Cloud Atlas 'astonishingly different' in US and UK editions, study finds

I remember studying "The Catcher in the Rye" back in high-school, and before we started to read we were provided with a few critical analysis pieces for orientation. One of these was a discussion of the difference between the UK and American editions the key of which was the omission of italics in the former. Now anybody who's read TCITR will know the importance of italics to the narrative. The UK editor didn't believe they were important, perhaps objected to them on aesthetic grounds and so they were stripped out. We did of course reading the re-released, American edition, printed as the author intended.

The reason why this happened is explained in the article itself; it's surprisingly common.

TL:DR; books are 'sold' by authors as intellectual property reproduction rights in distinct territories. Historically it used to be too expensive to ship lumps of paper across oceans, so the North American and UK/European English language rights were handled by local publishers in the USA and UK, and smart authors and literary agents unbundled them and sold them separately (rather than relying on a publishing house's sub-rights department to outsource the rights they can't make use of).

A smart author then goes and kicks butt so that their US editor talks to their UK editor (or vice versa), agree who's going to take lead on copy editing and typesetting, and then ensures that the author only has to double-check one version of the book as it goes through production. Note that publishers are usually happy to do this (except in KidLit/YA/educational material when tailoring to fit local national spelling/grammar rules is manadatory) because it's a lot less effort for one of them to just buy the InDesign files and slap a new cover on the product, and a lot less stress for the author (checking copy edits/page proofs eats multiple weeks during which the author would rather be writing the next book: you don't get paid for checking extra sets of edits).

Where this breaks down: author is inexperienced (as Mitchell was at the time), or one of the publishers fucks up (as also happened in this case), or there's a big dislocation in the schedule for the book and one publisher is left racing to catch up with another one to hit the same publication date (which they always try to do, in order to minimize the risk of retail imports eating their launch week sales -- the First Sale doctrine means you can't stop a customer from legally buying a copy in another country and paying for it to be shipped to them, and Amazon makes this really easy).

It's less of a problem these days as most of the major houses moved to their own brain-dead version of all-electronic production workflow in the past couple of years (mutter, grumble, Word change tracking and PDFs and email like it's 1995, baby). It took them so long because many authors are eccentric recluses who still insist on using fountain pens to make chicken tracks on paper, or WordStar 4.0 for DOS. (And, embarrassingly, you can't ignore them unless you want to kiss goodbye to GRRM's next installment in "Game of Thrones", for example: writing novels is one of the last truly artisanal art forms to still exist and have commercial potential.) Nevertheless, these cock-ups still happen even today: all it takes is editor A having a snit and refusing to talk to editor B, or having a wholly unscheduled stroke and being off-work for three months in a department that's already overstretched, and everything comes off the rails.

Source: this is my life.

"...I have a very low faff-tolerance threshold..."

Love that. I do some work as a fiction editor and I'm often amazed at how calm authors are about quite large changes occurring in their work. They can be obsessive about tiny things like a single word and then casually change the order of chapters, or cut whole sections without losing any sleep.