Hugh Macleod still musing on, (talking himself into more business with) Microsoft.

I'm a) answering both one ugly bit of sophistry that came up in the comments :

2. The US Government's paradox is that, while monopolies might be illegal, for Microsoft, as a publicly traded company to act in any way other than it has would also be deemed illegal.

This is clearly wrong. No company has a legal obligation to its shareholders to do illegal things. And even if they did have a legal obligation to do *wrong* things, anyone with any integrity would avoid them like the plague.

If you stop to consider the same argument being applied to, say, bribing foreign governments or abusing human rights, that should become obvious.

And then b) addressing the wider question.

I can understand that there's a genuinely interesting challenge to try to make Microsoft relevant and exciting again.

But I don't, honestly see how that can happen *yet* - they haven't been nearly humbled enough and still in transition between Gates / Ballmer and whatever is coming next.

Yes, they need a new idea. But chasing any new technological trend can't be it.

Microsoft's big idea, which has sustained them for 30 years, and really was visionary when Bill Gates was promoting it in the 70s, was that, with the right intellectual property laws, you could build a "pure software" company, selling directly to the end-users rather than be part of, or a supplier to, a hardware company. And as a pure software company you could get your product onto everyone else's hardware in every office, in every home.

I see Microsoft as the best-case scenario for a proprietory software company. But that turns out not to be good enough. We need more and more powerful software on a more complex ecosystem than anyone, even Microsoft, can keep under control. And the only way we can have that is through open platforms and protocols, free (open-source) software, "peer-production" and software-as-a-service. All of these are deeply inimical to Microsoft's core DNA of wanting to "own" software platforms.

If MS is to have a future, its "next big idea" can't be one type of device or another. (There's going to be a multiplicity of different types of devices. More than Microsoft can produce or even write drivers for.) It can't be "advertising" because advertising itself is under huge transformative pressures.

No, the big idea has to come after MS have exorcised the notion of "software as product" and started with a clean slate.