fonts—including categories of fonts, fonts in the Windows operating system, type styles and font families, and creating special charactersWELL-DESIGNED TYPE IS A MAJOR COMPONENT OF THE professional appearance and clarity of a map. Cartographers are as careful about how they choose lettering for maps as they are about the symbols they design to present mapped information. Type characteristics categorize and order features as well as establish the personality and legibility of a map. However, it is not just about choosing the right font. Type on maps almost always needs to be modified in some way to improve its contrast and readability against complex map backgrounds. Creating effective maps relies, in part, on creating effective map type.
Most of the useful type characteristics found in design software can be created using GIS, allowing complete map designs to be achieved in the GIS environment. Fonts and styles, sizes and spacing, and text effects can all be edited within text elements. Mapmakers also need to consider type issues beyond their GIS environment. Type effects export with varied success, and well-planned type can save time in map editing. Understanding font tools in the Windows operating system can assist in using type well.
Mapmakers build better maps when they understand the following:
fonts—including categories of fonts, fonts in the Windows operating system, type styles and font families, and creating special characters
label size—including character size, character spacing, and line spacing
type effects—including callouts, shadows, and halos
Nearly all of the fonts shown in the figures in this book are either part of the set that comes with Windows 7 and 8 operating systems or are free and open-source fonts that you can download and install.
Figure 5.1 The basic parts of a letter. Figure by E. Guidero, Department of Geography, The Pennsylvania State University (Penn State Geography).
Fonts are the personality of a map. They may be serious and authoritative or carefree and inviting. Whatever their tone, they need to be legible in the challenging contexts that are characteristic of mapping. They must be easily read at small sizes and odd angles in varied display media. They are often read among numerous other labels that are close together. In addition, map labels may include unfamiliar words that are indecipherable if they are not clearly legible.
Before you can begin to shape the temperament of your map by choosing a font, you should understand the basic categories of fonts and how to use their special characters. You should also be familiar with the potential breadth of font families. This topic describes the basic tools and details you should know when choosing fonts for mapmaking.
If you are curious to see what fonts are installed on your computer, look through the Fonts folder in the Windows directory on your system drive. It may be easier to look through the list in your word-processing or graphic-design programs or using an online tool that examines your system (for example, Flipping Typical, http://flippingtypical.com).
This chapter uses some font-specific terms, which are illustrated in figure 5.1.
Strokes are lines that make the shape of the letter. Typefaces can have either varying or constant stroke widths. The amount by which the strokes vary is called stroke contrast. For example, the o in “Foxtail” becomes narrower at the top and bottom, and then wider at the sides. It has high stroke contrast. In comparison, the O in “Orange” is almost the same width all the way around the letterform.
Serifs are the small finishing strokes on letters, seen as little “feet” at the bottom of the letter and on the ends of straight parts of letters, labeled on the capital F in the figure.
Terminals are the ending strokes on letters that do not have serifs. Sans serif fonts often have curved terminals that resemble serifs on just a few letters to aid legibility, such as lowercase t and l.
Baseline is the lines on which letters sit. The flat portions at the bottom of letters (such as capital F) sit right on the baseline. Curved elements usually extend slightly below the baseline.
Cap height is the height of capital letters without curved elements at their top or bottom (for example, the height of capital F). Curved elements, such as the top or bottom of the capital O, can extend beyond the cap height. Lowercase letters may also extend above the cap height.
X-height is the size of the lowercase x, from the baseline to the top of the x (where flat letters end). Letters with curved elements, such as the lowercase o, may extend above the x-height. Fonts with a large x-height relative to the cap height will appear larger when used with other fonts at the same point size.
Descenders are the part of a letter that extends well below the baseline.
Ascenders are the part of a letter that extends well above x-height.
Counters are closed spaces inside letters, such as the hollow space inside the o.
Bowl refers to the curved portion of a letter, describing the shape of the top part of the letter “e” or the bottom part of the letter “a.”
Technically, a font is one size and style of a typeface, but the more general use of “font” here is consistent with how typefaces are selected in GIS and other applications.
Recognizing and naming individual fonts takes much practice and is a more detailed level of study than necessary for most mapmaking. Five basic categories of font appearance are useful as you learn map design:
text
display
serif
sans serif
monospace
Text fonts are suitable for small sizes and continuous reading. Most typefaces you use for map labels are text fonts. Text fonts include a range of styles, and both serif and sans serif typefaces can be considered text faces. Overall, they are simply constructed, easy to read, and do not have flourishes or decorative elements (figure 5.2).
Figure 5.2 Calibri, Linux Libertine O, Source Sans Pro, and Source Serif Pro are all text fonts that can be used for map labeling. Figure by E. Guidero, Penn State Geography.
Figure 5.3 Impact, Giddyup, and Stencil are examples of display fonts that should be used judiciously.
Figures by E. Guidero, Penn State Geography.
In contrast to text fonts, display fonts have distinctive and decorative letterforms whose exaggerated design can be difficult to read. They have limited use in map design. You may use them in a title or logo, but avoid their use in the body of the map (figure 5.3). Some text fonts have certain variations, or weights, that are display versions of the text font. For example, Franklin Gothic has a book weight but also a heavy weight (figure 5.4). Franklin Gothic Book could be used for map labels, while Franklin Gothic Heavy could be used for titles.
Figure 5.5 shows a selection of serif fonts with example characters. Serifs, the small finishing strokes on letters, can make letterforms easier to distinguish. Compare the set of example characters above each font name to see how different the letterforms can be. Each capital I has serifs at the top and bottom of the stroke that are thin, blunt, or angular—a few examples of this variety. Serifs often assist with legibility.
Figure 5.4 Franklin Gothic Book, a text font, and Franklin Gothic Heavy, a display font.
Figure 5.5 Baskerville Old Face, Century Schoolbook, and Crimson Text are examples of serif fonts.
Serif fonts also include a subcategory known as slab serif fonts (figure 5.6). These fonts have blocky, slab-like serifs. Rockwell is a popular example. Although slab serif fonts do not often work well for small labels, they can be used for large labels and titles.
The next three examples are sans serif fonts—their letter forms lack serifs (figure 5.7). Trebuchet MS and other sans serif fonts designed for on-screen viewing cheat a bit by adding tails to letterforms, such as lowercase j and l. Technically, these are not serifs (except on the 1), but they do help with legibility.
Monospace fonts are not proportionally spaced like most fonts. Each character occupies the same width along the line, which produces a clumsy appearance in many design contexts (figure 5.8). Monospace fonts are useful for showing code snippets, but not for map labeling.
Figure 5.6 Rockwell, a slab serif font.
Figure 5.7 Fira Sans, TW Cen, and Trebuchet MS are examples of sans serif fonts.
Figure 5.8 Consolas and DejaVu Sans Mono are examples of monospace fonts.
Figures by E. Guidero, Penn State Geography.
Figure 5.9 The purple box around the word “palouse” represents the x-height of the Cambria font, as compared to the smaller x-height of Centaur. Both examples are the same point size (24).
Figure 5.10 Compare the legibility of the four fonts shown small, italic, and angled. This is a challenging but common situation for map labels. All examples were set at 12 points and captured at 72 pixels-per-inch resolution. A small amount of character spacing was also used to aid readability.
Figure updated by E. Guidero, Penn State Geography.
A single map will contain few fonts. Usually only two fonts are used, one serif and one sans serif. These two choices should complement each other. Choose two fonts with similar personalities, both informal or both formal in their manner, both modern or both classic. Fonts may be assertive, refined, official, welcoming, or casual. Choose a pair of fonts that give the right first impression of the map’s purpose. Often, a serif font is used to label hydrographic and other physical features, and a sans serif font is used to label cultural features. Getting carried away with fonts by including too many or choosing fonts that clash is a quick way to make an amateur-looking map. Pay attention to some of the small details of typefaces—they affect the legibility of labels as well as the overall map style.
Map type is often small, so mapmakers seek fonts with large x-heights, which are more legible. For example, at the same point size, Cambria has a larger x-height than Centaur (figure 5.9). When choosing map type, consider the ease of reading small letters when the map is seen at coarse resolutions on screen or when it is reproduced and letters are seen against patterns of tiny dots used to print background colors.
Consider stroke width as well. A font that has thicker strokes, or a lower stroke contrast (less variation in width), will remain legible after reproduction and against busy backgrounds. Thin strokes also cause problems with readability at coarse resolution. The thin strokes in Centaur are finer than the thin strokes of the same letters in Cambria. For example, the thinning of the lines at the top and bottom of the Centaur o are finer than those of the Cambria o (figure 5.9).
Recall the discussions of designing for media in chapter 4. If you wish map readers to use your map at the coarse resolution of desktop computers or television, for example, select a font that remains legible in that medium. Figure 5.10 shows four fonts with a challenging constraint: small, italic, and angled. Each label includes the word “minimum” above its name as an example with little to separate and distinguish letters. Some of these examples are almost illegible; Garamond italic, in particular, does poorly. Others maintain legibility fairly well. Verdana letters remain more distinguishable at the same point size and angle as the others because it has been designed for on-screen use.
Consider the state name “Illinois.” The capital I followed by two lowercase ls can be mistaken for each other, and when set close together, can be very hard to read (figure 5.11). Similarly, the lowercase l and numeral 1 can often be confused. This is an example of the importance of looking at whether letterforms have serifs or terminals that create differences among letters that are often confused.
The regular, italic, bold, and bold italic members of a font family are separate fonts that are installed individually on your computer. When you select a font from a list inside word-processing or mapping software, and then click buttons for bold and italic, you are usually choosing among four fonts, though only one name for the font displays in the font list. This is an important detail because a person working with a map on another computer will need all four fonts installed if the map includes regular, bold, italic, and bold italic styles.
In figure 5.12, you can see that italic is a separate font whose letter forms are different from regular, or roman, style. The shapes of the serifs are distinctly different, and the lowercase a takes on a whole new form. The italic is a companion font designed to complement the regular version. Italicizing is not simply a matter of slanting the characters.
If you do not have the bold, italic, or bold italic versions installed, some software programs will, however, slant and/or thicken the roman characters, creating what is known as faux-bold or faux-italic (figure 5.13). This is something that you should avoid. If you find that you need to use bold or italic but your system does not have it, use a different font.
Font families may also include many more weights than simply regular and bold. Figure 5.14 shows the full Alegreya Sans font family. Other, more extended font families include condensed versions, both sans and serif versions, and sometimes slab serif versions.
Figure 5.11 Notice how difficult it is to distinguish the different letters in some fonts, but not in others.
Figure 5.12 Regular and italic forms of the Adobe Garamond font.
Figure 5.13 Italic, faux-italic, bold, and faux-bold of the Adobe Garamond font.
Figure 5.14 The Alegreya Sans font family (OpenType).
Figures by E. Guidero, Penn State Geography.
You may be surprised by the way font names vary. Fonts that differ in appearance may have the same name, and fonts that look the same may have different names. These differences occur because different designers and companies seek to differentiate their products, and large companies offer the work of multiple designers and other companies in their libraries. Some font names have acronyms appended to clarify that it is owned by a particular company (such as MT for Monotype or MS for Microsoft) or to designate an intended use (such as UI for user interface).
Figure 5.15 shows three similar-looking fonts with the same name. Monotype Garamond was designed in 1922, and the other two were designed in 1989 and 2005, respectively. Although the latter two look similar, each has small differences. Monotype Garamond looks sufficiently different from the other two to appear to be almost a completely different font family. Its strokes have much higher contrast between the thick and thin sections. If you look closely, you can see that the serifs are shaped differently, and the letters have a squat appearance. While these may be individual small differences, they will collectively alter the look and feel of the map. If using one version of a font, such as Garamond, try not to include another version in the same map—it will look “off” to the viewer.
Figure 5.15 Three similar-looking fonts: Monotype Garamond, Adobe Garamond, and Garamond Premier Pro by Adobe. Figure by E. Guidero, Penn State Geography.
If you are using only Western characters (Latin-based languages), try to avoid typefaces designed for other character sets. For example, Microsoft Windows includes many typefaces with glyphs for non-Western languages, such as Japanese, Cyrillic, Cambodian, Urdu, Chinese, Thai, Hebrew, Hindi, Korean, and so forth. These typefaces include DokChampa, Batang, DaunPenh, Angsana, Browallia, Meiryo, DotumChe, Gautami, Iris, and Jasmine. These fonts are designed primarily for the non-Western glyphs, and the Western character sets included with them for convenience are not well designed and not recommended.
Maps often contain labels with special characters. An obvious example is the degree symbol ( ° ) and the characters representing minutes and seconds of a degree ( ´ and ´´ ). These are not the same characters as curly (or smart) single and double quotes ( ’ and ” ). Some place names have accents on letters or other diacritical marks that you will need to include to produce a correct map.
Figure 5.16 The Calibri character map with the degree symbol selected.
How do you produce these strange marks and characters? One way is to copy them from the character map included with the operating system and paste them into a text element in your map. Figure 5.16 shows the Calibri character map with the degree sign ready to be copied and pasted into a map.
The character map is also a good choice for examining the many symbols available in specialized fonts that do not include letters and numbers. One of the Wingdings fonts with icons and bullets is shown in figure 5.17.
Remember to test whether the special characters you use will export to the final format you intend for your map. Worst-case scenarios are that the special symbols will not display in the export, or they will crash other software. Or a special character may convert to a regular letter, a strange rectangle, or a question mark. Always check before you do too much custom character work.
Figure 5.17 The Wingdings character map with a mouse icon selected.
Fonts are technically miniature programs, installed within the operating system as a set of files. These files contain instructions to be used by the operating system and other software that display the fonts correctly. Remember, each letter in a font is a set of precisely set lines and curves. Font files include these geometric specifications as well as tables that determine how far apart the letters should be spaced when placed next to each other. The most common types of fonts used for graphic products are OpenType (O icons), TrueType (TT icons), and PostScript (red a icons). WOFF (Web Open Font Format) and EOT Lite (Embedded OpenType Lite) are used to show fonts on web pages.
Your computer probably has about one hundred typefaces pre-installed, and many applications install more fonts as the software is installed. These typefaces will probably be sufficient for your needs, but if you want to look further afield, beware of the glut of “free” fonts on the Internet. Many of these are poorly designed and do not contain the full character sets or enough additional styles (such as bold and italic) that you will need when you make maps. Be critical of the source of your fonts. Fonts are commercial products that vary in their quality of completeness and precision of the drawings of letterforms. Cheap fonts may be poor copies of the original finely drawn characters. There are many free and open-source fonts, however, that are created by reputable designers. Websites that offer these typefaces include Font Squirrel (http://www.fontsquirrel.com) and FontShop (http://www.fontshop.com/people/fontshop-team/fontlists/free-fonts).
It may seem like fonts that were installed by default on a computer are free. Although they were included in the purchase price of the computer and various software packages, you should know that, unless explicitly noted, fonts are intellectual property that are licensed when purchased. To legally produce a map with a customized design, you may need to purchase a font. If you give font files to other people so that they can see your design with the intended fonts, you may be breaking the license agreement you made when you or someone else purchased the font.
When you open a map file but do not have all the fonts used on the map, software will often prompt you to substitute another font. Font substitution may disrupt the look and fit of labels and symbols, length and spacing of text lines, and other design aspects that were carefully set within the map based on characteristics of the original fonts. Rather than redistribute the font files, which may be against the license agreement, one map design strategy is to use open-source typefaces that are distributed under the SIL Open Font License. This license allows the user to freely download, install, and redistribute the typefaces, even in commercial, for-profit products. Published or sold work may need to credit the font designer. Another option for preserving fonts is to choose an export option that can be set to embed fonts in the file (making the file size larger), such as PDF. This will depend on the specific types of permissions associated with the font. Some map readers may not have the administrative ability to install fonts on their work computer or mobile device, so this alternative is important.
Larger labels are used for more important map features. There are three ways to change the size of a text element on a map:
Characters can be set smaller or larger.
Space between letters can be added so that a text element is wider.
Lines of text can be spaced so a text element is taller or more compact.
Point size, character spacing, and leading are adjusted to accomplish these changes. This topic describes each of these characteristics.
Figure 5.18 Relative point sizes: 4, 12, 72 (72-point type is approximately 1 inch).
Figures updated by E. Guidero, Penn State Geography.
Type is measured in points. One point is 0.353 millimeter or 1/72 of an inch. Paragraphs in a book are typically set in 10- or 12-point type. The minuscule type in an ingredients list on a small food package might be only 4 points, and the large type for a wall map could be 72 points (about 1 inch high) (figure 5.18).
Different fonts can vary markedly in size even when you specify the same point size. In figure 5.19, all fonts are set to 16 point. The x-height of Goudy is indicated by the purple line, and you can see that the x-heights of Merriweather Sans and Bernard MT Condensed are much larger. In addition, uppercase letters, heights of ascenders (for example, the top part of d), and lengths of descenders (the bottom part of p) vary in size among fonts. Notice that the Merriweather Sans font (middle column) has short descenders and wide letters, and that Bernard MT Condensed has a tall x-height and narrow letters. Setting a particular point size is actually a fairly inexact measure because of the variation among fonts.
You will rarely be able to literally measure part of a 10-point letter and find it to be exactly 10 points high. The reason behind this variation harks back to the origins of type. The “10 point” size refers to the height of the small pieces of metal that carried the raised characters that were inked and pressed to paper. These metal blocks were necessarily larger than any of the individual characters in a font so they could accommodate their full range of shapes. The distance from the top of the ascenders to the bottom of the descenders will approximate the point size. This detailed knowledge is important for two reasons. You will need to experiment if you want to match a font size used in other text related to your map; you will only get an approximate size by measuring the letters. If you prepare a map design with one font and then change to another, it is likely that many of the labels will not fit together as you intended.
Figure 5.19 Three example fonts all set to 16 point. From left, Goudy Old Style, Merriweather Sans, and Bernard MT Condensed.
It is often tempting to create very small type on maps so the labels fit closely to very small features. But be considerate of your reader and remember that some map users have a difficult time reading tiny labels even with their glasses on. If you size type below 6 points, you may be creating an unreadable map. In addition, you need to size type larger if the map will be read from a distance or displayed at coarse resolution.
Kerning is the space between two individual letters. Tracking refers to all the spaces among a group of characters. To spread a label across an area on a map, one modifies the tracking (often referred to as character or letter spacing). GIS and design software allow you to set character spacing either in absolute points or as a percentage of the label point size. In ArcGIS, tracking units are expressed as a percentage of the point size of the text element. The example in figure 5.20 shows a label meant to span a large park on a map. The character spacing of the text is set to 200 percent, which means that another letter would fit in the space between letters. Given the variability of character size for different fonts, this is an approximate measure that requires some experimentation to get the desired effect.
It is common to increase tracking slightly for map labels to improve their legibility at small sizes. A small increase in tracking can also improve the appearance of curved type by making the letters less likely to tilt into one another, as shown in figure 5.21.
Figure 5.20 Text with 200-percent tracking.
Figure 5.21 Adjust letter spacing slightly so that characters are not crowded on curves.
Figures updated by E. Guidero, Penn State Geography.
Leading (pronounced “led-ing”) is the spacing between lines of type. The standard measure of leading is the distance between baselines, but it can also be specified as a percentage of point size or as an absolute amount by which spacing is increased. Large leading values are used to spread out a stacked areal label to express the extent of the region named. Leading is often combined with character spacing for this purpose.
It is common in map design to use slightly less than the default leading to move lines closer together. For example, when multiple line labels appear in crowded areas of the map, tight leading can unambiguously link them to their point features (figure 5.22).
For a paragraph with multiple lines, it is common for leading to be set at 120 percent of the type size. Lines of 10-point type measure 12 points from baseline to baseline, leaving a 2-point gap between the bottom of the descenders and the tops of the ascenders. This gap will appear larger or smaller depending on the style of the font (some descenders and ascenders are longer in some fonts than others). ArcGIS treats leading differently from most graphic design programs—the default measure of 120 percent leading is specified as zero points of leading (figure 5.23).
Figure 5.22 This map pair shows labels with regular leading (A) and the same labels with tighter leading (B). Data sources: US Geological Survey (USGS), NHDPlus, US Census Bureau. Maps by E. Guidero, Penn State Geography.
Although the leading number in ArcGIS is set in points, this number does not reflect the actual baseline-to-baseline measurement. Instead, it indicates increasing or decreasing spacing from the default. For example, if you set leading to 6 points, line spacing increases by 6 points from the default 120 percent gap (here, 2 points), resulting in a total gap of 8 points (figure 5.24).
Leading is an approximate measure for the same reasons that different fonts have different inherent sizes, even when set at the same point size. Figure 5.25 shows the same leading setting for Fira Sans, Adobe Text Pro, and Perpetua fonts. The baseline-to-baseline distance for Perpetua is indicated by an orange box. Notice that the leading is looser for the other two fonts, even though each is set at “zero leading” in ArcGIS.
Figure 5.23 In ArcGIS, 10-point type (enlarged) with 0-point leading (equivalent to 120 percent leading) creates a total of 12-point leading, leaving a 2-point gap between lines (the dot grid has 1-point spacing).
Figure 5.24 The leading for the Eugene Airport label was increased by 6 points. If you count using the 1-point grid, you can see that there is now an 8-point gap between the bottom of the descender (the g in Eugene) and the top of the uppercase letter (the A in Airport).
Because of these differences, it is important to settle on a final typeface early in the map design process. You may have to painstakingly change the point size and leading values if you start with a narrowly leaded font (like Perpetua) but change to a widely leaded font (like Fira Sans).
Figure 5.25 Note the differences in line spacing for these three fonts. All three are set to the same point size with zero leading in ArcGIS.
Figures 5.23–5.25 updated by E. Guidero, Penn State Geography.
Figure 5.26 Leader line positions were dynamically selected by ArcGIS as the “Nesting site” label was moved manually around the feature. (Only one of these labels would appear on the final map). Figure updated by P. Limpisathian, Penn State Geography.
In addition to the basic type characteristics of font, size, style, and spacing, there are a variety of effects you can use to augment text on maps. Callouts clarify the link between a location and its label, which is especially useful in densely labeled areas. Shadows and halos improve the contrast between text and nearby map features. Callouts, shadows, and halos do not change the form of characters in a text element. They are additions to the characters. Thoughtful use of type effects improves map clarity and enhances design.
Callouts use graphic elements, such as leader lines, that explicitly link a text label to a point location. Use callouts sparingly. Reserve them for when you need to identify points in a densely labeled or otherwise inaccessible location on the map. Setting a callout for every label on a map produces visual clutter. A leader line can connect to any part of the label: the beginning, the end, or the middle—whichever is closest to the feature (figure 5.26). The goal is to use leader lines sparingly and to make them as short as possible.
A callout styled as a cartoon balloon is too dominant for most mapping contexts, though in journalistic mapping it can be appropriate to highlight a few locations discussed in a story. A filled callout or callout balloon also blocks underlying map information. GIS and graphics software provide a lot of flexibility in changing the design of callouts (figure 5.27).
Figure 5.27 The “Belmont Harbor Light” label shows ArcGIS’s default callout style with color adjustment. Three other styles are suggested. Figure updated by E. Guidero, Penn State Geography.
Shadow effects, when skillfully applied, increase the legibility of text on maps. A shadow is simply a graphic copy of the shape of the characters, offset, and rendered in a contrasting color to enhance text prominence. It is defined by the offsets applied, which adjust the perceived height of the text above the page (figure 5.28). A smaller shadow can increase legibility of text by adding contrast with background colors and features. The shadow shown in figure 5.29 is offset only 0.5 points right and down. These thinner shadows enhance the text but are less distracting than shadows that are well offset from the text.
Figure 5.30 demonstrates how shadows improve contrast between both light or dark type and a multicolored imagery background. Without a shadow, the white label is hard to read over light areas in the image, and the dark label is hard to read over dark areas.
Use shadows sparingly in map design, and remember to check how a shadow looks in the final medium for the map. Make sure your shadow effect exports properly if you will be moving your map out of the GIS environment. Vector export formats may omit, offset, or reshape the shadow letter forms (see chapter 4).
Figure 5.28 Emphasis has been added to text using a gray shadow with a significant 2-point offset to the right and down. The result is that the text appears to be doubled. Figure updated by P. Limpisathian, Penn State Geography.
Figure 5.29 This shadow with a half-point offset down and right will provide subtle contrast against other map detail. The result appears to be slightly floating above the surface. On a crowded map, this style of shadowed text is more legible than the effect in the previous example. Figure updated by P. Limpisathian, Penn State Geography.
Figure 5.30. Labels are difficult to read against the satellite image background (A). Shadows below white and black text improve contrast against the image (B). Data source: USGS. Maps by E. Guidero, Penn State Geography.
Halos function much like shadows. A halo can be used to decorate text or to subtly improve contrast with the background. The halo effect is also called letter casing by cartographers and a stroke outlining letters in design software. In figure 5.31A, light labels are difficult to read on light roads. Figure 5.31B shows the halo color matched to the background color. The halos break lines where they are close to text so that text remains legible.
Halos are best when they are subtle and relatively thin, especially when used for smaller text and unobtrusive breaks, although occasionally a bolder halo may be useful. The text has a 0.5-point halo in figure 5.32A. However, because the background is so busy, it is difficult to see the text. With that kind of background, a larger halo is needed, as shown in 5.32B. Notice that this size halo fills most open areas within letters and small spaces between letters, without masking too much of the area around the text. With a simpler background, as shown in 5.32C, a narrow halo is a better choice. The goal when selecting a halo size is to clean up small pieces of line or other content that show between letters while masking as little of the underlying map information as possible.
Figure 5.31 Labels are difficult to read against road lines (A). A 1-point halo on the 14-point type breaks the lines and increases legibility (B). Data source: US Census Bureau. Map updated by E. Guidero, Penn State Geography.
Figure 5.32 A 0.5-point halo around 10-point type (A) should be enlarged to improve legibility against a busy background (B), but is sufficient for a simpler background (C). Figure updated by E. Guidero, Penn State Geography.
In figure 5.33, the road names set in 8- and 10-point Segoe UI are offset 1 point away from the road features (A). A thin 1-point halo was then applied to the labels to mask interfering road segments (B). The challenge is to choose a halo size that is an adequate compromise between improving legibility while not obscuring too much map data. For example, the halo here is not quite large enough to mask the road segment that appears between the “Laporte” and “Ave” portions of the label along the left side.
However, it seems some of the halos should be a bit thinner because lines are being unnecessarily nibbled near the bottom of letters. For example, the “Romona Rd” label (upper middle of map) chews into the road line it names. This problem could be improved by manual repositioning of labels or by adjusting their automatic placement specifications. The contradiction of halos that seem both too large and too small suggests that a suitable halo thickness has almost been achieved. If you were preparing this map for print, it would be important to check halo widths in a high-resolution print because small sizes are rarely rendered accurately on screen.
Figure 5.33. Map A has no halos. Notice the interference of street lines with the label letters. Thin halos that produce breaks in the street lines for the labels were used in map B. Data source: US Census Bureau. Maps by E. Guidero, Penn State Geography.
Map labels sometimes need to span many background colors, both dark and light. This varied contrast makes it difficult to select a single color that remains readable. As with shadows, halos that contrast with the text and background colors (rather than matching the background) can improve legibility. For example, type that is the same color as one of the vegetation symbols is nearly invisible (figure 5.34). Halos add the contrast necessary to read the labels (figure 5.35).
Figure 5.34 Text that borrows its color from underlying features is unreadable on satellite imagery of Glacier National Park.
Data source: USGS. Maps by E. Guidero, Penn State Geography.
Halos are useful for mapping because they can help ensure that text is legible on varied backgrounds. They also have drawbacks. They do not export well, they increase processing times, and they can produce excess visual noise around labels. Trying to create subtle halos that match the background can be a challenge if multiple background colors are used on a map. Use of contrasting fill, line, and type colors are alternatives to haloing type, so do not become overly reliant on this useful effect. ArcGIS’s advanced symbol-level masking is also an alternative, particularly when labels overlay multiple line and polygon symbols.
Figure 5.35 Halos that contrast with both the text and the background improve legibility.
