Wednesday, February 29, 2012

Microsoft and the epitomy of fail

Ever wonder why Microsoft lost the Internet? They did. Nobody uses Microsoft products for Internet services except a few losers that don't know any better. The cloud? It's a Linux world, baybee.

Well, here's the deal that I've found out this past six weeks or so: Developing for Microsoft products is a painful and expensive process. What that means is that Microsoft has lost the student and hobbyist markets, and that's where the people who create new stuff come from -- they don't come from stuffy old companies that have $10K hanging around for Microsoft licenses for a single engineering team.

So anyhow, I recently created my first tray app. It used a couple of MFC functions where I could not find an equivalent anywhere else in the system that would do what I needed done. Because of that, it would not compile with Microsoft's "student and hobbyist" product, Microsoft Visual C++ Express. Once I did a workaround for that (temporary install of the demo version of the full Microsoft Visual Studio just to get access to the MFC include files and library), then there was another issue: Visual C++ Express won't create a .msi installer file for distributing your program. I worked around this by using an open source program called WiX to build an XML template for my package, and successfully managed to create a .msi file. I installed it. I used the control panel and uninstalled it. Yay.

But here's the problem. I'm a hobbyist when it comes to Microsoft software. As a hobbyist I only write Open Source software (if you want closed source software out of me I'm happy doing it, just pay me $$$, but that's not my thing with hobby software). As a hobbyist I don't release software that can't be compiled unless you pay Microsoft hundreds of dollars for the full Microsoft Visual Studio, or do possibly-illegal workarounds like installing demo versions on top of crippleware. So you'll never see this tray app and it's probably the last time I write anything for Windows that I'm not paid to write.

Multiply that decision by tens of thousands of hobbyists who look at the same situation and instead go write software for Linux, and you understand why Microsoft lost the Internet. I did this because I needed to learn how to work with Microsoft's software to do some stuff at work, but most people simply aren't as driven as me when it comes to learning new things. They take the easy path... and that's Linux.

Too bad, Microsoft... your stuff actually isn't that bad, it's ugly and a bit incoherent but then so is Linux. But if you make it hard for people to get used to writing software for your platform on a hobby basis, you lose. It's that simple. Microsoft Visual Studio Express is useless -- you can't write real programs for Microsoft without MFC, and creating MSI files should Just Work rather than having to go search for a third-party tool. With Linux, it Just Works -- fire up Eclipse and start developing C++ or Java programs, that simple. With MacOS, it Just Works -- fire up XCode and start developing Mac programs in Objective C or other supported languages, that simple. Windows? It's a painful experience trying to get around cripple-ware. And once that happens, there's one word to describe your company: Fail.

What's Microsoft's future? Microsoft really has no future other than as a legacy company. They lost the Internet and the Cloud because they preferred wringing money out of their devtools division (and it can't be a huge amount of money even) to fostering the next generation of innovators with free or inexpensive tools to use to write software for Microsoft's platforms, and they'll lose the next major innovation to happen in computing too. And one of these days, the accumulated sum total of these innovations will render Microsoft as irrelevant as Unisys -- just another legacy company milking its legacy products for service income long after they've become irrelevant to the majority of the industry.

-ELG

No comments:

Post a Comment