Taupe Hat Systems

In my various travails as an ed-tech person, one of the things that has been a recurring theme (aside from poorly-written software ) or perhaps related to same, is the preponderance of educationally-oriented webapps being written for IIS and .NET. There are a number of problems I have with this.

[More:]

• Poor standards compliance. I have yet to see a dynamically-generated IIS page that was remotely close to being W3C compliant.
• Browser incompatibility. If your site requires ActiveX, you've just arbitrarily limited visitors to people using MSIE on Windows. Given the notorious diversity of platforms found in your typical K12 environment, this is a really dumb thing to do. Setting aside ActiveX, .NET pages tend overwhelmingly to work better on IE/Win than any other platform or browser.
• Market realities. Microsoft is unlikely to try and fix the above, as it would be contrary to their business interests to make life better for people not using their software. I'm not necessarily trying to criticize them for this, but it's a point worth considering. The only thing that will change them is more people using alternative browsers and platforms and demanding better support. So far, their response to this demand has been tepid at best.
• Programmers. Ask yourself if this pattern seems familiar: someone comes up with a great idea for a web-based application, and decides for various reasons to do it on .NET. A hiring call goes out and results in a group of relatively green programmers, but they do a decent job of getting the project off the ground. Sometime near the second round of major debug, a sizable chunk of the best programmers leave the project for greener pastures. The people left tend to be less creative or capable, and just as apt to create bugs as fix them. This is a pattern I've seen in .NET programs on several occasions - not to say that it's unique to the language, but it is certainly more prevalent. My guess is that this is because .NET jocks are often hired straight out of college, and the ones that are good quickly move on to better development environments (or much bigger organizations).

My take on .NET is that it's useful for two things: quickly and easily moving windows apps to network apps that work on windows and masquerade as web apps, and for prototyping. On the other hand, I've done prototyping in Dreamweaver using php/mysql, and it's pretty darn easy.

At any rate, the last thing in the world you should be doing is trying to write anything web-based on .NET that's intended to go into production in diverse environments - it simply isn't cut out for it, unless you can afford a very big and very skilled staff. Since the big selling point of .NET is that it's supposed to be easy to get off the ground with - MS conveniently fails to mention its weaknesses in maintainability - most organizations doing .NET work are neither big nor wealthy, and are doomed to either sink (in the private sector) or cause havoc and lots of extra work (in the public sector).

One last note. I live in Oregon. Here, we're blessed with an incredible surfiet of really good programmers who specialize in working with PHP, Apache, Linux, etc., all of which results from the location of OSDL in Beaverton, an excellent CS program at OSU, and the "halo effect" of having Linus Torvalds living here. Anyone claiming that FOSS programmers are hard to find or expensive in Oregon is too busy eating Microsoft's FUD to see what's right in front of them.

Moral of the story: If you're thinking of offering a webapp to schools using IIS/.NET, don't!