Technique G227:Using the Map Equivalent Purpose Framework to provide an equivalent purpose text alternative for digital geographic maps

Generalized

Generalized means that "something" is being used to approximately represent the real world. The primitive map symbols are points, lines, and polygons. The Generalized category asks whether every point, line, and polygon for every object on the visual map is present on the text map. Every feature on the visual map needs to be present on the text map. There should be no shapes or features on the visual map that are not present on the text map, even if the feature lacks any label.

• Points are zero-dimensional features (a building entrance marker, a bus stop, a city dot, a point of interest).
• Lines are one-dimensional features (a road, a river, a contour line, a defined route, a power line).
• Polygons are two-dimensional features (a building footprint, a park boundary, a body of water, a census tract, a country outline).

If the visual map shows an unlabeled polygon for a body of water, the text map needs to contain a polygon for that body of water. If the visual map shows a thin gray line for an unnamed service road, the text map needs to contain that line. Omitting features the sighted user can see, including unlabeled features, fails the Generalized criterion.

Spatial Information

Spatial Information is about where features are, and what they are like (shape, size, orientation, location, and other properties). The Spatial Information category asks whether the shape, size, orientation, name, type, and any thematic, temporal, or sensory properties are present for each object on the text map, in enough detail for someone to redraw the feature. For each object on the map, the text alternative needs to contain, whenever the property is present on the visual (or non-text) map:

• Shape: the detailed borders of the object. A simple identifier such as "rectangle" is sufficient for a simple shape; an irregular complex shape (a U.S. state outline, an irregular building footprint) requires enough detail to reproduce the border.
• Size: the dimensions of the perimeter of the object in real-world units (or map units when real-world units are not possible). Example in real-world units: Blue Tavern has 3 walls. The two long walls that connect at a point are 50 meters long, and the short wall leading onto the patio is 20 meters long. In map units: Blue Tavern has 3 walls. The two long walls that connect at a point are 5 cm long on the map, and the short wall leading onto the patio is 2 cm long.
• Orientation: how the shape is facing and how the object relates to its bordering objects. Example: the Blue Tavern is a triangular building with the point of the triangle facing northwest at 11 o'clock. The point touches Fillmore St.
• Name: the label of the object as it is referenced when discussing information on the map (Fillmore St.). Every object on the text map needs a name of some kind, even if it is just "unlabeled body of water".
• Type: the categorical descriptor of the object (Street, Walkway, Restaurant, Park, Census Tract). Type may be derivable from the name (Fillmore St. has type Street).
• Thematic information: for thematic maps, the one or more numeric or categorical variables associated with each object that are visually encoded on the baseline map. Example: California: 543 Total Cases, 5 Total Deaths, masks required. Thematic information is distinct from Type, because Type is the object category and Thematic information is the overlaid data being visualized.
• Temporal information: the change in geographic, numeric, or categorical data over time, when the visual map exposes time-varying data through a timeline, animated layer, line graph, or date selector (for example, monthly COVID-19 case counts on 1/1/2021, 2/1/2021, 3/1/2021, and 4/1/2021 for each state).
• Sensory information: how the object is represented on the original non-text map. Example: the water polygon is a light blue color.

If a property is not present on the visual map for a given object, mark it N/A for that object.

Spatial Relationships

Spatial Relationships are the connections between multiple objects: the location, distance, topological intersections and adjacencies, and direction between objects that are core to understanding the information in the representation. The Spatial Relationships category asks whether the reader can position any drawn feature in its correct relationship to every other feature, equal to what is present on the visual (or non-text) map. For each pair of objects on the map, the text alternative needs to contain, whenever the relationship is present on the visual baseline:

• Distance: how far each object is from every other object, typically using Euclidean ("as the crow flies") distance and not accounting for intervening objects. Example: the green room is 23 meters from the red room. Distance can be expressed in real-world or map units.
• Direction: the angle between each object, using compass headings, degrees, clock positions, or directional relations precise enough to reproduce the layout. Example: Fillmore St. is 85 degrees from Clay St.; Sandwiches and More is at 2 o'clock from Men's Haircuts.
• Location: both absolute and relative location, whenever the visual map exposes them. Absolute location is the exact coordinates of the object (the center of the Red Room is at 45.0013722° N, 8.9998094° E). Relative location is provided through the combination of distance and direction between every object pair. When the visual map shows graticules, the distance and direction between graticules and feature edges need to be included.
• Topological relationships: the relationship between two objects across three dimensions (interior, boundary, exterior) for both objects. When two objects intersect, the intersection needs to be detailed across the three dimensions. When two objects only border, the borders need to be indicated (the square purple room is placed completely inside the rectangular Colored Room Zone, 10 meters from the top and 4 meters from the left edge). When objects are spread apart, the exterior relationship is captured by the Distance and Direction criteria above.

If a relationship is not present on the visual map for a given pair of objects, mark it N/A for that pair.

Routes

Routes are the defined paths on the map: how to get between two or more features on a map following a particular defined path, including the Spatial Information and Spatial Relationship elements for the route object, with the route information as prevalent in the text map as it is on the visual map. Typical defined routes include a turn-by-turn route, a designated trail, a bus line, or a defined evacuation path. If the visual baseline map has no defined routes, this entire category is N/A. If the visual baseline map shows defined routes that are not present on the text map, this category fails.

For each defined route on the visual map, the text alternative needs to contain:

• Route as a feature: every defined route on the visual map needs to appear in the text map, with all applicable Spatial Information properties (shape, size, orientation, name, type, sensory information, distance markers along the route). Example: there is a blue line tracing the shape of Fillmore St. Every 75 feet there is a blue marker showing the length. The line is straight, running south to north, and represents 94 meters. The line starts at 3321 Fillmore St. and ends at 1283 Fillmore St. The line angles from south to north at a 2 o'clock angle.
• Route in context: every defined route needs to be presented in relationship to the other objects on the map (intersections, intersection types, buildings, landmarks, and other features the route passes), with sufficient detail to reproduce the route within the surrounding geography. Example: the line begins at the center of the 3321 Fillmore St. building and continues north 10 meters, 3 meters in front of the building on the west side of Fillmore St. The route then crosses Clay St. for 4 meters at a stoplight intersection. On the north side of Clay and Fillmore the route continues for 5 meters in front of Dunkin Donuts, 10 meters along Lululemon, 20 meters along Wells Fargo, 10 meters in front of a parking lot, 5 meters past Boba and More, 40 meters in front of a Safeway, and 10 meters along the front of 1283 Fillmore St., ending halfway along 1283 Fillmore St.
• Route prominence: defined routes need to be easy to find and chunked into their own section. The time required to find route information on the text map should be comparable to the time required to find the route on the visual map. Example: a section titled "Routes" with each route presented as Route 1, Route 2, and so on, with the route content nested under the route name.

If a property is not present on the visual map for a given route, mark it N/A for that route.

Equivalent purpose decision

The text alternative serves the equivalent purpose when every applicable criterion (excluding N/A) in each of the four categories above is satisfied for every object, every object pair, and every route on the visual baseline map. If any applicable criterion fails, the text alternative does not serve the equivalent purpose, and this technique is not satisfied.