>The GUI interface stuff and the display of PEMS in C# is pretty good "as is". So we give up the Report Writer, which frankly I would miss. We give up VFP databases. As much as I like VFP tables, it would be better to have a single file database.

For most app development you are totally right. OTOH, currently I am "scripting datamining". I can honetly say that NOTHING I know of would fit the task better than vfp - just happy coincidence that I am quite fluent there<g>. The possibility to check different models/assumptions just by replacing single tables with the OS is worth gold, even if I have to dance around the 2 gigs in 15% of my queries, which is growing to be a PITA. Having the script language database aware AND oop cuts down programming time like crazy. But this is a very special case.

>Which lead me back to why I think that a new Fox, built by a modern day Dave Fulton, could be product that would make money.

Have my doubts here. One of the pitfalls of the *whole* xbase market was the splintering of file formats: from the esperanto of interchangeability grew better, but incompatible file formats. Not really secure in any way since the base dbf-format can be read easily with nothing more sophisticated than vfp's LLFF, but business integration was hindered. Hence XML-bloat became fashionable...

>Which now leads to another thought. Open sourcing is not the way to go. It is development by committee. No open source project has really become commercially viable or has captured a significant market share. I think that most of the really good languages, the ones that programmers like, were all developed by small companies that were VERY focused.

You already mentioned Python. The languange is a very clean design, is easy to grok for the first steps, very dynamic and has some aspects of oop licked which vfp doesn't even begin to target. And even the MS implementation is remarkably "open", if I understood the legalize correctly. But Python HAS the BDFL (benevolent dictator for life), which probably helps a lot in the committee area, but is hard to measure.

IT as a whole has probably left the arena where successful languages / OS might be built by startups. Pockets are usually not deep enough to go against large companies, so probably the most realistic chance to start something "new" is open source - another witness might be Ruby on Rails, since there are chences that a critical mass of programmers not costing company money enhance the raw diamond of a good idea into a widely usable product. INternet-based ideas still can have enough novelty, but commercially successful database development systems/languages going against everything from oracle via DB2 / Sql Server down to PostgeSQL on the enterprize side to vfp via Codebase, VistaDB and a few others down to SQLite on the desktop side ?

my 0.02 EUR

thomas