VMware can afford to be the IBM of virtualization and charge substantial prices. But there are limits to how far that can go, as it found out with the charges based on vRAM use (the Vtax). Nor would I underestimate the capabilities of all three competitors -- Citrix, Microsoft and Red Hat -- to start at the low end, then move up, taking over more of the market as VMware struggles to produce the software-defined data center.

If it concedes the low end of the market, as it appears to be doing, I'm not sure VMware will be the type of company it's been so far. Software pricing strategies are notoriously tricky, and underselling competitors is also not the answer, not for a company like VMware. But it needs to keep some elements of that dual approach, the way Rosenblum and Greene did.

4. Give Virtual Desktop Infrastructure A Comprehensive Story

Virtualizing the desktop in one sense is easy. In another it is incredibly hard to do it right. With smartphones and tablets, the nature of end-user computing is changing so fast that it opens up new possibilities almost as fast as it opens up new exposures.

Desktop virtualization author Shawn Bass has written how virtual desktops are no more secure than physical ones and much work remains to be done to make them reliably defended. Citrix Systems has done yeoman's work in this area with the Air Force and Defense Intelligence Agency.

Bromium, a promising startup co-founded by Simon Crosby, former Citrix CTO, is doing additional work in the realm of micro-hypervisors, or microvisors, that isolate end users' risky tasks.

What's VMware's story on this all important subject? It's labored mightily to build out the features of its own virtual desktop infrastructure to the point where they're a match -- or a near match -- to the technical leader in the space, Citrix. But security remains a difficult conundrum. VMware is going to have to provide a leading example of how it solves security for multiple devices to get more lift in this space.

5. VMware May Supply A Magical, "Next Generation" Application

VMware has done two things concurrently in the application space. It's acquired end-user application vendors, such as SocialCast and SlideRocket, and it's built out a developer-attracting platform in Cloud Foundry, where developers may build their own applications.

So is VMware a supplier of infrastructure for next-generation applications -- a virtualization environment with developer platform support -- or is it a supplier of the applications themselves? Maybe both?

Is this issue what the Pivotal Initiative is meant to resolve? VMware is spinning out its Pivotal Labs acquisition, Spring Java framework, Cloud Foundry, the vFabric part of vSphere, and other elements into a new, 1,400-employee business unit. It may end up trying to be an application platform, good for building software to run in VMware's private cloud environment.

If VMware wants to directly supply applications, can it provide the combination of elements that end users can't resist? VMware would like to find a leading application that prompts customers to adopt its virtual desktop infrastructure, and then use apps as a way to capture the next generation of end-user computing.

But think of the competition on that front, the difficulty of displacing Microsoft Office and the speed of end-user computing's evolution. When the market is rapidly moving onto smartphones, tablets and other mobile devices, finding a magic application is not going to be easy. On the other hand, VMware has many elements in Spring and Cloud Foundry and the underlying architecture of SlideRocket that could be put into a developer platform, with virtualization allowing it to display on different types of devices. If VMware can find a way to enable secure application output in that vein, it might have something.