On an HHKB

Welcome back to our series on designing custom keymaps! After looking into how good/bad QWERTY is, the power of layers, and the potential of custom keymaps, last time we took the first real step by examining your options for compiling a corpus. As a recap: The corpus is simply a big chunk of text. We use this collection of textual data, often a single text file, to characterize your typing habits (calculating various language statistics), and feed it directly or indirectly as an input into layout optimization algorithms – to find the optimal keymap for you! Today we’ll expand on this idea by exploring your options if, like me, you prefer a personalized corpus rather than grabbing some general (and mostly irrelevant) data available online. Image 1: Letter frequencies – the most basic use of corpora – in this very post We've seen that a well-crafted, personalized corpus plays a key role in determining the outcome of the optimization process. We've also explored...