>>Has anyone used Lightswitch?
>>Any feedback?
>
>Used, no. I did see three presentations about it, including one by the group director, and my impression is that it's sort of stuck in a bad middle niche. It's not powerful enough for real developers but too complex, as Craig says, to be an end user tool. (Remember the Rocky and Bullwinkle cartoon show? "Again?! That trick never works!")


>>not powerful enough for real developers


I've seen some comments like that from several reviewers, Mike.

But I keep my eye out for things that might be useful to clients as a way of accelerating development and reducing development costs while producing outstanding applications.

My use of .NET so far has led me to conclude that while that it's a really powerful platform that exploits hardware and O/S resources well, it often requires an awful lot of code to accomplish relatively simple tasks.
I've built a simple framework for tasks like table maintenance, I/O, email, ftp, report generation etc and that cuts hours if not days from development time, but as soon as things get a bit complicated I wind up with humungous gobs of code that I inevitably have to go back and redo when I'm finished and really know how I should have done it in the first place.
There has to be a better way, I think.
I've looked at the frameworks out there and found them either too complex, too proprietary, too restrictive, too pricey or all of the above.

Lightswitch's architecture is attractive to me in that it's data-model- based. That's a good start, it seems to me.