He cllimbed the beach

If you type “He climbed the beech” into Google Translate, the French comes out correctly as “Il a grimpé le hêtre”. But I mistyped it just now as “He cllimbed the beech”. Out came “Il a grimpé sur la plage” — he climbed the beach!

It would take a better computer psychologist than I am to know exactly what happened here with Google’s machine learning algorithm, but the obvious guess is a rather cool one. I’m guessing “cllimbed” established me as a sloppy speller, from whom the most likely interpretation of “beech” was a sandy strip by the sea.

[29 August 2021]

Newton’s precarious PhD descendants

The Mathematics Genealogy Project allows you to track a mathematician’s students, their students, and so on. For example, I am listed with 75 descendants, and Isaac Newton has 21507.

But unlike me, Newton nearly died out! If the Project’s data are correct, he was followed by eight generations consisting each of one man having just one student, for 150 years (from Smith through Sedgwick, to be exact):

Isaac Newton, Cambridge 1668
Roger Cotes, Cambridge 1706
Robert Smith, Cambridge 1715
Walter Taylor, Cambridge 1723
Stephen Whisson, Cambridge 1742
Thomas Postlethwaite, Cambridge 1756
Thomas Jones, Cambridge 1782
Adam Sedgwick, Cambridge 1811
William Hopkins, Cambridge 1830

And then suddenly, we get the Cambrian explosion. As British science flourished in the Victorian era, William Hopkins had students Cayley, Galton, Maxwell, Routh, Stokes, Thomson, and Todhunter, all famous names still and all with hundreds or thousands of descendants of their own. It is to these that Newton owes his 20,000 or so descendants alive today.

If any link in the chain had been missing, Newton would have had no descendants in the present. Of course his actual impact would have been undiminished, so this is a reminder that there are more forms of inheritance than genetic.

[8 August 2021]