It’s been my experience that our intuition of what is and isn’t a pathological or malicious input is about as accurate as time estimates. Requirements change and any and all attack vectors will be found and exploited. The mental burden of remembering the assumptions made to correctly check for pathological or malicious inputs in all cases and keep them updated during refactoring is far too enormous to successfully maintain. — Almost Always Unsigned

Seeing Theory: A visual introduction to probability and statistics. I already mentioned this amazing website but it is still worth a retweet.

From Last.fm, this week

Haskell tutorials:

• Course on functional programming - Haskell (Clément Hurlin)
• Haskell training: 101 and 102 (Google)

Happy New Year Eve!

Anatomy of a Project Euler problem, and, some remarks on the OEIS. #math #python

macOS Setup after 15 Years of Linux. #apple

Why is my Rust build so slow?. #rust

10 things I love about Julia. #julia