VFP testing has been largely automated for quite some time; there are about 4000 tests that are run for each OS/Language combo in a full test pass. Automated testing is great - before I left I had it where you went to a web page, selected a bank of machines and a series of tests and off you went. So automation allowed us to run a huge number of tests with minimal effort.

My understanding is that these test runs were done for SP2. The guy I trained as my replacement moved on so I don't have any direct test contacts anymore.

Now, one observation I think may have a bearing. The most difficult thing to test is UI because it's all "mechanical turk" stuff and very difficult to automate. If you don't have a large testing staff it's quite easy to miss UI issues.

Something else - Calvin, Aleksey, and Richard could code VFP in their sleep. None of those guys are prone to checking in buggy code no matter how many other tasks they have.


>Yeah, I'd agree that three developers is fine. In fact, small teams often work better and faster. I would say the problem is that they are stretched thin. Plus, VFP is not a career-builder for them anymore. They are busy learning and contributing to .NET and when someone is stretched too far the quality of their work is likely to suffer. I don't care if you're Einstein, if you have too many irons in the fire you aren't going to figure out that E=MCsquared (well, should have been superscript, but I'm not sure off-hand how to do that or if the UT will allow it). And I'd wager that the testing is mostly automated now, and that's not some magic bulet. I don't think the testing in the last couple of releases has been given the resources it needed. M$ never gave a hoot about this product from an upper mgt. standpoint, why should they now?
>
>>Javier,
>>
>>I disagree that 3 developers is not enough. We're talking 3 highly professional and experienced developers. In fact, these same three were solely responsible for VFP 9 RTM and most of the VFP 8 work.
>>
>>These guys are under pressure to perform in other areas in Visual Studio as well.
>>
>>
>>>I guess the bottom line is that Microsoft is behaving like a small (very small) software company, and in a very irrespectul way towards its developers and customers; in the end VFP users are their customers, like it or not, they bought a product from them and deserve a response much better than SP2.
>>>
>>>When you buy software from Microsoft you expect that more the 3 developers are working on your solution (as John Koziol said in a thread). That's too little people by any MS standard, and a insult to VFP developers given MS size. I do not know if something can be done, but if there is any petition to sign I would be more than happy to stamp mine. It is not to tell MS to develop 5 more versions, it is to tell them to stand by their quality standards, and their promises (this SP2 is not support till 2015). If they promise support till 2015, please give it, and at least keep the product running in the most recent MS operating system (ie fix the little Vista issues at least). Frankly, I thought MS was a more serious company.
>>>
>>>If they don have the money, I wouldn't mind to pay a monthly fee to support MS developers to keep working in SP2 and Sedna, but seriously working. And I know they have the money, so what's the point in doing this? They have stopped releasing more versions, that's OK, we all know, but please release a decent SP2 and Sedna as you promised. Are MS going to loose so much respect for so little money? It seems so, and I just cannot understand it, I would not be able to make such a thing to any of my customers.
>>>
>>>>I've read some of the SP2 threads. And I've read various complaints about certain features no longers working in SP2. The bantering seems to go back and forth as to whether or not features have been broken by SP2.
>>>>
>>>>What is the bottom line?
>>>>Does SP2 break some traditional features of VFP9?