Greetings SIGCIS,   My colleague Brian White has done quite a bit of work to successfully recover a demo/game of Nim, written to run on the Whirlwind computer in late 1955.  We know that the Whirlwind project had an ongoing work item to produce a variety of demonstrations to impress their steady stream of VIP visitors with the breadth of applications that could be run on this new-fangled general-purpose computing machine.   The thing that's a puzzle is just how much work they put into "generalizing" Nim.  The normal game allows a number of tokens to be removed from a bucket on each turn.  The WW Nim can be configured to allow/require the players (human against computer) to remove a number of tokens from a number of buckets in each turn.   I don't think the Nim game is very well known these days, but it seems to have had a bit of a moment in the early days of computing as an algorithmic problem that was more challenging than (e.g.) tic-tac-toe, but more tractable than chess.   I've seen reference to Nim on a couple of special-purpose Nim-playing machines (Nimatron and Ferranti NimRod), but is anyone aware of other "game theory" approaches to Nim algorithms from that era that might have prompted the Whirlwind efforts?   Thanks! /guy ps,   All the material is posted on my Whirlwind simulator github site, although the user-interface is so un-intuitive (there's graphical output on a CRT, but input is all in octal) that it really needs a fairly extensive user-guide to be playable.  You can see the original Whirlwind memo on the game at http://www.bitsavers.org/pdf/mit/whirlwind/DCL-series/DCL-113_Generalized_Ni...