I'm in the middle of a three-day training workshop at work for a pricey new web application that our university bought. I was actually trained on it about 18 months ago, but it's taken them this long to get it installed and working and ready for deployment, so I'm being trained again. (To be fair, some of that time was intentional delay on our part to just wait until the next version is released.)
I am grandly disappointed in this product. After 18 months and a new version (which turned out to merely be a "maintenance update"), I expected far more progress towards the product being usable. I've been holding off on projects for the last year because it was always "just around the corner," and we've been getting our ducks in a row so that when the time came, we could start our migration quickly.
As of this writing, though, I am seriously considering turning my back on it altogether and rolling my own solution. Our users are used to a much more clean, simple, and task-focused web application than what we bought. I don't doubt it's good for some people - I don't know who - but I just can't see it working for us. (The fact that this means I'd have to develop and maintain the alternative, instead of washing my hands of the entire responsibility, should give you an idea of just how much I don't care for this product.)
There are many limitations to the product, but the one that really struck me while using the app is the inscrutable interface. With the current crop of emerging web interface design principles, it's really hard to look at web interfaces that feel like 1999. It's been a long time since I've worked with a web application that was so solidly presented from the point of view of the internal implementation - the "programmer's view" of the system - instead of being focused on helping the user get done what he needs to get done. It really is staggeringly hard to use.
The sad thing is that it should be much easier to use. It makes me want to write a letter to the CEO and say, "Please, fly me out to your headquarters for two weeks, give me the power to have your programmers change the web interface, pay me reasonably well for my time, and let me make your product better." I'm no Nielsen or Zeldman, granted, but I'm sure I have enough suggestions to keep 'em busy for a while.
I'm going back tomorrow for more training. Yay.