The problem with your reasoning is that Microsoft is also going hardware. And not by buying Nokia... Look at surface, it's only the start. So your main argument that it's not good to rely on the product of a competitor... is weak.
And it's not at all a reasonable strategy. A reasonable strategy is when you minimize the worst possible outcome of the strategy. This is maximizing it... while at the same time maximizing the best possible outcome. There is no way this is reasonable. It's a bet. A risky bet strategy. Might work... but computer history will remember it as a bold risk.
They are only serious about it if they push it to their channel and compete directly with the OEMs. They don't have enough stores to go it alone. If you read between the lines of the recent article in Vanity Fair, it seems Ballmer will not take it seriously and wants it to die by not being successful enough, because they never gave it a chance.
And it's not at all a reasonable strategy. A reasonable strategy is when you minimize the worst possible outcome of the strategy. This is maximizing it... while at the same time maximizing the best possible outcome. There is no way this is reasonable. It's a bet. A risky bet strategy. Might work... but computer history will remember it as a bold risk.
[Edited for typos]