Fiction:
- Jude The Obscure, Thomas Hardy. The way Hardy 'smuggles' Jude's viewpoint into the narrative is so ridiculously well crafted, a pleasure to read.
- Legend Of The Galactic Heroes - a big anime from the 80s which was based on Japanese soft SF novels, which are now finally being translated. Worth a read if you're into space operas and large scale politics, I'm happy that the age of ebooks allows for such 'niche' interests to get translated.
Non-fiction:
- Secondhand Time: The Last Of The Soviets, Svetlana Alexievitch. It's her 'usual' style, a selection of interviews with a 'chorus' of shorter interviews, this time about the fall of the Soviet Union. Lots of interesting stuff from people whose world was replaced with another world overnight.
- The Pursuit of Glory: Europe 1648-1815, Tim Blanning. This is actually an entry in the long row of Penguin History of Europe, but I haven't read the others. It examines certain aspects of change in Europe (starting with how transport networks moved from mud roads to proper streets and culminating in how the nation state was invented)
- Vanished Kingdoms: The History of Half-Forgotten Europe, Norman Davies. As the title says, a collection of essays on small kingdoms and countries which have only existed for a short time, some for which no current country claims 'ancestry'. It's always important to remind yourself what a random patchwork current European borders are.
- Eckermann's Conversations with Goethe. This is a collection of notes on young Eckermann's conversations with the aging Goethe. Goethe is famous for always challenging himself, always trying to create, make and be involved until his last day, it's absolutely inspirational. Before this book I didn't understand why he's such an icon for Germans (and I read a few biographies), but I get it now, he lived the 'man is his own marble' saying.
Computer stuff:
- Introduction to Machine Learning with Python, Mueller and Guido. This is essentially a more detailed version of the scikit-learn documentation with more elaborate exercises (even though some of it overlaps), highly readable. The scikit-learn documentation itself is among the best python package documentation out there and is something you can read 'cover to cover'.
- Write Great Code - I've only started to read this one, it's very much about the basics of what a computer does when you're running high level code, so you can adjust your coding style. So far it's very useful.
Stats:
- Discovering Statistics Using R, Field/Miles/Field. A very opinionated, highly amusing (the constant humor may annoy some), huge waltz through statistics and how to use the methods and interpret their output in R.