Financial Investment companies here in the US often use the phrase "past performance is not an indicator of future returns". The same holds true with software. Just because old applications continue to run fine does not mean they will continue to do so in the future. For example, there are currently issues with FoxPro and Visual FoxPro running on Vista. I doubt Microsoft will go back and fix these issues with Fox versions that are no longer supported.
---
We have been a VFP shop for almost 10 years. It is a great environment for quick development and stable applications. That said, we're done with new development in VFP. Here's why:
-The won't be a 64-bit version. In my opinion, this is the biggest end-of-life indicator for this product--even more important than Microsoft's support until 2014. Sure Vista and server 2007 have 32-bit compabitility mode; but I expect by the next Windows OS releases (3-5 years down the road), running 32-bit software will be a dying trend. Some larger companies may even have initiatives to run 64-bit software whereever possible. Why process 1/2 the bits at a time when you don't have to? By 2014 VFP will be equivalent of modern day Cobol. There will be a bunch of legacy apps left that should have been converted years ago.
-Have you seen the new .NET 3.0 stuff? WPF offers extremely flexible anchoring capabilities--something that works okay in VFP, but not great. WCF Services offer new levels of sercurity and binding methods.
-DLINQ and XLINQ. These 2 language enhancements are going to save .NET developers lots of time. Simple tasks will be simple code. Plus there is a tool that will build business objects from a relational database.
---
To say I'm certain would be an overstatement. It's sort of a gut feeling comprised from the last 3 years of VFP-related information. There are almost no new books...there isn't currently a plan for version 10...the publication we were getting for over 7 years became redundant and focused on specific uses with limited use...the newsgroups I watch have slowed down to a crawl--sometimes no messages for a several days...etc.
Another indicator--the Sedna code has to do with interaction between VFP and VS, but I haven't seen anything that gives VFP the ability to host managed code. We can access managed code, but we can't use the nice interfaces or benefit from the new controls. I think we're being quietly coerced away from VFP to VS. They're giving us tools that expose us to VS, but in truth they want us to see what we're missing. Maybe they're right. I haven't used VS enough yet to know for sure. I do know ClickOnce deployment is very nice. In 15 lines of code I can have my application automatically check for new versions upon load and update itself.
My hope is that development time will significaly reduce and my users will still experience a signficant difference. My biggest fear is that development and/or training time will increase significantly.
---
This has been almost a touchy subject for me. I started using foxpro about 3 years ago and as soon as I really got into it I was hooked. When I heard that VFP was being "phased out", I was, to say the least, somewhat devistated. I keep hoping to hear that Microsoft has changed their mind and will continue to release new versions, but I'm not as hopeful as I used to be. I haven't personally dived very much into the new .NET world yet, but I see that it is the way to start heading. Not only is there less support and less forums and what have you, there's also quite a lot less VFP jobs out there and most of them that you do find is usually converting legacy code as opposed to creating new innovative apps. I'm not really looking forward to it (yet) but I'm going to have to start really delving into .NET - gotta keep up with the times.
---
Microsoft has never said that VFP.Next (Sedna) will be version 10. I talked to Alan Griver (yag) last month in Germany and he said the decision on what to call it or how to package it hasn't been made yet. My bet is that it will be more of a service pack or "plus pack" or "VFP9-R2".
Also, Microsoft has stated that there are no current plans for anything after Sedna, but that's the way it's always been with each release. That the next version hadn't been planned, but there was one.---
Microsoft want everyone to believe that 64-bit is the way to go, but unfortunately this is a rosy-tinted view of the real business world. Companies simply do not upgrade hardware and software just because Microsoft release something new! Things like "Return on Investment" come into play and unless there is a good reason to spend money on new machines and operating systems it just wont happen! Imagine the cost of switching thousands of users over to 64-bit and then look at the benefit of spending that money....
---
At some point 64-bit still seem inevitable. My Dell rep tells me that many larger businesses cycle their hardware every 3-4 years. In smaller businesses hardware lifespan doesn't usually exceed 5-6 years. At some point the purchase of new hardware is inevitable. If the new hardware is 64-bit, VFP ends up in "Compatibility Mode". That's a bad place for any software product to end up.
---
Regarding 64-bit, I believe it's sort of over-rated and is being hyped pretty heavily. You know IBM AS/400s had 64 bit chips and OSs in the early 90s. Plus you just had to recompile existing apps to get them to run as 64-bit. While the AS/400 is popular with certain industries, I bet most people have never even heard of it or it's capabilities let alone beat a path to it to take advantage of the 64-bit capabilities. Some things I've heard about Intel/MS 64-bit - it will only have advantages for extremely large databases and may actually slow regular applications...
---
we know that businesses in Africa and other developing areas worry about cost and maintainability given how low GDP is in many of these countries. Given that most westerners use only a small part of the capability of packages like VFP 3>, excel 3> , is there not an argument for the startup distribution company in (say) Zambia to be able to use an earlier OS and an earlier software like VFP 5 to be made avialable and supported locally rather than tell them they need the latest bells n whistles super dooper OS and software when in reality it could be years before they need this capacity. If we in the west/north truly want to help Africa, it seems crazy we basically want to bin so much fabulous software prematurely. What do you 'pros' reckon?
---
