Microsoft's big picture: Whistler, Blackcomb and .NET
We take a look at Bill's big .BET...
Special report Microsoft's .NET initiative can generally be seen as a view of the future where applications are broken down into online services that can be utilized over the Internet. These application services will be rented or leased. While packaged software isn't expected to disappear entirely, many believe that this new model will radically impact on how businesses and consumers obtain up-to-date products from companies like Microsoft. Individual and corporate users will be able to dial-up applications on demand through either a subscription agreement or rental arrangement.
.NET will offer a new approach to user interfaces for Windows and the company's other products, the intention being to facilitate easier interaction with Web-based services. Compare this with the Windows 95 Explorer; Explorer was designed to make it easier for people to work in a document-centric world so that the previous focus - on applications - would be pushed to the back. This will be a lot more obvious in .NET; users will work more seamlessly with the Internet so that the boundaries between the local system and the "Internet cloud" are even less defined. The .NET user experience is ptiched as a logical extension of Explorer, and will rely heavily on XML technology, which will enable easier customization and integration with the Internet.
The actual .NET Framework consists of three primary layers. The universal run-time engine handles the lowest level of services, which includes thread management. This engine is conceptually similar to the Java virtual machine, which manages environmental matters. A common class library rests on the universal run-time engine. The code objects can communicate with applications written in any programming language, thereby making it possible to utilize large amounts of legacy code for .NET enabled environments. Active Server Pages Plus is the highest layer and serves to separate scripts from Web-based code.
For the .NET strategy to come together, a number of things need to happen. First of all, Microsoft needs to partner with third parties that can supply the infrastructure for a subscription-based service. The company already has good relationships with Conxion and other application hosting services, and we can assume that these companies will be vying for Microsoft's business. Microsoft's products will need to be upgraded first, and then totally overhauled, to become compatible with .NET services. At a very low level, this means that Microsoft's products will have to be XML-enabled, which is something the company is working on right now. All of the company's products that ship in 2000 and beyond will be XML-enabled to varying degrees, from server products such as SQL Server 2000 and Exchange Server 2000 to client applications like Internet Explorer and Microsoft Outlook. In some cases, this will require two generations of upgrades.
For example, Microsoft will begin beta testing Office 10 this summer, but Office 10 isn't the first version of Office.NET. Instead, Office 10 will offer a few .NET services, such as Smart Links, while the company prepares a future release that will completely exist in the .NET world. In some cases - such as Windows - the company will probably have to create versions of the product that are .NET enabled and versions that are not. This is because it will take at least two to four years for the .NET strategy to unfold.