Microsoft's New Win-Win Strategy

Over the next week or two, I'll be blogging about Microsoft's product strategy.  I have received a variety of interesting reactions to Windows Vista: Past its Due Date.  I obviously struck a chord with many people both inside and outside of Microsoft.  In the "Vista" article, I related a tale of another time a major product met its demise: dBASE.  I was there for that one.  But, history repeats itself, and it's happening with Windows.  Windows is headed down the same path.

Unless Microsoft does something dramatic.  I call it Windows TNG.

Microsoft's best, and perhaps only opportunity to take their products to the next level involves four simple, bet-the-company steps.  Each follow-up post will go into these steps in detail.  Honestly, I don't think Microsoft has what it takes to take such bold moves anymore. But, here they are:

1. Scrap the Windows codebase forever.  Release Vista, and announce publicly that it will be the last version of Windows based upon the NT/Win32 platform.
2. Use Linux as the base operating system for the next generation of Windows.  Do not modify it, do not "Microsoftize" it.  Do not try to own it.  Exploit it.
3. Reinvent the Desktop.  Call it Windows.  Windows: The Next Generation.  Outdo Apple, outdo the current platform, outdo every "Linux desktop" effort in existence.
4. Put applications first.  Office TNG, Project TNG, Excel TNG, Outlook TNG.  Do not port.  Rewrite.  Do not create a Win32 compatibility layer.  Do it right.

To some readers, this is an obvious win.  To others, it's ridiculous.  Some would say it is heresy.  The most analytical would say that it throws away Microsoft's biggest IP asset, the Windows codebase, and puts Microsoft head-to-head on a level playing field, making them far too vulnerable.  The stock price would plummet.

Maybe so.  But, once you realize that the codebase is the problem, you also realize Microsoft has to devise something better.

My early career was fueled by the exitement that Microsoft brought to the playing field.  They not only created a C compiler for me, they created a better one.  They not only gave me a graphical desktop.  They gave me a better one.   They not only created a better way for the novice to write windows applications, they amazed the world with VB and revolutionized the desktop development environment.  They were hungry, unfettered by legacy strategies and technologies.  They were the underdog, and they were my champion.

So, rather than bitch and moan (which is so easy), I'll go into details about these steps in a multi-part post.  I hope it's useful.

Comments

[zzz]

Why Linux and not Singularity? There's next to no benefit in writing native code when you take into account everything and not just the performance. I think a new kernel should be one that'll give the best environment for running managed code which is what majority of developers are writing if not today, by the time there will be an entirely new OS from Microsoft available.

So why is Linux is the better base for managed apps rather than Singularity?

http://www.google.com/search?hl=en&q=singularity
Look the fifth link.

[mysterious.e]

This is really a pointless as it's not going to happen.

But ... you would have better luck orchestrating a cabal in the Linux comunity to take on Microsoft and Apple. That could be useful and profitable. It will take more than the LAMP stack and Google to lead the way.

1) Utiltize the soon to be open source java as THE app dev environment. Eclipse! Ahhhhhhhhhhhhh!
2) Build an integrated storage engine around MySql. (Check out the 5.1 xml features - they are cool!)
3) Take the 'universal canvas' approach to the office suite. Scrap open office. Think simple. And it ships with the OS of course. In fact, the browser is just a universal canvas app.
4) Buid a windowing environment that is both approachable and revolutionary. Yes - revolutionary.
5) Build out a compatability layer that allows allows win32 apps, cocca and carbon apps to run. Make it plugable. Find a way to map all the protcols to WS*. Quad core is on the horizon. 64-bit is here. Process technology will allow gigabites of ram to be dirt cheap in 18-36 months. This is not impossible. After all, if they can bring their luggage their is a much better chance they'll stay.
6) All apps are JIT compiled. They just work. Nobody compiles anything.
7) No one will settle for less so don't even dare bank on the contrary.

Compete against Apple and Microsoft and out think them. Take Linux from the realm of the command-line wonk and make it workable for the masses. Realize the vision of Java as an application patform. Make MySql the keeper of all bits. If you executed spot-on it could take 4-5 years but it would be worth it. But most importantly you have to be much better and simpler. There has to be a compelling reason for the consumer that to embrace this OS will make their lives and their wallet so much easier that they will leave everyone else in droves. It just cannot be similar. You have to best them from all angles.

It's time for the Linux community to put up or shut up. Stop preaching and solve the problems. Come out of the dark caves where the servers and shadows lie and into the realm of the mortals that just know how the get to the internet by clicking on the blue 'e' icon, or think that their Mac is truly invincible. Linux is not better becasue it is an altruistic endevor or becasue of truly great server apps such as mysql. This kind of thinking keeps Linux as the reciepent of data from web forms and as a cheap replacement for cray super computers. To be great it has to be everywhere and out of the way at the same time. It should be like the lightbulb or the refrigerator in ease of use an ubiquity. Microsoft nor Apple can claim that.

[Gary Wisniewski]

"Why Linux Instead of _________".

My argument for Linux is strategic rather than technical, though I think Linux has technical merits to be sure.

Microsoft has proven time and time again that it can enter markets and dominate by assimilation, adoption and control. Suggesting that Microsoft can create an operating system better than Windows is putting it on the same path it's on now. They might as well try to fix Windows.

Linux, on the other hand, is ripe for Microsoft's classic "embrace and extend" culture. Microsoft is much better at assimilation than innovation.

My subsequent posts will elaborate on this. Stay tuned.

[Ed Kaim]

I've heard ideas like this a lot in the past, so I posted some similar thoughts I've had on the other side of the subject: http://www.sharplogic.com/blogs/ed/PermaLink,guid,f298e9a0-0c11-4faa-9d55-52d54c4d9a84.aspx.

[AndyC]

Right...

And who exactly is going to rewrite (not port) the literally billions of Win32 apps that businesses rely upon daily? Nobody, because this is clearly a ridiculous Linux lovers pipe-dream...

[Gary Wisniewski]

"who exactly is going to rewrite (not port) the literally billions..."

I didn't say ALL developers should rewrite everything, just that Microsoft should not port or rely upon a compatibility layer. Maybe a lot of code in MS office is already portable. If so, it would be easier. My main argument is not to take the lazy path, but rather to innovate. They'll have exactly the same problems with "The Singularity". The have code written for "the old operating system" and need to move it "to the new one". If it turns out that a compatibility layer really is the best for the user, then that's a reasonable option. Providing a compatible platform does not dictate using the same codebase. It's just the easiest way, up to a point. And we've reached that point.

Second, I'm surely not a "Linux lover". As I said, most all products I've developed commercially have been Windows-based. I am very familiar with how it works, and also how Linux and Unix work. I've been using them all for a long, long time. I've done development under RSX-11M, RDOS, AOS/VS, VMS, OS/2, SunOS, Solaris, and others. I don't love or hate any of them. They are just operating systems.

It's true that I do feel Linux has advantages to Microsoft as a operating system platform, but my ideas are based upon logic, not some irrational "belief" in one or the other. Instead of Linux, it could be a completely new Unix-like kernel, but Linux has political and competitive advantages to Microsoft.

Third, is it a pipe-dream? Well, I'm the first person to say "probably". But, sometimes when you have something in your head, you need to get it out. I wouldn't be writing this if I didn't think it had merit, despite the unliklihood of such an outcome.