Fun article in The Verge about deciphering Indus Script using statistical methods. I used to play with this stuff with the Voynich manuscript. But that text is sort of a honeypot, a largely irrelevant medieval European one-off, and quite possibly a hoax. Indus Script matters, there’s thousands of unreadable objects with Indus writing on it. The Harappan civilization is important to human history and so little is known about it.
Mostly I just wanted to make a note that Indus Script was proposed for Unicode in 1999. About 250 code points. It’s still pending.