I find walking incredibly useful when programming - for those moment where a bit of lateral thinking is required, to think outside the box a little box, or step out of the function and look at things anew.
As the article states, it's not very good at helping you solve a particular logic problem though.
Oh I spend most of my time thinking in the box. I find that when I'm out walking I can see that box more clearly - I think that's part of the "out of the box" thinking.
I agree. I can recall a few relevant instances when I was working on my OS (for a class). I would get stuck on a problem (sometimes design related but also sometimes debugging), give up for the evening and head home. By the time I'd walked halfway across the football field, the answer would come to me. Not sure if it was the walking or the fresh air, but this happened at least twice throughout the 7-week project.
As the article states, it's not very good at helping you solve a particular logic problem though.