Microsoft's outgoing CEO couldn't see the desktop for the tiles.
It should come as no surprise that Steve Ballmer is out as Microsoft CEO. The company's vision for client computing is miserably off base, and that's on him. How could he have blessed such a misguided strategy for client devices?
On Ballmer's watch, Microsoft rolled out Windows 8, a product that is destined to go down as one of the most colossal missteps in computing history. I wouldn't be surprised to see it morph one day into a verb for undermining your own product. Like, for example, "the way they eight their core product like that, it's no wonder they went belly up."
Any first-year b-school student would tell you that Microsoft disregarded the basic tenets of business expansion with Windows 8. If you want to grow your business, you've got two logical options: Extend the reach of your existing products into new markets, or develop new products for your existing customers.
There's a simple two-by-two matrix to guide such decisions. You can make one for yourself by drawing a cross inside a square so that there are four quadrants. Label the two columns "new customers" and "existing customers" and the two rows "new products" and "existing products." Expand your business from the "existing products/existing customers" square to either extend current or develop new products. The square at the far end of the two-by-two matrix – "new products/new customers" – is better left for the start-ups.
Maybe Keanu Reeves' character Neo might see how Windows 8 fits into the matrix. But here in this dimension, there's no option for taking an existing product with a large stable of existing customers -- the cushiest square on the board -- and essentially upending it by transforming it into a new product.
A common ploy in the software world is to get existing customers to pony up again for existing products by making and selling new-and-improved versions. That's what Microsoft had done through Windows 7. And it's what the industry expected the company to do with Windows 8.
This is a whole lot easier to grasp if we set aside the Windows numbering scheme for a moment. The existing product is the Desktop, and the new product is the Start Page. Shortly after the iPad came out in 2010, whispers were circulating that Microsoft was developing touch for Windows 8. Many industry participants were excited. They took that to mean that the software giant would be adding touch capability to the desktop.
They expected the company to update the menu structure so that it was more finger-friendly. And they expected to see gesture equivalents for mouse and keyboard commands. And they hoped to see some APIs so that hardware vendors and app developers could run with it.
No such luck, though. Instead of upgrading the desktop, Microsoft spent most of its efforts on the Modern UI and relegated our beloved desktop environment to a tile on the Start page.
So why did Ballmer embrace such an ill-fated direction? It's certainly possible that the Windows 8 nomenclature fogged his vision, and he just assumed that we'd all automatically upgrade from Windows 7.
More likely, he got caught up in the one Microsoft vision. It must have been intoxicating to think of a wide swath of products, all with a universal UI and programming model. But it shouldn't have been so intoxicating that he couldn't see that it was a really bad idea to upend the PC -- the market where Microsoft actually has a presence -- so that it would play nice with products in markets in which Microsoft barely registers.
Maybe if Ballmer directed the company to approach the problem from the other side of the fence -- force Windows phones and tablets to adapt to the PC desktop -- everyone might be better off, huh?
As it happens, it's too late for Ballmer. Sadly, he was so close to the product that he couldn't see the desktop for the tiles.
Join us for a roundup of the top stories on InformationWeek.com for the week of April 24, 2016. We'll be talking with the InformationWeek.com editors and correspondents who brought you the top stories of the week!