How to assess The Federalist Papers' authorship with statistics

The Federalist Papers comprises of 85 articles written in the 1780s by founding fathers Alexander Hamilton, James Madison and John Jay. They wrote under a collective pseudonym, Publius, so keep their involvement secret. But who wrote what? There is much dispute. Let's try K-Means Clustering.

K-means clustering aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean, serving as a prototype of the cluster. This results in a partitioning of the data space into Voronoi cells. For our example we’ll have 68 observations (papers) into 2 clusters (2 authors, Madison and Hamilton).

Once the data has been converted into workable features, we can fit them onto a 2-cluster model. This is unsupervised - we are effectively just pouring in our (ideally) significant data and telling it that there are two distinct sets within it, and to try and extricate them.

Spoiler: most of them were by Hamilton. Read the rest

Preventing typography disasters, from the Oscars to prescription bottles

This Vox video illustrates the fact that type design was the true culprit in last month's Oscars cockup, and how easy it would have been to prevent. But backslapping award shows are only the beginning of bad design when it comes to type.

The 2017 Oscars ended with a pretty shocking mix-up. Announcer Warren Beatty incorrectly named La La Land as the Best Picture winner, and the mistake wasn't revealed until crew members had already started giving their acceptance speeches. A lot of things went wrong for the snafu to happen the way it did. But what if typography was one of them? A better announcement card design could have made for a very different Academy Awards show — not to mention a much less embarrassing Miss Universe show for Steve Harvey back in 2015. But the implications of bad typography don't end there: poorly designed ballots in the 2000 presidential election arguably could have swayed the outcome, and illegible type on medicine bottles could be causing nearly 500,000 cases of drug misuse per year in the U.S.

The way pill bottles are turned into incomprehensible ads for the pharmacy--and how graphic designer Deborah Adler proposes to fix it--is intriguing, not least because there are always a lot of unseen pressures and constraints that design is bound by. Nothing will ever be done to fix it: informational text gets perverted by conceit, branding, regulation and corporate bikeshedding until it is worse than useless for any purpose.

Except one, perhaps! I'm especially fond of how manuals for electrical appliances are mostly regulatory safety warnings that succeed both in limiting liability and making the device less safe, because they are so forbidding and badly written no human will ever comprehend them. Read the rest