Little experiments like this one by Errol Morris point to a macro shift in the visual arts from subjectively assessing design elements to objectively measuring design elements. It will take years to fully manifest itself, and will not go down well in much of the creative community, but it’s inevitable. When it is simple and fast to test font A versus font B, or concept A versus concept B, clients will demand it, and designers will have to accept the results.
Michael Beirut explains the experiment:
But I also know that the ingredients used by graphic designers — colors, shapes, typefaces — are fundamentally mysterious. What do they mean? How do they work? Why does one work better than another? What criteria should we use to choose?…This ambiguity can be maddening, especially to clients, who in desperation will invoke anecdotes and folk wisdom to help control an otherwise rudderless process…
To Morris’s surprise, the results of the test showed a clear difference between the performance of Baskerville and other fonts — not just Baskerville and Comic Sans (no contest); or Baskerville and Trebuchet or Helvetica (a clear serif versus sans distinction); but even Baskerville and Georgia (a lovely, and arguably even more legible serif by Matthew Carter). Compared to versions in the other typefaces, the passage set in Baskerville had both the highest rate of agreement and the lowest rate of disagreement. This led Morris to the inevitable conclusion: Baskerville is the typeface of truth.
First seen on Kottke.