And thanks to Jon Udell for taking an interest in my comments on Microsoft's not-so-new LINQ technology. I think that my comments there stand on their own merits, but I would like to respond to Anders Hejlsberg's comments.
Anders, I lead a Delphi developers mailing list for several years with hundreds of participants and even lamented Borland's abandonment of that tool in a Delphi Informant editorial in December of 2000. So I have tremendous respect for your skills, and if you ask George Goley, I know plenty about the benefits of FoxPro's tight integration between langauge and data. (See my earlier blog post about my first big project, where I leveraged that capability to turn a 100K LOC system to a sub-2K LOC system.) But you lost me when you allowed your talents to be bought by Microsoft, who continues to act against the interests of the larger developer community and FOSS efforts in particular.
So, when you equivocate that this is something new for "mainstream programming platform" (though Ruby is well on its way there) or in a "rich framework like .NET," I wonder if you realize how silly that sounds. Nothing like a lot of vendor lock-in with a little innovation.
So, I am still wondering — will you credit Lisp and Ruby for leading the way, or continue to insist that you are doing something novel?
I do agree that a type system that seems typeless to the user but that is smart about figuring out type-specificity at run or compile time is the best of both worlds. Just so sorry to see that you and George went the way that you did.