Douglas Bowman, Visual Design Lead at Google, has decided to leave Google for an undisclosed opportunity. In a farewell note posted on his blog, he takes a swing at Google's data-centric, engineer-driven culture.The Google that Bowman describes is unlike the quirky, benevolent, intellectual playground that one might picture from reading employees' blog posts. The Google that Bowman saw was a sterile data mine, where minute design decisions demanded statistical support.

"I had a recent debate over whether a border should be 3, 4 or 5 pixels wide, and was asked to prove my case," he says in his blog. "I can't operate in an environment like that. I've grown tired of debating such miniscule design decisions. There are more exciting design problems in this world to tackle."

To emphasize his point, Bowman refers to a New York Times article about Google VP of search products and user experience Marissa Mayer. The story recounts how Mayer asked a design team to test 41 gradations of blue to determine the one favored by Google users.

Can you imagine that happening at Apple?

Google's overreliance on numbers appears to extend beyond design to hiring. A person claiming to be a technical writer at Google offers the following criticism at Glassdoor.com: "Google also tends to focus hiring assessment on academic scores, which is entirely appropriate for new grads, but they do this even for people with multiple years in the field. That's just silly, and it actually does run contrary to the whole performance process within the company itself, where effort and accomplishments are recognized. Thus, it makes no sense to disregard this type of assessment for new hires whose professional contribution can far outweigh their academic performance by many years."

While acknowledging the utility of using data to validate actions, Bowman argues that Google's engineering culture has paralyzed the company and made it incapable of daring design decisions.

There may be some truth to that. Anytime there's a dominant culture, groups in the minority face a challenge to be heard.

At the same time, data and design aren't necessarily antithetical. Consider how much better Facebook's home page redesign might have gone if those in charge had consulted more of the site's users.

Out of over a million Facebook users who have chosen to voice an opinion about the social site's new look in an online poll, 94% expressed disapproval.

So if Google's calculating design process is unlikely to lead to a home page to rival the artistry of the Sistine Chapel, neither is it likely to produce aesthetically refined products that sell poorly, like Apple's G4 Cube.

For a large company with millions of users worldwide, maybe that's the way it should be.