I'm the kind of person who'd rather quickly plagiarize a mountain of useful code and give credit to the original programmers (thus, of course, turning the action into "copying" and not plagiarism) than to slowly write my own useful code, however sophisticated, and take credit for it.
Here's another way of saying the same thing - I'd rather get to look at a lot of useful examples that I could probably reuse later, than to read a whole lot of useful text but have to write my own code.
If your thinking is anything like mine and you do even a little bit of work in the world of data mining, then you owe it to yourself to look at http://www.sqlserverdataming.com. The site is a virtual cornucopia of useful and reusable content. For example, they actually have a live instance of SQL Server 2005 Analysis Services attached to the website so you can play with their sample applications right away. There are literally too many great samples and content to list them all. One other thing that I really like about the website is that you can subscribe to any of the major content sections, such as whitepapers, to receive an alert any time something new appears on the page. There's also a good newsletter worth subscribing to.
I hope this helps!