Unrelated to all the terrible things, even the most un-noticed fonts have fascinating and beautiful histories: https://aresluna.org/the-hardest-working-font-in-manhattan/.

I still have my dad's Keuffel & Esser set, plus a couple of loose sticks of circuit diagram symbols

I have to say, this is the most damning discussion of quantitative security risk management I've seen in ages.

It also pretty much tracks with everything I've seen, with the exception of a few orgs that have had both massive resources for security work and the discipline to only measure real things that exist and make a difference to technical security outcomes, like "unsandboxed line-of-code executions per day".

You can read Colette Alexander's thesis here: https://lup.lub.lu.se/student-papers/search/publication/9148570