>>You probably did not need to interface different data "stores" ranging from excel up to DB2 and create ad hoc selection results. Rapid turn around in interpreter, easy OOP to build a fwk of similar query attributes, easy scripting via command window, output options for any target from flat file over excel up to direct inserts into DB-Backend...
>And yes, integrated data engine helps a lot.
>
>My data stores I have to interface with are sql server, oracle, access, excel files, word files, text files, loose xml files, email messages (exchange), Sharepoint data, MOSS, web services, Streaming video / audio just to name a few and they ALL share data. Our environment is massive and the flow of data from a multitude of different sources is staggering.

No doubting that - I was also never in the camp that said you could not write an app in dotnet (would not want to return money for the things they paid me in java for<g>) - but I wanted to show that there are areas where vfp still is better than dotnet.

I was harping more on the quick development of throw away products, sometimes just reports for the board on last weeks data, and always with some new "twists". Creating a SQL-generator fwk in Python would be the next approximation: but thinking, generating and testing in the same language gives at least me a boost.