Glenn Snow wrote on 08/03/15 at 14:20:14:
But in the end didn't you decide on 3...Nf6?
Indeed, though in a roundabout way, for many scattered reasons. At first, the Anti-Sicilians section of my book was intended to be very short, just providing some guidance on move order issues and giving a bit of short coverage (this was at Everyman's suggestion - I wanted to write a more comprehensive Kalashnikov book, and they wanted a repertoire book with a small amount of material on the Anti-Sicilians - they thought it was a more marketable book) of all of White's options, etc. And so in order to keep it manageable, I thought it might be easier to try and swerve into ...e6 related options, e.g. 1.e4 c5 2.Nc3 Nc6 3.g3 e6, 3.f4 e6, 2.Nf3 Nc6 3.Bb5 e6, etc. But then some irritating move order stuff popped up, e.g. 2.Nc3 Nc6 3.Nge2!? e5 4.g3, etc. And so I decided that it should be a more focused around fianchetto approaches. I had a lot of material already on these systems, as I had been playing my recommendations in the Closed and GPA for a long time already. This ...g6 stuff is also a little bit more versatile if people intended to use the book for just the Anti-Sicilian coverage or if they wanted to eventually move away from the Kalashnikov and to something else. Eventually I moved away from meeting the Rossolimo with 3...e6 because I found a few problems, it was no longer part of a thematic framework, and because I had a lot more experience and a few new ideas after 3...Nf6. I picked it up a long time ago from
The Easiest Sicilian and had always found White knew less there than in other places.
And as you can see, the Anti-Sicilians coverage also ballooned up into a book in and of itself - it felt weird for me to write 20-50 page chapters on the Kalashnikov and then really short chapters on the Anti-Sicilians. They'd be inconsistent and useless compared to the rest of the text, so I just did what I felt was right anyway!