InfoWorld: What are [Microsoft's] chances that they can become a serious competitor to Linux as a Internet services company, given their business model going forward?
Young: They probably can make it an important part of their business in the not too distant future, but can they really give up this heroin-like addiction to their product sales? Looking at their balance sheet today, it would be a miracle if they could give away their OS and still hope to meet their investor's expectations any time soon.
InfoWorld: Compare the kernel technologies in Windows 2000 with what you will have this year in Versions 2.4 and 3.0 of the Linux kernel for things like scalability, multitasking support, and support of pervasive devices.
Young: Keep in mind that the Linux kernel is evolving forward faster than it ever has before. The number of very high-quality engineering teams now contributing to it from Oracle, SAP, and Transmeta is really quite high. Microsoft has its work cut out for it because the only people allowed to contribute to the Microsoft project are Microsoft engineers. I don't care how big Microsoft is; they don't have the engineering talent that the whole Internet represents.
InfoWorld: Are you taking any heart in the report that there could be some 60,000 imperfections in the shipping version of Windows 2000?
Young: To be fair, if there are 60,000 bugs in Windows, then there has to be arguably 30,000 bugs in Linux. The reason most are there is largely because most [users] never run into most of those bugs. The catch is when you do run into a bug, it could be a material one for your particular application. Under the proprietary model, all those bugs that don't impact your system are of no interest to you. But if you can't do anything about them, they could impact your system in material ways.