Now that is "real world" - problems even before release!-) Seriously, though, what is the RWH authors' plan for tackling the eternal frustration of Haskell book authors, a moving target? There used to be a time when one could guess the poster's Haskell book from their question topics: - 'HGL' doesn't work: Hudak's book - 'fromInt' issues: Thompson's book The problem arises because the books survive far longer than the version of the code base they refer to. The only way to reduce this friction has been to avoid fast-changing details, eg, focussing on programming, problem solving, reasoning (Hutton, Thompson, Bird), and even that isn't safe ('n+k', 'fromInt',..). Neither is keeping a separate library - 'SOEGraphics' has been revived several times, on different GUI/graphic libs, even different Haskell implementations, but has always fallen prey to bitrot again. I've only scanned RWH briefly, but it seems to be on the other end of the spectrum, focussing precisely on the details of stuff that is likely to evolve much faster than the book. Is the plan to keep someone employed over the years to keep the online version updated? Claus