As late entrants into most of these platforms, their strategy is to leverage their existing technical capital as much as possible. So, we're seeing cloud versions of Outlook, Office, and now Visual Studio. I don't think they're trying to move people away from desktop (yet), but they are making their company's fate less tightly bound to desktop.
I've been working as a developer in Redmond in "Server & Tools", now called "Server & Cloud", for the better part of a year now. I know for a fact there are teams that work to address piracy, but I personally haven't seen it as a driving force in specific goals.
Without wanting to give away key strategy (not that I even know much) Look at how the company is structured and where the revenue comes from. Our division serves primarily enterprise customers. Enterprise customers are totally loving the cloud and are the primary driver of this demand. So we're focused on delivering a first-rate cloud platform (Azure) and first-rate developer tools (e.g. VS 2013 and cross platform Azure libs) to develop for it.
It seems to be working, our deployments are growing steadily. Public reports claim that we are deploying semicustom data center capacity in markets that impose some localized requirements.
Look at the investment in infrastructure capacity being made here. Such a fundamental shift wouldn't make sense (to me) for the goal of reducing the number of over-deployed licenses, when that could probably instead be addressed with far cheaper (even if annoying) DRM-like solutions.
So again, our customers told us they want cloud, so we're going to build them the best darn cloud ever, not forgetting our experience with their particular needs and requirements.
I wonder if there's something to be said for getting web-based dev environments working well enough in a private cloud setting....I could see that as a future. And then the bonus of not even caring which device you access it from.
Modern virtualization capabilities somehow collide with the webapp approach and i wonder where this will go.
I wonder if Microsoft sees the fleeting profitability in owning the desktop environment (more competition, shrinking market, higher expectations). So they're trying to apply the same concepts that really helped propel windows as a popular platform (easy to develop for) to their cloud platform (azure) in an attempt to coax their existing user base to start adopting it over other cloud platforms (like google apps)
As a windows developer myself, I see this as a great strategy. I've been questioning staying with C# and windows for a few years now... it started with buying a Mac (while running Parallels) and learning several new languages. But I still LIKE C#. So it might be a great transition plan.