recorder, but the data recorder is required to store the information and to transfer that information to a computer.

vector - a mathematical representation of a line, arc, or circle, indicating the coordinates of the starting point, direction of travel, and distance of travel. A vector image is a screen or paper image generated from the mathematical representations of the drawing entities. The scale of the image can be enlarged or reduced to any extent desired, and the image will still be correctly shown, with all parts properly scaled and related to one another.

vectorize - to turn an image consisting only of pixels (a raster image) into an image consisting of vectors.

version control - controlling access to computer files to make certain that all additions and edits are made on current files and that back-up files and historic copies made during the project are not confused with or mistaken for current files. There should be only one copy of each current file available for editing. Back-up copies should be kept separately and/or labeled in ways that prevent confusion.

virtual reality (VR) - a general term used to indicate some form of extremely life-like computer representation. Some would restrict use of the term to computer-generated environments that are immersive. (In immersive environments the 'visitor' wears goggles with projected images and is positioned in a room such that he/she can move about and seem to be moving in the virtual space. Movement is reflected in the goggles, and he/she may even have gloves with sensors so that movement of the hands can be noted.) Others would broaden the term to include very life-like computer displays that may be manipulated in real time. In either case, real-time interaction between the computer and the user is a hallmark of virtual reality systems.

wire-frame - a model is said to be a wire-frame model if it consists only of lines in space. Those lines may imply shapes (boxes, cylinders, etc.), but there is nothing in the model to specify the existence of more than simple lines. As a result, no drawing of a three-dimensional object can appear fully realistic, since there are no explicit surfaces and, consequently, nothing to specify which part of a scene should by hidden by an intervening surface. A wire-frame drawing is simply a drawing made from any model (wire-frame, surface, or solid) that exhibits the characteristics of a drawing from a wire-frame model, i.e., with nothing hidden from the viewer by intervening surfaces.

Return to
Chapter 1.
Return to
Chapter 2.
Return to
Chapter 3.
Return to
Chapter 4.
Return to
Chapter 5.
Go to Table of Contents.

About this document: