Frameworks
Message in a Bottle
Microsoft is looking for a way out of the Windows legacy code bear trap and if 'Midori' is any indication, it appears Redmond is working to do just that.
The dog days of summer are known for producing little news. But late last month, the software industry got a bit of a buzz from leaked documentation about a Microsoft project -- code-named "Midori" -- that, according to a report in SD Times, may scrap the compatibility-obsessed Windows codebase in favor of an entirely new operating system.
Speculation that Redmond was getting ready to jettison the 3.2 bazillion lines of legacy code in Windows is hardly new. The entire Windows Vista development cycle evolved into a years-long parade for clean-break propagandists, who believe Windows carries far too much baggage for its own good.
The Midori documents seem to strongly support what Redmond magazine columnist and frequent RDN contributor Mary Jo Foley has been reporting for months: That Microsoft is looking for a way out of the Windows legacy code bear trap. And if Midori is any indication, it appears Redmond is working to do just that.
But it's clear that Microsoft is looking for a lot more than a simple escape route. The list of key Midori technologies reads like a menu of breaking-edge technologies. Virtualization, managed code, distributed computing, parallel programming and concurrency, and lots and lots of abstraction. Midori seems to be as ambitious a project as Microsoft has ever conceived.
What is certain is that Midori will drive further down the path being paved right now by Windows 7. Despite the determined silence coming from Steven Sinofsky and his Windows and Windows Live Engineering Group, we know that Windows 7 will be a largely incremental refinement of Vista, with a focus on enabling Microsoft's Software plus Services (S+S) and Live services-based application models. Microsoft hopes to tightly wire up Windows 7 to the services-savvy cloud, creating plenty of leverage for its own S+S vision in the process.
Midori, significantly, hews to that same vision. It's a brilliant bit of maneuvering. Leverage Windows to promote and enable your services model, so you can then leverage your services model to promote Midori.
Beyond the artful tack and jibe of Microsoft's strategy, Midori raises the existential question of what will happen to our existing code and what developers should be considering as they write programs moving forward. Redmond is clearly plotting a years-long transition cycle, whereby Midori-enabled solutions live in harmony with existing Windows-based environments. But it's clear that development managers need to begin thinking today about the possibility of a post-Windows future tomorrow.
With so much still at question, how do you plan to prepare for Microsoft's bold vision? And what do you think Microsoft must do to succeed in its mission to not only break the compatibility chokehold of Windows, but enable a compelling, forward-leaning compute platform for the decades to come? E-mail me at
mdesmond@reddevnews.com
and let me know!
About the Author
Michael Desmond is an editor and writer for 1105 Media's Enterprise Computing Group.