Last week, 2 years after Microsoft announced its sweeping plans for .NET, Microsoft Chairman and Chief Software Architect Bill Gates hosted an event at the Microsoft campus to give analysts an overview of the company's progress thus far. The verdict isn't pretty: Like a principal cracking a yardstick on a recalcitrant student's knuckles, Gates delivered a grade of C to the company for its delivery of Web services. That grade might be a bit generous. Here's my own analysis of where Microsoft stands regarding .NET so far.
Envisioned originally as a universal online identity authentication service, .NET Passport gets no respect today because the company forces the service on the hundreds of millions of people who use MSN Hotmail. In fact, if .NET Passport weren't tied to Hotmail, I find it unlikely that more than a few dozen people would even take advantage of it. .NET Passport's eWallet feature goes virtually unused, and a short-lived .NET Passport for Kids promotion failed to garner serious interest from parents. .NET Passport is a prime example of a solution without a problem, and if Microsoft is serious about .NET, I suspect the company will scale .NET Passport back in the coming years and ultimately let a more universal solution, such as that offered by The Liberty Alliance, to displace the service. (For more information about The Liberty Alliance, go to
Microsoft purchased Hotmail in January 1998—when the free email service had only 9 million users—for about $400 million. Today, with almost 250 million active users, Hotmail is the most popular email service in the world. Although Microsoft is slowly ripping features out of the free version to influence customers to sign up for an extra-cost subscription service, the free service still works, and the per-fee Hotmail service is still cheap. Like .NET Passport, Hotmail predates .NET, but Hotmail's success is probably the most visible sign that Microsoft can deliver Web-based services to a large audience.
On a darker note, Hotmail is marred by spam. Microsoft recently admitted that more than 80 percent of the mail transmitted to and from Hotmail is spam that the company's spam filters don't catch. That's more than 1 billion pieces of spam, every single day. To give you an idea of how bad this situation is, I maintain a Hotmail account that I never use: Over the past 5 days, I received 273 spam messages, so I cleared out the Junk Mail folder last night. When I woke up this morning, the account had received 34 new spam messages. Unbelievable! I hope Microsoft will do something about the spam problems in Hotmail, even if doing so requires a paid service.
.NET My Services
Last year, Microsoft began touting .NET My Services, which was to have been released by late 2002. According to the company's plan, .NET My Services would build from .NET Passport to supply a suite of Web services to enterprises and consumers providing, among other things, online authentication and identity management, and calendaring and email services. However, .NET My Services had one fatal flaw: It required Microsoft to store user data on its servers, raising serious privacy and security concerns. For large corporations, an even bigger problem arose: No company was interested in allowing Microsoft to store proprietary customer information.
When not one of its partners endorsed .NET My Services, Microsoft canceled plans for and retooled the service. The current plan is to ship a server-based product sometime in 2003 that will enable corporations to roll out services similar to .NET My Services but using a company's internal servers. Thus, Microsoft's single biggest Web services plan has been completely recast into yet another standard server product, raising serious doubts about whether the company can ever truly deliver on its .NET vision. Say what you will about .NET My Services, the project was a major misstep for the software giant.
Visual Studio .NET/.NET Framework
It took years to complete, but the move from Visual Studio (VS) 6 to Visual Studio .NET—which enables developers to create .NET services and applications—was well worth the wait. Visual Studio .NET is an amazingly mature development environment with support for all the latest .NET technologies, including the .NET Framework, C#, and ASP .NET. Developers have quickly embraced this tool and are now racing to deliver .NET-based software. In a field of otherwise disappointing products, Visual Studio .NET is the shining star, and the .NET Framework it supports is logical, well-designed, and full-featured.
.NET Enterprise Servers
Nothing is wrong with the .NET Enterprise Servers—among them Application Center 2000, BizTalk Server 2002, Commerce Server 2002, Exchange 2000 Server, SQL Server 2000—other than the simple fact that most have very little to do with .NET. Perhaps the company was riding a .NET high when it decided to preemptively name its server line the .NET Enterprise Servers, but I think Microsoft should have waited until these products had integrated .NET Framework support and performed more than a modicum of XML Web services interoperability. Today, the .NET Enterprise Servers just confuse the .NET issue for customers, who might logically wonder whether this group of products was what they had been waiting for all along.
Articulating the .NET Vision
I've always thought that Microsoft has done a supremely poor job of communicating what .NET is and why customers should want it, and the events of the past year have only reinforced this belief. "One degree of separation": Can anyone on this planet explain what that phrase means without having to think about it? In contrast, I think Microsoft is taking the right approach with Palladium, its secure computing initiative, which won't be released until 2005. Unlike .NET, Palladium wasn't introduced to the world with a flashy event but rather was leaked slowly to the press, letting people simultaneously get a feel for what the technology is and also understand that the technology isn't going to emerge any time soon. With .NET, Microsoft has always given the impression that we're very close to seeing something palpable, when in fact the only .NET products the company has been able to release for end users—.NET Passport, Hotmail, and the .NET Enterprise Servers—all use pre-.NET technologies. This strategy seems like a smoke screen or a delaying tactic. Yeah, we get it: You're moving to Web services. But, years later, that day still hasn't arrived. Talk to us, Microsoft.
Overall, .NET isn't much further along now than it was 2 years ago: It's still a good idea, and Microsoft is still betting the company on it. But .NET hasn't altered the way we compute, and that's a problem. You can't change the world, you know, unless you actually change the world.