Hi, Craig.

>>>Personally, I really hope they would stay away from SQL
>
>I can't believe you just said that. Do you mind ellaborating on not only the inherent problems you see with the SQL Standard (SQL2 or SQL3 - you pick), but also what you see as a better approach to working with data contained in RDMS?

Well, I never really liked SQL although I've been using it for 20+ years. SQL has -in my opinion, so that isn't very important, anyway- a weird syntax, no type system, is not suited for anything except RDBMS (and there are a lot other information sources out there), and is absolutely detached from object oriented models.

For me, tweaking C# to make SQL fit into would be a kludge. It is not so different from what we have today. I'm not absolutely sure what the better implementation could be (I played with that, but no one funded me for a real project -yet <g>), but I guess it should be something closer to what a good ORM is, without all the impedance problems, and with all the efficiency a language-built thing could be.

I'm still doing research on my own about this space and I have a deadline on mid-September when I have to present about language implementatin at a Microsoft local event, so I can get back with some of this and comment if you're interested. I'll be talking about language research space, specifically about functional languages, contracts, aspects, dynamic languages, extended type systems, and of course, data integration. And -of course- I won't fail to mention VFP in front of an audience of "beard and sweater" .NET architects. 8-) Other languages I'll be showing include Eifell, Boo, Smallscript, C omega, and quite probably brand new implementations of Lisp and Prolog.

Now that I wrote all this, I'm starting to consider wether it is too much... 8-P

Regards,