Thursday, October 30, 2014

Introduction to italics, part one

At this point, we’ve completed the entire ASCII block (save for the at sign (@) ).

Now, we could go on to do one of four things. We could add more glyphs to the font—fill out the Latin Supplement block with accented letters like ‘ å ’ and a few moderately common symbols like ‘ ¢ ’ and ‘ ° ’. We could go back and actually kern the letters we’ve already made to make the spacing look nicer. We could also derive the bold weight, or we could create the italic version of the font.

I recommend doing italics first. That is because between these four projects, creating italics will add the most value to your font. A functional font family includes a regular, or roman style, and an italic style. In many cases, italics are mandatory, for setting names of works of art, for example. They are also the best way to emphasize words in text. What about deriving the bold weight, you may ask? Well, bold is much, much less commonly used than italics. A font family does not need a bold weight to be functional, though it can be a selling point. Bold is usually used in contexts where it would also be appropriate to use a completely different typeface (headings, for example). The bold weight is also much more straightforward though tedious to design. It is also much easier to embolden both a roman and an italic typeface to create the bold and the bold italic than to italicize both a regular and a bold font to make the italic and bold italic. That is why I like designing italics before bold.

What are italics?

Italics are much more different from the regular style than the bold style is. In serif type, it is possible to machine-produce a passable, though ugly, bold font given a regular font, since emboldening is a relatively straightforward and mechanical operation. Indeed, dozens of weights can be automatically interpolated (with quite good results) from just three base fonts (extra-light, regular, and black). However it is impossible for a computer to automatically produce the italic style of a serif font (something the opposite is true in sans serif type, where the “italic” is very close to a machine slanted version of the original, and the different weights are given much more consideration). That is because the serif italic letterforms themselves are drastically different from the roman letterforms. Italic letterforms are much closer to calligraphy than roman letterforms are. In fact, the italic hand is one of the most natural styles of calligraphy. This is evident in the strokes of the italic letterforms, which are more cursive than those of roman letterforms.
(Somewhat messy) italic calligraphy of some lyrics from How You Get The Girl, in Taylor Swift’s 1989. While handwritten, this calligraphy is essentially the same as italic type. Compare this with the Proforma Italic sample below.

For comparison, here is a sample of the italics from the Proforma font (grouped by letter shape):

The letterforms are much more free-flowing and give a completely different texture than the roman style does. Here are some important differences.

Sunday, October 26, 2014

Design notes: the ampersand ( & )

The ampersand (especially its italic form) is one of those glyphs that sees fairly little usage but for whatever reason type designers love obsessing over.

The ampersand, like the question mark, percent sign, and several other punctuation marks evolved from, you guessed it, a ligature—in this case, an ‘Et’ ligature—Latin/French for “and”. In most typefaces, the ‘E’ and ‘t’ have become so tangled up in each other that they are barely recognizable (italic ampersands are much more reminiscent of the ‘Et’). The pretzel-like form of the ampersand is very difficult to describe—your best bet is to just “look” at ampersands and try to copy them. The symbol has a leg that resembles the leg on the ‘R’, and its two bowls are more asymmetrical versions of the ones on the numeral ‘8’. It also contains a serif on its arm, which is similar to the one on the ‘v’ and rather unremarkable except that it is usually skewed with the longer side facing inward. Ampersands are usually the capital height, but have the stroke weight of the lowercase letters.
An ampersand from Floribunda

Saturday, October 25, 2014

Design notes: miscellaneous punctuation ( * ^ ` ~ )

There exists within the ASCII block a handful of assorted miscellaneous symbols—the asterisk ( * ), the raised caret ( ^ ), grave mark ( ` ), and tilde ( ~ ). The forms of these symbols are poorly defined, so type designers have a lot of free reign over these symbols. Except for the asterisk, these symbols are also incredibly rare, serving no purpose and having no use in proper typesetting, so they are kind of a typographical backwater.

Asterisk ( * )

The asterisk, or “star”, is the only one of these four symbols to be typographically significant. All asterisks resemble a star shape, but renditions of this glyph vary wildly among typefaces. An asterisk can have five, six, seven, or eight points, and occasionally many more.

The arms of an asterisk can be wedge shaped, pointing inward or outward, or they can be teardrop shaped. Sometimes the arms are not joined at the center. An asterisk is usually about half as tall as a capital ‘I’, and its upper edge is often affixed to a font’s capital overshoot. Many asterisks, particularly in old style typefaces, are rotated about ten degrees clockwise from a normal “up-pointing” star orientation.
Properties of the asterisk and caret in Floribunda.
Properties of the asterisk and caret in Floribunda.

Friday, October 24, 2014

Design notes: slashed characters ( < > # % $ )

From the various hairline characters and slashes we’ve designed, several more glyphs can be derived— ‘<’, ‘>’, ‘#’, ‘%’, ‘$’. These symbols, however, require slightly more artistic consideration, and their design varies much more among fonts than that of the hairline characters and slashes.

Greater than and less than symbols ( < > )

The greater than/less than symbols are chevron-shaped characters used in mathematics, computer programming, and occasionally as terminal punctuation, serving a purpose similar to ‘?’ and ‘!’.
Less than signs from various typefaces. The acute angled form is by far the most common.
Less than signs from various typefaces. The acute angled form is by far the most common.
The design of ‘<’ varies greatly among typefaces. There is no single angle prescription, like there is for ‘/’ or ‘v’. Most fonts compress the glyph vertically, rendering a shape that’s wider than it is tall (giving it an acute angle). Others arrange the arms at right angles, like Proforma does. Neither form is easier or harder to design than the other, nor is one form superior to another. The orthogonal form can be made from the ‘+’ glyph, its arms stretched to √2/2 times the x-height. The stretched form can be made from two rotated slashes, perhaps with beveling added to the vertex.
Some properties of the ‘<’, ‘>’, and ‘#’ symbols

Octothorpe (number/pound sign, hashtag) ( # )

The octothorpe, also known commonly as the hashtag, or less commonly, a “number sign” or “pound sign”, is a grid-like character which is usually paid little attention in type design. The octothorpe is smaller than the capital height, and centered within it. It is often, though not always, slightly oblique; less so than the slash though. The horizontal cross strokes may be made thicker than the vertical/oblique strokes. The four strokes should appear to visually delineate nine congruent spaces.

Sunday, October 19, 2014

Design notes: brackets and braces ( [ ] { } )


Brackets are squarish symbols which are somewhat common in prose (indicating an insertion in a quotation, or a footnote), though less common than the parenthese. They are also heavily used in mathematics, and as an uncommon emoji.

Brackets are not all that hard to design at all. They can be made from one side of a capital ‘I’, stretched to the dimensions of the parenthese. Oftentimes the bracketing (no pun intended) of the serifs is made more horizontal to give the bracket character a more shelf-like shape. The stem of a bracket is thinner than a normal stem, but it is usually thicker than the vertical bar and the hairline characters. Brackets, while resembling them, are not hairline symbols.


Braces (sometimes called “French curly brackets”) are a third type of parenthetical character. Braces are not naturally occurring glyphs. They have no usage in any form of writing, except as a math symbol used to denote sets such as x = { 1, 3, 5, 6 }. The only reason they even made it onto the standard keyboard is because computer programmers make heavy use of these glyphs in their code (which should be set in monospaced type anyway).

Braces are difficult to describe. The best way to learn how to draw them is to look at examples from existing typefaces. Braces are also perfectly symmetrical (a rarity in type design), so you can draw the top half and just mirror it to make the bottom half.

Saturday, October 18, 2014

Design notes: hairline symbols ( | / \ + = _ )

There exist a handful of lineal symbols—slashes, the equals sign, etc. which are relatively simple to design, even if some confusion exists about their dimensions and placements. In almost all typefaces, these characters are plain rectangles.


Slashes—‘/’, ‘\’—sometimes called solidusses (or solidi) are hairline characters derived from a third glyph—the vertical bar ‘|’.
Dimensions of the various slash characters
Dimensions of the various slash characters
  • The vertical bar is optically the same width as the en-dash (or em-dash), though since it’s oriented vertically, it needs to be a bit heavier (like 0.044 ems vs 0.040 ems).
  • The vertical bar is supposed to extend across the entire em height. That is, the vertical bar is 1.0 ems high. But not all fonts obey this rule. There is some confusion over whether or not the bar is continuous. The vertical bar is always continuous—the broken form ‘ ¦ ’ is a separate character.
  • Spacing of the vertical bar varies across typefaces. I like to go with sidebearings that are twice the width of the bar itself.
  • The slash is the same width as the vertical bar, just inclined exactly 18 degrees away from the vertical. This figure is incredibly consistent across typefaces; in fact, many designers will naturally create slashes tilted 18 degrees without even knowing about the rule, simply because that angle looks the best to the eye. The backslash is an exact mirror reflection of the slash.
  • For most typefaces, the top of the slash terminates at the ascender line. The descent of the slash varies—some fonts bring it almost to the descender line, others to the bottom of the parentheses, and others just below the baseline.
  • For many typefaces, the slash is just a long, thin, tilted rectangle. Many other designs however, include slashes that are shaped like parallelograms, with flatly-cut bottom edges.
  • Spacing the slash is more important that most other punctuation marks, since its leaning shape and common occurrence adjacent to letters (like in ft/sec) make it prone to collisions. The general rule is to space the slash so that if it extended two-thirds the way to the baseline, the ends would just barely extend over the shank (metrical width of the glyph, shaded in the above diagram). This holds even if the slash itself terminates above or below the “two-thirds” line.

Design notes: terminal punctuation ( ? ! )

The question mark and the exclamation point are some of the most well known punctuation marks, despite the fact that they are fairly uncommon in most prose—the reason I didn’t include it in the essential punctuation posts. The question mark is not rare, though it is not common in most writing—the exception being text with an abundance of dialog. The exclamation point is rarer; most manuals of style actually recommend against using it (though in conversation, the exclamation point is close to default, and it is the period which must be used sparingly).

A long time ago, the question mark and exclamation point were ligatures of a ‘Qo’ and an ‘Jo’—the ‘Q’ standing for “question” and the ‘J’ standing for “joy”, respectively (the ‘o’ comes from the Ancient Latin versions of these words, which both ended in ‘o’). The ‘Q’ degenerated to become the curved stroke over the ‘o’, which collapsed into a dot, forming the ‘?’. The  ‘Jo’ degenerated in a similar way, with the ‘J’ becoming the stem of the exclamation point (though the ‘J’ was written as a straight vertical stroke, ‘I’, back then anyway).

The question mark

The question mark is one of the hardest punctuation marks to design.
  • For whatever reason, the stress on the question mark is opposite that of conventional western calligraphy. Why? It probably has something to do with the fact that when this glyph is drawn in pencil or ballpoint pen, it is most natural for us to put the most pressure on the diagonal strokes.
  • The stroke of the question mark terminates about one period’s width above its lower dot. The stroke’s finial rarely “points” at the dot, rather it’s directed to a location somewhat to the right.
  • The height of the question mark varies among typefaces. Some make it the capital height, others raise it all the way to the ascender line (usually more modern style typefaces). Many make the question mark somewhere between those two heights.

Friday, October 17, 2014

Design notes: loopy numerals ( 6 9 and 8 )

The numbers ‘6’, ‘9’, and ‘8’ are certainly the hardest numerals to design. They contain no straight lines and no obvious guidelines to design from. These numerals largely have to be drawn by eye, though we can employ a circle-method similar to the one we used for the ‘s’ to make the ‘8’.

6 and 9

  • Perhaps the best guide we have for drawing the ‘6’ is the lowercase ‘b’, as the stress on the two glyphs’ bowls is similar. The ‘6’ is also about the same width as the ‘b’.
  • The joint between the tip of the bowl and the numeral’s arc (left side) is rarely smooth. Usually there is a vertex on the inside.
  • The terminal often extends almost all the way to the right extreme of the glyph.
  • The ‘9’ is a perfect 180 degree rotation of the ‘6’. There are very few adjustments that should be made to it (though some designers like to shrink the bowl of the ‘9’ slightly). That means that if your ‘9’ looks funny, then it’s probably indicative of a poorly drawn curve in your ‘6’. In this way, the ‘6’ and the ‘9’ are, in a way, designed simultaneously, as each can help you improve the other.

Wednesday, October 15, 2014

Design notes: the numerals 2 5 7 and 4

The ‘2’, ‘5’, ‘7’, and ‘4’ can be derived from the ‘1’ and ‘3’.

2 and 5

The ‘2’ and the ‘5’ take the upper and lower bowls of the ‘3’, respectively, and combine it with a stressed horizontal stroke.
  • The ‘2’ is a hard glyph to design. The upper bowl of the ‘3’ has to be extended out all the way to the baseline, where it meets a stressed arm that ends in a beak serif.
  • The thickest part of the ‘2’ is the upper right portion of its bowl. The thinnest part is the joint between the bowl and the arm.
  • The joint is never a loop (as the ‘2’ is often handwritten), and it almost never bends back out (like it does in some poorly designed sans serif fonts). The vertex is thinner than the arm is.
  • The arm of the ‘2’ is slightly wider than the bowl is. It also carries stress, but is geometrically thinner than the stem width for optical reasons.
  • The lower bowl of the ‘5’ is almost traced from the lower bowl of the ‘3’. The terminal curves up less, however, and at least in the lining ‘5’, the top part of the bowl angles up before curving down.
  • The vertical hairline linking the ‘5’s flag (horizontal stroke) and bowl is sometimes angled slightly to the right.
  • The flag of the ‘5’ either has no serif (shown above), or has an overshooting upturn serif.

Tuesday, October 14, 2014

Design notes: The numbers 1 3 and 0

1 and 0

  • Numbers are often slightly shorter than capital letters. It can be convenient to set the capital line as the overshoot for the numerals, and the numeral line slightly below it. Lining numerals are also the same weight as capital letters.
  • The ‘1’ is a combination of the stem of the capital ‘I’ and the head serif of the ‘l’. The serifs of the ‘1’ are significantly longer than either letter’s.
  • The flag (top serif) of the ‘1’ is almost never curved, like it is in handwriting and some grotesque sans serifs.
  • The ‘0’ is about the same width as the lowercase ‘o’, just taller. Remember to match weights with the capital ‘O’.

Monday, October 13, 2014

Introduction to Arabic numerals: what are old style and monospace numerals?

Arabic (or more accurately, Indian) numerals are single character numerical glyphs like ‘1’ or ‘6’. Perhaps because of ease of access, Arabic numerals are somewhat overused, but they are mandatory for writing out numbers with more than two digits.

It’s not commonly known that numerals come in both capital and lowercase forms. Capital, or lining numerals, are the type most people are most familiar with today, probably because they are almost always the default. Lowercase, or old style numerals, are rarer, but still common in professional typesetting. Some type designers refer to lining and old style numerals as lining and old style figures—this means the exact same thing as numerals.

Numerals also come in different widths—monospace (or tabular) and proportional. Monospace means that all the numbers (glyph plus sidebearings) have the same width; proportional means that they only take up the space they need. This means that a font might include up to four different styles of numerals—monospace lining (usually the default), proportional lining, monospace old style, and proportional old style. Many fonts also include lowercase versions of symbols like the percent sign and currency symbols to go with the old style figures.
A font might include up to four different styles of numerals.
(The font used here is Warnock Pro, by Robert Slimbach).

Sunday, October 12, 2014

Design notes: essential high punctuation ( – — ‘ ’ “ ” )

Many amateur type designers are unaware that the punctuation marks that fontforge shows by default in the ASCII block (or the Latin Supplement block right after it) are not the only punctuation characters in common use. High punctuation marks—characters like the dashes ( – — ) and curly apostrophes and quotes ( ‘ ’ “ ” )—live in another Unicode block (glyph encoding scheme) called General Punctuation, buried beneath a myriad of Vietnamese glyphs and Polytonic Greek letters.

General Punctuation encodes about a hundred assorted punctuation characters that couldn’t be fit into the original ASCII block. Most of them are whitespace characters (we’ll get to those) and obscure glyphs like ‘※’. About a third of them however are commonly used in professional typesetting—including but not limited to the aforementioned dashes and apostrophes, the primes ( ′ ″ ), the ellipsis ( … ), the asterism ( ⁂ ), the daggers ( † ‡ ), and the permille sign ( ‰ ). Any decent font should eventually include these characters (though the primes and asterism are often overlooked), but the essential members are the dashes and curly apostrophes.

To jump to the General Punctuation block, you have to change your font’s encoding to Unicode ( EncodingReencodeISO 10646-1 (Unicode, BMP) ) and go to ViewGoTo and select General Punctuation from the dropdown menu on the dialog (or just scroll down until you see punctuation marks). Once you touch the slots you want to create, you can reencode your font back into the regular Latin-1 encoding and fontforge will remember the extra glyphs and include them at the end of your glyph table.


Dashes ( – — ) should not be confused with hyphens. They have a different grammatical meaning, even if they are often misused. Dashes come in two lengths—a short dash called a nut or en dash, used for ranges (120–145 miles), conflicts (Iran–Iraq war), scores (54–7), and occasionally as an abbreviation for the word “to” (the New York–Albany train)—and a long dash often simply called a dash or more specifically, an em dash. The em dash is a pause mark that has a meaning somewhere between a semicolon, comma, and colon. Many people who don’t know how to type these characters—or are using a font that doesn’t contain them—replace the en dash with a hyphen (-) and the em dash with two consecutive hyphens (--). Which is wrong. Others simply don’t use dashes, which is just silly. Look in any professionally typeset publication and you will find an abundance of dashes.
The em dash is almost one em long. The en dash is about half as long, and the hyphen about a quarter as long.
The em dash is almost one em long. The en dash is about half as long, and the hyphen about a quarter as long.
Dashes are almost always perfect rectangles, with no beveling or other calligraphic traits. Historically, the em dash got its name because it was supposed to be one em long (the width of a letter ‘M’, in the old days). The en dash was half as long (the width of a letter ‘N’ back then). Of course, just as the ‘M’ and ‘N’ are no longer bound by such strict proportions, neither are the dashes. Em and en dashes in most fonts are about ten percent shorter than a perfect em square and 1:2 em rectangle, respectively. In Floribunda, the en dash is 0.42 ems (420 font units) long, and the em dash is 0.9 ems long. Dashes sit at the same height as the hyphen does, but are often somewhat thinner. Thinner dashes tend to look more elegant.

Saturday, October 11, 2014

Design notes: The first punctuation marks

There are about seventeen punctuation marks in common use in prose:  . , ; : ? ! ' " - ( ) ‘ ’ “ ” – and —. The nine simplest will be covered in this post.

The period and the comma

  • The period is basically the same as the tittle (dot) on the ‘i’, though often slightly larger.
  • Periods and other very small baseline-dwelling punctuation marks require padded sidebearings so that they are legibly distinct from the letters before them. Their sidebearings are twice as wide as a typical rounded letter’s.
  • Periods sit on the baseline, with a very small undershoot.
  • The comma is a period with a tail. The tail extends about one period-height below the baseline.
  • The head of the comma is usually a bit thinner than the period, since the comma is longer. It is also a good idea to have the head of the comma be somewhat evocative of the terminals in its typeface.

Friday, October 10, 2014

Testing your font in print

We’ve made test texts before, but those were all at the sentence or the paragraph (at a stretch) level. Now that I have a full upper and lowercase glyph set, spaced and all, and some rudimentary punctuation marks, I can set entire passages of text in Floribunda. (Since the last post, I’ve spaced the capitals and added some basic punctuation marks— ‘.’ ‘-’ and ‘,’ . Spacing capitals is done basically the same way that the lowercase letters are spaced; punctuation will be covered in a later post).

Producing test pages is not hard. Just grab some text from a random Wikipedia article (preferably one that contains few numbers or special characters), and format it into columns set in your font. Throw in different-point-sized and ALL CAPS text generously—the capital text in mine is set in faux small capitals (the most bearable of the four typographical ‘faux’s, in my opinion anyway)—and you might want to include some knockout and grayscale type as well. Then just print it out on paper (please be kind to the environment and reuse some computer paper that’s been printed on on the other side—type design does not require blank new paper). Here is the page I used; the passage is taken from the Wikipedia article on Auroras.
Note that the ideal body text font size to use is much smaller than you’re used to specifying. For most typesetters, 12 point type is clumsily gargantuan, and generally reserved for headings and other display text. Most body text in magazines and textbooks is set in about an 8–9 point font (this corresponds to about a 10–11-pixel em square). Don’t believe me? Print out some nine point text and compare it to a copy of TIME magazine. The type will probably be about the same size.

Tuesday, October 7, 2014

Design notes: The capital R and K

The last two letters to design are the ‘K’ and the ‘R’. The ‘K’ should be made first, since the tail of the ‘R’ inherits from it.


  • The capital ‘K’, like its lowercase version, is built from a stem (‘I’) and the letter ‘X’. The lowercase ‘k’ itself can also be helpful for modeling the tail. ‘K’ is about the same width as the ‘X’, sometimes slightly thinner.
  • Inner serifs, of course, are shortened, and the serif on the upper arm expanded, just like with the lowercase ‘k’.
  • Often times the arm is extended back past the vertex, so that it connects to the stem (covered in the lowercase ‘k’ tutorial). Here, I made a disjointed ‘K’, where the arm and leg meet at a point.
  • If you make a disjointed ‘K’, an optical correction may be in order to prevent the vertex from looking bent down. This involves adding a slight upturn to the vertex. Shown below is an uncorrected ‘K’ (left) and a corrected ‘K’ (right).

Monday, October 6, 2014

Design notes: miscellaneous capital letters ( U B S )

At this point, I’ve covered almost all of the “hard” letter forms, and by now, you should be familiar with many of the subtle and mundane rules of type design. So it makes more sense now to just release a few design notes on each of the remaining letters pointing out some oft-overlooked features of their letterforms.


  • This letter is about the same width as the ‘N’ and the ‘V’.
  • Like the ‘M’ and ‘N’, the ‘U’ defies the typical calligraphic stress rules. While some typefaces such as Cambria follow the rules strictly, making ‘U’s with two thick stems, it is overwhelming convention to make the upstroke a hairline, just like in the ‘N’. In fact, the stems are often exactly identical.
  • While this varies somewhat, usually the bowl of the ‘U’ is a hairline as well, with stress only appearing left of the letter’s minimum. In other words, as you trace the letter from left to right, the stress disappears at the minimum, just as you’d expect calligraphically, but when the curve comes back up to finish out the letter, the stress never reappears.
  • The left stroke is usually identical to the top half of the ‘I’ glyph.
  • Curve onset usually occurs about a two-thirds the way down the letter. The bowl usually occupies a rectangle with sides in a 2 : 1 ratio.

Sunday, October 5, 2014

Glyph design: zigzag capitals ( M N )

There are two letters in the capital alphabet composed of diagonal strokes contained between vertical stems—‘M’, and ‘N’. Because of the letters’ architecture, they violate the typical rules of stress. Strictly, the ‘M’ should go thick–thick–thin–thick (from left to right), and the ‘N’ should consist entirely of stressed strokes. But just like with the ‘z’ (which, strictly, would be entirely hairline), some strokes in the ‘M’ and ‘N’ are arbitrarily made thin so that the stress alternates with each stroke. Owing to the pointed-nib calligraphy pen, which stresses in part based on the direction the stroke is drawn in—up being thin and down being thick—the usual convention is to make the ascending leg of the ‘M’ and both stems of the ‘N’ into hairlines.
Sabon and Minion Pro have slanted stems, while Proforma has vertical stems on its ‘M’.
Sabon and Minion Pro have slanted stems, while Proforma has vertical stems on its ‘M’.
 The ‘M’ can be made either from the ‘W’, or from the ‘I’ and ‘V’. It depends on the slant of the outer stems. Some typefaces, like Sabon and Minion Pro (and a few sans serifs, like Futura) have slanted stems, others like Proforma have vertical stems. Floribunda will have an unslanted ‘M’, so this tutorial will use the ‘I’ and ‘V’ method. The two ‘I’ glyphs should overlap with the ‘V’ just enough so that their intersection points lie at roughly the same height as the lower edge of the ‘I’ brackets. The left ‘I’ should also be thinned to the hairline thickness.

Friday, October 3, 2014

Glyph design: letters with bowls, part one ( D P )

The ‘D’ and ‘P’ are perhaps the most difficult letters of the uppercase alphabet to nail. They require you to blend straight letter forms, like the arms of the ‘E’,  into a round bowl, like the one on the ‘C’.

The ‘D’ can be made by taking the left side of the ‘E’, and attaching it to the bowl of a ‘C’, rotated 180 degrees (never reflected horizontally!).