The purchase of Fox Software in 1992 was very strategic for Microsoft, and was the biggest corporate purchase Microsoft had ever made up until that time. Aston-Tate's dBase was still popular, Borland had Paradox, and there was PowerBuilder. Microsoft needed three things from the deal - the Fox developer team, the Fox technology, and the customer marketshare of FoxPro/FoxBase. Microsoft was just starting work on Access and it was more targeting power users, but there was still some overlap. Visual Basic was still in it's early days. It took almost 4 years, after VFP 5.0, for Microsoft to see the long term trend and their own strategy around VB and Access to decide on less strategy for VFP. Basically, the VFP customer base and sales went from increasing to decreasing, and as I said, in business if you aren't growing you are dying. There was an estimated 500,000 FoxPro developers at the peak around 1995, and millions of computers with FoxPro apps running (either DOS or Windows based). Some at Microsoft not real connected with the product and community under-estimated how important the Fox community was or how VFP developers would not move to VB (or .NET) that easily. Hope this clarifies the timeline around the strategy a bit more.

Just keep this in mind. The two people at Microsoft who convinced Bill Gates to buy Fox Software were Adam Bosworth and Tod Neilsen. Adam was the general manager in charge of Access 1.0, and he went on to start a competive company with Tod called Crossgain which got bought by BEA where he was a VP, then he was one of the VPs at Google for several years. Tod ended up being the first main product manager for FoxPro (he was in the keynotes at the first few big Microsoft VFP DevCons) and then became VP of Marketing for all Microsoft developer tools, before leaving to Crossgain in late 2000. Tod went to BEA, then he became CEO of Borland and then killed Delphi (ironic). Tod is now the COO of VMWare, and Richard McCaniff (first general manager for FoxPro 2.x and VFP 3.0, when the team was the largest) is now the VP of Developmentat VMWare. In 1992, Adam was also in charge of the first ODBC 1.0 technology, and he was interested in seeing if I would be a C++ developer for that team. It wasn't something I was interested in at the time, as I wanted to help bring VB GUI features/tools to VFP written in VFP. Adam once told me that the primary reason MS bought Fox is for the key developers. Dr. Dave Fulton said just a few years ago (in the video we had him to for VFP 9.0) that he didn't expect FoxPro to evolve at Microsoft as long as it did. I think he and many others thought it would end after VFP 5.0, and like I said, it did in a meeting I was at with the entire Fox team in which Tod Neilsen made the annoucement Fox was disbanning, and he cried (although myself and others thought it was more marketing/morale acting than sincere).

Eric Rudder, who was the architect of VFP 3.0 and reported to Richard McAniff during that time in the early 90s, ended up being in charge of the VB4-5/VFP5 team, then in charge of VS 97, then became Bill Gates' technical assistant for 4 years before running the entire developer division launching .NET. I think with many people in management and marketing wanting to kill VFP after 5.0, Eric was the one to ordered it to continue, meaning, he was the one who said there would be a VFP 7, then 8, and finally 9, against the wishes of many managers and people at MS. Eric understood the Fox technology, the community, and also the importance of all the runtime apps means to Windows and overall MS customer loyalty. Things were difficult with VFP 8 and 9 since the VFP management was under the VB management (including budget). Hey, there are useful in cool features in VFP 9.0 that still won't be in the next 3 versions of Visual Studio. :)

So, would this make a good book excerpt?? :)

>I should have replied to your original post. It was quite interesting. I emailed it to some people. Could you clarify something? You said "Craig was right" and he - if I remember correctly - was saying the EOL decision was about revenue (in the less-popular diminishing form). But at the same time you indicated how early on the execs wanted to kill VFP and how it competed with Access and VB. Certainly they could see that up front even before the purchase, right? It sounds like the idea was to take what technology they wanted from it and kill it to begin with, which is what I feel the evidence has always pointed towards. The diminishing revenue is then just "planned" (your word) and really an excuse. The root cause is that they simply wanted to kill it and how they did it is irrelevant (and the method and final "reasoning" certainly shouldn't be thought of as the root cause).