In carlislerainey/ggcarly: Miscellaneous functions to combine with ggplot2

knitr::opts_chunk$set(echo = FALSE)

a theorem

Carlisle's Fundamental Theorem of Slide Design: Bigger is better until things start to overlap.

the theory

In general, details are hard to see in low-resolution formats like conference presentations. Imagine squinting, from the back of an unevenly lit room, at a slide projected by an an old, buzzing projector onto a tiny screen.

Where we have room to make things bigger, we should.

We usually have extra room along the axes to make axis and tick labels easy-to-read. There's a big ROI in having readable axis labels. This isn't important in the print format because it's easy (perhaps easier) to read small text.

There's not necessarily more room inside the plot for larger points, text, lines, etc.

Implication: For a low-resolution presentation format, it makes sense to use bigger, bolder axis labels than in a print format (holding the size of the text/lines/points in the plotting region constant).

classic theme; default size

library(ggplot2)
library(ggrepel)
library(ggcarly)
p <- ggplot(mtcars,
  aes(wt, mpg, label = rownames(mtcars), colour = factor(cyl))) +
  geom_point() + 
  geom_text_repel() + 
  labs(x = "Weight (in 1000s of lbs.)",
       y = "MPG",
       color = "Cylinders")
p

classic theme (manually resized in chunk options)

p 

carly_presents theme (manually resized in chunk options)

p + theme_carly_presents()

carlislerainey/ggcarly documentation built on May 7, 2019, 5 a.m.