- Author: Mark Monmonier
- Full Title: How to Lie with Maps
- Category: books
Highlights
- In the sense that all maps tell white lies about the planet, small-scale maps have a smaller capacity for truth than large-scale maps. (View Highlight)
- Map projections, which transform the curved, three-dimensional surface of the planet into a flat, two-dimensional plane, can greatly distort map scale. Although the globe can be a true scale model of the earth, with a constant scale at all points and in all directions, the flat map stretches some distances and shortens others, so that scale varies from point to point. Indeed, the bar scale sometimes included in the lower right of an online map can be blatantly misleading when the user zooms out to show an entire continent. Moreover, scale at a point tends to vary with direction as well. (View Highlight)
- Perhaps the most striking trade-off in map projection is between conformality and areal equivalence. Although some projections distort both angles and areas, no projection can be both conformal and equivalent. Not only are these properties mutually exclusive, but in parts of the map well removed from the standard line(s) conformal maps severely exaggerate area and equal-area maps severely distort shape. (View Highlight)
- Map projections distort five geographic relationships: areas, angles, gross shapes, distances, and directions. Although some projections preserve local angles but not areas, others preserve areas but not local angles. All distort large shapes noticeably (but some distort continental shapes more than others), and all distort at least some distances and some directions. Yet, as the Mercator and gnomonic maps demonstrate, the mapmaker often can tailor the projection to serve a specific need. (View Highlight)
- Graphic symbols complement map scale and projection by making visible the features, places, and other locational information represented on the map. By describing and differentiating features and places, map symbols serve as a graphic code for storing and retrieving data in a two-dimensional geographic framework. This code can be simple and straightforward, as on a route map drawn to show a new neighbor how to find the local elementary school: a few simple lines, labels, and Xs representing selected streets and landmarks should do. Labels such as “Elm St.” and “Fire Dept.” tie the map to reality and make a key or legend unnecessary. (View Highlight)
(View Highlight)- Selection, which serves geometric generalization by suppressing some information, promotes content generalization by choosing only relevant features. Classification, in contrast, makes the map helpfully informative as well as usable by recognizing similarities among the features chosen so that a single type of symbol can represent a group of similar features. (View Highlight)
- Electronic graphics introduce further complications when a map author who is pleased with the image on a computer screen ignores the distortion that might occur when the map is printed with ink or toner, reproduced as a poster with an ink-jet printer, projected onto a screen in an auditorium, or viewed on a tablet computer. Colors might not look the same on a different display. (View Highlight)
- Maps are like milk: their information is perishable, and it is wise to check the date. But even when the map author provides one, the date might reflect the time of publication, not the time for which the information was gathered. And when the map was compiled from more than one source or through a long, tedious field survey, the information itself might be so temporally variable as to require not a single date but a range of dates. Particularly troublesome is the carefully dated or current-situation map that shows obsolete features or omits more recent ones. These errors might be few and not readily apparent; a map that is 99.9 percent accurate easily deceives most users. (View Highlight)
- Maps based on electronic data files can be highly erroneous when several sources contributed the data and the user or compiler lacked the time or interest to verify their accuracy. Obviously inaccurate data from careless, profit-driven firms have unpleasantly surprised purchasers of street-network information, and even data from reliable vendors occasionally have infuriating errors, as when roads and boundaries captured from adjoining map sheets fail to align at the common edge. Without careful editing, streets and streams are more easily omitted or misplaced in a computer database than on a paper map. Electronic data are incomplete without metadata (data about data) that address quality assurance and help data users determine the compatibility of diverse features that might be combined. (View Highlight)
- Around the turn of the twenty-first century, maps moved online. The change has been massive, much like the digital revolutions that swept the news business, entertainment, and political discourse, all of which at times seem curiously conflated. Scholars and pundits assert that maps are now ubiquitous, an obvious exaggeration but not far from the truth. More significant is that maps have become remarkably fast in both their creation and dissemination, and so strikingly that twenty-first-century electronic maps collectively merit the label fast maps, mostly because no other general term adequately accounts for both their velocity and their diversity. These maps can deceive in the traditional ways that have been discussed throughout this book, but they also raise unique issues that never arose with static or paper maps. (View Highlight)
- After a newspaper in a New York City suburb used a FOIA (Freedom of Information Act) request to obtain an address listing of pistol permits in two counties, it posted a zoomable map showing who in one’s neighborhood might own a handgun (plate 16). Outraged gun owners complained that the data not only ignored assault rifles and shotguns but made permit holders, who might not have actually purchased a firearm, vulnerable to intrepid burglars, no doubt the worst kind. Despite applause from gun-control advocates, the newspaper took down the map and hired armed guards to protect its staff and property. (View Highlight)
- Web cartography depends on a storage and retrieval strategy known as tiling, which divides an electronic world map into a large number of small rectangles known as tiles. Figure 14.1 describes this approach as a succession of zoom levels that start with a whole-world view at the top of a pyramid for which each closer, more detailed step involves markedly more tiles, each covering less territory in greater detail. To support the rapid zooming of web maps, sets of tiles must be generalized and stored for as many as twenty-two specific levels of detail. A user requesting a map for a particular address receives a display several steps up from the most detailed level. Clicking the plus sign to zoom in or the minus sign to zoom out typically reveals more or fewer types of features. A jump in the map’s labels is also apparent, as when street and business names appear or disappear when zooming in or out, as exemplified by the three successive levels in plate 17. Boundaries between tiles are never visible, and the number of tiles does not increase by a factor of 4, as figure 14.1 might imply. (View Highlight)
- A subtle consequence of web-map tiling is an invigorated use of the Mercator projection, denounced roundly for its infamous distortion of area on small-scale world maps but an indispensable tool for marine navigation and field artillery in the pre-satellite era (see chapter 8). The Mercator map is well suited to tiling because it does not distort angles. This mathematical property, called conformality, allows only a minimal, largely unnoticeable distortion of local shapes and distances on individual tiles at the more detailed zoom levels. The Mercator map also provides a rectangular grid of meridians and parallels, which accommodates the rectangular screens of laptop computers and mobile devices as well as the massive electronic storehouse of tiles, served up rapidly when a viewer pans across the map or requests more detail. A single projection assures an acceptably smooth transition from one zoom level to the next. (View Highlight)
- Cartographers attuned to the mathematical details of map projection acknowledge that the Mercator projection used for web mapping is not quite the same Mercator projection used for large-scale topographic maps, which are cast on an ellipsoid rather than on a sphere. Ellipsoids, which provide a more accurate geometric framework for military and scientific applications, are spheres flattened strategically at the poles to more closely approximate Earth’s actual shape. Extreme accuracy demands different ellipsoids for different continents, but this precision would be lost on web cartography, which gets by quite well by assuming a less computationally demanding spherical planet. Purists call this approximation the “Web Mercator projection” and note the use of different mathematical shortcuts—which means nothing to most users. (View Highlight)
- More problematic is the fully zoomed-out whole-world map, as in figure 14.2, on which Greenland looks inappropriately similar in size to South America. Coping strategies available to web-map services include naive acceptance of the traditional Mercator worldview (conveniently truncated to avoid a pointlessly humongous Antarctica), restricting the degree to which viewers can zoom out, and jumping to a more appropriate nonrectangular geometric framework for a whole-world map enhanced with environmental data that include polar regions. Even so, any cylindrical projection—not just the Mercator—is wholly inadequate when a viewer wants to zoom in on Antarctica, which is better served by a polar conformal projection, as in figure 14.3. (View Highlight)
- A web map service (WMS) is a business supported by venture capitalists confident that advertising revenue will not only cover operating costs but earn them a decent return on their investment. An efficient, well-promoted WMS attracts millions of viewers, whose attention is sold to advertisers, whose messages appear on the map’s margins, and sometimes on the interior, as when brand-name coffee bars and hotel chains pay to put their logos on a map rather than be represented by a generic icon like a coffee cup or bed. Putting a business’s name on the map is free, but a familiar icon increases the likelihood a viewer will click to find the phone number, determine hours of operation, or make it a destination for the GPS. A firm can add also an interactive street map to its website—perhaps for free, unless the map is heavily used. (View Highlight)