That didn't stop Thompson from advocating for Windows Phone's demise, positing that as a more viable long-term strategy than what Microsoft is now pursuing.
At first glance, the Nokia acquisition made sense, said Thompson in an interview Friday. "The acquisition of Nokia is certainly not a black-and-white situation, but on the surface if you are committed to a devices and services strategy, which Microsoft says it is, it does makes some sense, even if they had Nokia under their thumb earlier," he said, referring to the 2011 agreement under which Nokia bet everything on Windows Phone. "There are efficiencies that accrue [with owning the hardware], as Apple has demonstrated. But the problem with the acquisition is that I disagree with the underlying strategy that drove [the Nokia purchase]."
As Thompson saw it, Microsoft had a choice: It could either try to sell hardware — the "devices" part of the strategy — to a fraction of the potential market, or try to sell services to anyone and everyone with a connected device.
But Microsoft wanted all the cake. "Instead of choosing one or the other, Microsoft wants to do both," said Thompson.
And that's a bad move. Not simply because by doing all, Microsoft risks doing nothing well, but because its strengths clearly lie in the services side of the strategy.
"It's never been a hardware company," said Thompson. "I don't see any reason to expect that they now will become one. The danger is that the services that ought to be pushed, like Office 365, which could run on every platform, are going to die on the vine because of the emphasis on [Microsoft's own] devices."
Thompson was high on Office as a platform, in part because of its importance to the company's revenue — in the most recently reported quarter, the division responsible for Office accounted for 36% of the firm's total, the most of any group — but also because of Office 365, the software-by-subscription service Microsoft significantly expanded earlier this year.
"Over time, value moves up the stack," said Thompson. "First from the chip to the OS, then to the software, and on to the services. That's just the way technology goes. I think there's room for a premium service ecosystem built on their Office layer. Office is further up the stack than the OS, and they could leverage that to be the major player in the cloud."
In a post to Stratechery last week, Thompson was more specific in his advice. "They ought to pursue a strategy — services — that entails being everywhere," he wrote.
The problem is that while "services" is part of Microsoft's new slogan, the purchase of Nokia makes Thompson suspect that that half will not be the one calling the shots.
Sign up for CIO Asia eNewsletters.