A short but sweet entry today, I’m afraid. Don’t worry, though; with a bit of luck, I should have something rather special for you next week. For now, though, on with the show!
First up is an article on Samuel Arbesman’s Social Dimension blog over at Wired, entitled “The Rarity of the Ampersand: Frequencies of Special Characters”.*
Arbesman is an applied mathematician, and he writes about an experiment carried out by Michael Dickens to determine the relative frequencies of characters in a variety of contexts such as prose, casual writing, and programming. Dickens confined himself to the component characters of the now decidedly old-school ASCII code, but the results are instructive nonetheless. There are some familiar faces at the bottom of the list: &, <, %, @, #, ^, `, and ~ bring up the rear for both punctuation characters only and all symbols combined. Perhaps the tilde, percent sign and caret should be in the running for Shady Characters 2!
Continuing with the theme of character frequency analysis, I recently came across a site called Context of Diacritics, whose creator Ondrej Jób describes it as “an analysis of diacritics made to help type designers with refining the character sets of their fonts.” Ondrej and his collaborators have mined Wikipedia articles written in a variety of different languages to derive a list of frequencies for characters with diacritic marks (ä, ř, đ, and so on) and pairs of letters with diacritics (çõ, ół, and so on). And though CoD is ostensibly a tool to help type designers focus on the most widely used combinations of characters and diacritics in their target languages, it’s a visual and typographic treat even for non-type designers like myself. Take a look! What’s your favourite diacritic/letter pair?