Source: alexpoole.info Translation: Vladimir NikonovIn 1998, when Times New Roman was still widely used online, my then-boss demanded that we design our websites in Arial because she hated serif fonts. Did she find sans-serif fonts more legible, or was it just a matter of taste? In 2003, as part of my master's degree work, I reviewed over 50 practical studies in the field of typography and found a definite answer. IntroductionSeveral decades in scientific and typographical There is an ongoing debate in circles over a seemingly insignificant question: is a serif font more legible and a sans serif font less legible? To this day, no one has provided a definitive, conclusive answer to this question.1. Definitions Legibility vs. Readability Legibility of a font depends on the accuracy of its elements, which usually means the ability to recognize individual letters or words. However, readability is directly related to the optimal layout and layout of the entire body of the text. An illegible font, no matter how you arrange it, cannot be made readable. But the most legible font can be rendered unreadable if it is too wide or too large (or small) in size for certain purposes (Dowding 1957; Lund 1999). Typographical Features There are many elements in a typeface that affect its legibility. Serif font /sans serif Serifs are the small strokes that end each element of a letter. A sans-serif font does not have these small finishing touches. Sans-serif and serif fonts Point size Point size is the parameter most often used to describe the legibility of a font's appearance, but it can be deceiving. The size of the printed character is a legacy of the printing system, when each letter was placed on a separate metal pin. The size of the printed character depends on the size of the metal pin, and not on the actual size of the letter. The letter does not have to occupy the entire surface of the pin, so two fonts with the same nominal size of the printed character may well have different sizes. (Bix, 2002). The difference between actual letter size and point size. Height of lowercase letters without leaders (x-height) The height of lowercase letters without leaders is determined by the height of the lowercase letter “x” in the font. It is often a more accurate indicator of actual type size than the size of the printed character (Poulton, 1972; Bix, 2002). x-height Intraletter spaces (counters) Intraletter spaces are empty spaces inside a letter. They are also good indicators of actual font size. Intraletter gaps (counters) Protruding and hanging strokes (ascenders and descenders) Protruding strokes are strokes of letters that protrude: for capital letters - beyond the top line of capital letters (ЁИ), for lowercase letters — beyond the top line of lowercase letters (beif). Hanging strokes are strokes of uppercase and lowercase letters that hang beyond the bottom line of the font (DSchTsruftsu). Ascenders and descenders 2. Facts Review of comparative studies of serif and sans-serif fonts for legibility There is a wealth of research demonstrating the complete lack of differences in the perception of serif and sans-serif fonts. (Tinker, 1932; Zachrisson, 1965; Bernard, 2001; Tullis, 1995; De Lange, 1993; Moriarty & Scheiner, 1984; Poulton, 1965; Coghill, 1980). However, a number of specialized studies claim to demonstrate the superiority of serif fonts (Robinson , 1983; Burt, 1959; But these studies have been criticized from a methodological point of view (Lund, 1997, 1998, 1999). Particularly interesting is the case of Sir Cyril Burt, well known in psychological circles for falsifying research results. It turned out that he is really inclined to fabricateresults in their typographic work (Hartley & Rooum, 1983). Unfortunately, many researchers, typographers and graphic designers tend to uncritically use Burt and Weildon's materials, so many informal web resources continue to use serif fonts, unjustifiably most effective. The most frustrating thing, however, is that in more than 100 years of research into type legibility, scientists have failed to formulate a specific theoretical framework for the role serifs play in type legibility (Lund, 1999). Nor did they manage to gain any tangible significance for their work within the typographic community (Spencer, 1968). Arguments for serif fonts Serifs are used to give a horizontal direction to the eye; the absence of serifs promotes a vertical direction of view, which is considered unfavorable compared to a horizontal direction (De Lange, 1993). These are the most common arguments in favor of a serif font. However, it cannot be said unequivocally that serifs “direct the path of the eye.” In 1878, Professor Emile Javal from the University of Paris found that the eye when reading does not move smoothly along the line of text, but makes a series of jumps, which he called saccadic jumps ( Spencer, 1968, Rayner & Pollatsek, 1989). Unfortunately, many graphic designers and printers continue to use this argument to support the use of serifs, due to the lack of consistency with the scientific world. Serifs help increase the space between letters and words, which improves legibility. Serifs are not needed to control the spacing between letters and words—in fact, serifs are unfortunately not good for this purpose. In traditional typography, letter spacing is achieved by lengths of metal placed between letters and by the distance between the letter shape and the edge of the printing block. It is even easier to manage letter spacing on modern computerized printing equipment (Sassoon, 1993; Rubinstein, 1988). Serifs are used. to enhance the contrast between individual letters, which aids in letter identification. Carefully conducted research has shown that words can be recognized as quickly as letters during eye fixation, and that individual letters can be identified more quickly when they are part of a word. This “word superiority effect” proves that serifs are not required to differentiate individual letters (Reynolds, 1979). Serifs link individual characters into a “single word.” The simple gestalt created by the distance between words is quite enough to connect the letters into a “single whole.” Moreover, it is obvious that additional elements such as raised and drooping strokes have a much greater effect on word recognition than serifs (Poulton, 1965). Readers prefer texts written in serif fonts as they are more legible. Many previous studies had in fact shown a preference for serif fonts (Tinker, 1963; Zachrisson, 1965). However, Tinker commented that such texts were perceived as more legible, in large part due to the fact that these fonts were more familiar to the reader.40 years ago, sans serif fonts were not as common as they are today. And if such a study is carried out now, it is not surprising that the results will be completely opposite. In fact, many recent studies show that Internet users prefer sans-serif fonts for reading text online (Boyarski, 1998, Bernard, 2000-2001, Tullis, 1995, Reynolds, 1979). It is important to remember that in almost all studies factors influencing reading comprehension, reader preference, or perception of legibility, and user behavior are different and inconsistent (Lund,1999). Serif fonts are still used because sans-serif fonts cause fatigue. It is often claimed that reading a lot of text written in sans-serif fonts causes fatigue. But there is no obvious evidence for this, since there was no test for fatigue levels differences were noticed when reading texts written in serif and sans serif fonts. Moreover, “there is no objective method for measuring fatigue levels. The subjective determination of the level of fatigue is subject to the combination of a host of factors that may be completely unrelated to the subject of our experiment" (Reynolds, 1979). Arguments in favor of a sans serif fontSerif is already a historical artifact. This is largely true, especially in light of that all attempts to justify serifs look unconvincing in retrospect. Many researchers attribute the invention of serifs to the Romans, arguing that “the ancient Romans... when writing on stone, finished each stroke with a stroke (serif) in order to smooth out the surface irregularities created by the tool” (Craig, 1980; Bix, 2002). Others claim that "the brush sketch, before being executed in stone, resulted in a serif at the end of most strokes" (Bigelow, 1981; in Rubinstein, 1988). Whatever the reasons for the origin of serifs, serifs have been in use for so long that legibility the perception of a serif font can be attributed to habituation of the font—readers tend to find the font they are most accustomed to more legible (Tinker, 1963; Zachrisson, 1965). Sans serif fonts are better readable on the Internet, although screen reading research suggests that There is no difference between reading on a screen and reading on paper (Dillon, 1992; Bernard, 2001), this statement has some inaccuracies. When fonts are digitized for computer use, the shape of the letter must fit a relatively small pixel grid. Which often leads to so-called “stepping stones” (Rubinstein, 1988). Many web professionals, such as graphic designers, find that this relatively low resolution is not efficient enough to produce quality serifs. And that a sans-serif font lends itself more naturally to digitization and is clearer, and therefore more legible. Digitized fonts must fit into a relatively small pixel grid. However, this is not supported by the results of recent research (Bernard, 2001, Boyarski, 1998, Tullis, 1995, De Lange), and evidence that for legibility of perception it does not matter what font is typed. on the Internet. Sans serif font is better at small sizes. A sans-serif font can easily withstand copying and blurriness due to the simplicity of its shapes. There is reason to believe that serifs can create visual noise at small sizes, because they take away the area of ​​the main body of the letter (Morris, 2001). But this has not been confirmed by continuous reading tests (Poulton, 1972). Other factors, such as stroke thickness, letter spacing, and the height of lowercase letters without descenders, are likely to have a much greater influence on the identification of a letter shape, regardless of whether it has been copied or reduced in size (Poulton, 1972; Reynolds, 1979). .Sans serif font is better for teaching children to read. Books for children are often printed in sans-serif font because... teachers believe that the simplicity of a letter's shape makes it more recognizable (Coghill, 1980, Walker, 2001). But studies with children have found no difference, no matter what font the text was written in (Coghill, 1980; Zachrisson, 1965; Walker, 2001).3. Conclusions What initially appeared to be a clearly dichotomous issue of serif versus sans serif fonts has resulted in a whole body of research consisting of weak opposing statements; one after another, studies were carried out only to come to the same conclusion: “there is no difference.” Why is this so?