John,

Great post, I think it summs up the answer of the question why some of those vocal .NET-ers bitch about VFP. They never unleashed the true power of it.

>Recently I realized that many VFP users, including "gurus", used VFP like VB: as a front end to SP accessed via SPT, or even using ADO as recommended by one leading prophet. For such a person, moving to NET is a natural step; it truly does everything they did in VFP. This explains why some people have been saying since 1.1 that NET does everything VFP does. That's not a criticism of anybody, it's just an explanation for the heated disconnect that was so frequently seen around here.
>
>As for memory-disk spanning: not sure why you chose to raise it again ;-) but loading resultsets of indeterminate size and munging them repeatedly at speed without hogging resource is an awesome ability. For those who've never done it, it really is one of those "you don't know what you don't know" scenarios. MS has clearly stated its intention of providing this sort of feature in NET and if they can do it without memory-disk spanning (and several potential mechanisms have been suggested so far) then I'll happily hoot and holler (for joy) along with the best of them.
>
>Apart from that: I think you're completely correct in your gloomy assessment of the corporate bespoke development market, especially after the MS announcement. However, I'm not convinced that everybody in the market noticed the announcement or could care less. I guess we'll see as time goes by. IMHO the fascination with development tool is vendor driven and will dissolve as people start to rely more and more on personal devices, which MS says will be the norm within 5 years. Rather than choosing a development tool, people will choose their service provider- just as people already choose Google/online shopping etc etc without caring whether it's written in NET or Java or PHP or whatever.