The author wonders what important came out of, say, Farmville, and I think the answer is something along the lines of the importance of business intelligence -

The author wonders what important came out of, say, Farmville, and I think the answer is something along the lines of the importance of business intelligence - the attention paid to user telemetry, A/B testing, and product iteration has grown based on this has grown by an order of magnitude since Zynga started bragging that they could determine the ideal size of e.g. a call-to-action button. This makes it obvious there's an unstated assumption at work here, though, i.e. that the unimportant work plays a significant part in developing this technology. I don't think this is necessarily true: the opposite premise, that "it's GPUs time when it's GPUs time", seems equally plausible.