Kent

I see that there are a few new interesting pages on MSDN this morning. I can tell they're sexy, because they have a black background. And there are reflections! (Reflections are the new black). While their overall "Web 2.0"-osity is hampered by the MSDN chrome, it is interesting to see MSDN at least taking a chance at creating interesting pages.

Wonder who built those pages?

Which leads me to think a recurring thought while I was at said fish shoppe: Do visually interesting pages work for information? One of the requests we got a lot was for "sexier" pages on MSDN, particularly where new products like Express were concerned. People would say, "Make them look as good as Apple's pages." (usually pointing at the Trailers page, or some other showcase site. Heavily graphical sites are certainly something that people can look at and think, "Wow, that's interesting", but when it comes to providing information, they tend to collapse into unpleasant, text-heavy, link dump pages fairly rapidly. I really don't think that there is anyway around it -- organizing large chunks of technical content is not a pretty job. You can create a few high-level buckets and load them up with content, you can create intricate nested hierarchies that are mostly meaningless to most people other than the hierarchy creator, you can try to organize the content for a specific audience (but mostly forget that it's a perpetual exercise), or you can forget organization and give them cake. All answers are satisfying in one sense and for some people, unsatisfying for others.

Knowledge is of two kinds. We know a subject ourselves, or we know where we can find information on it. (Samuel Johnson)

It's easy to point out the flaws in one organization or another, another to make the mistake of trying to suggest another. Personally, I tend to use "The Google" to find almost everything, including on MSDN.  I continue to think that a "Personal Guide" is still the best way to find content - I think that this is why About.com, W3Schools and TechEdBloggers continue to be useful (and I guess why Wikipedia is as well). Automating any system leads to false positives and negatives (who hasn't waded through  pages of entries on WinCE to get to the 'real' APIs?) However, I think we need some sort of common way of tagging and identifying content. Is something like DotNetKicks or DotNetSlackers the answer? I'm not entirely convinced (sorry Sonu).

Well, I guess another "no conclusion" post. I guess I need to think on organization some more. There must be something I'm not finding.