APP-V, MED-V, and missed opportunities

Holding back on its legacy emulation play is a mistake for Microsoft

As those who follow this blog will attest, I've made no bones about my dissatisfaction with Microsoft's enterprise desktop strategy. The bloated heap of incompatibilities known as Windows Vista has forever soured me on the company's post-XP direction, and no amount of UI smoke and mirrors will make me warm up to its imminent successor, Windows 7. As far as I'm concerned, the position of desktop standard bearer -- once claimed almost exclusively by Windows XP -- remains vacant.

Part of the problem is the company's refusal to address the legacy compatibility issue. Industry pundits and enterprise customers alike have been clamoring for a long-term alternative to the ever-expanding collection of patches, shims, and compatibility band-aids that make up the current Windows runtime. But instead of the clean break that so many wanted, Microsoft is sticking us with yet another round of duct tape and twine in the form of Windows 7.

[ See how Windows 7 beta performs in InfoWorld's benchmark tests. ]

It didn't have to be this way. Thanks to some prudent acquisitions -- including Softricity and Kidaro -- Microsoft now has the technology pieces in place to create a viable compatibility layer that would allow legacy applications to run unmodified on an otherwise legacy-intolerant, managed-code-only OS platform.

But instead of weaving these bits into the Windows core, where they could benefit all users, the company is withholding the productized versions -- Microsoft Enterprise Desktop Virtualization (MED-V) and Microsoft Application Virtualization (APP-V) -- and reserving them as carrots to entice enterprises to sign up for its Software Assurance licensing program (as part of the Microsoft Desktop Optimization Pack, or MDOP).

When Microsoft first acquired Softricity, I had high hopes that the company would soon kick the legacy habit. However, after two years, the product formerly known as Softgrid remains a bit of a bastard child, caught in a state of perpetual limbo between the Windows desktop, server, and applications groups.

The former Kidaro Workspace product has been similarly relegated to afterthought status. In fact, aside from some rebranding and a bit of management interface tweaking, neither product has evolved much since arriving at the Microsoft campus. Certainly, they haven't been integrated with the Windows desktop stack in the way that IT shops were hoping. I mean, you can't even deploy APP-V encoded packages to PCs running the 64-bit versions of XP, Vista, or Windows 7, and there's no relief in sight before 2010.

I, for one, consider this to be a missed opportunity for Microsoft. The company is already on the defensive after the Vista debacle. The inclusion of a robust legacy emulation layer, consisting of an integrated APP-V sequencer (for capturing legacy application installers) and a fallback MED-V workspace (for handling the really tough-to-virtualize applications), could have been just compelling story they needed to woo all those XP Refuseniks.

Instead, we're asked to swallow yet another round of broken applications and consumer-oriented fluff. So what if Windows 7 is prettier and (maybe) a bit faster (than Vista)? If they really wanted to win over the IT crowd, all they had to do was make the clean break we've been asking for. That, and roll in APP-V/MED-V to help us isolate those legacy programs we'd all like to kill off but that nobody can seem to live without.

Copyright © 2009 IDG Communications, Inc.

How to choose a low-code development platform