>>IF MS accepts 80% coverage as adequate (and I realize you did NOT say that) then they can only do themselves a disservice in the long run.
>>
>>In any case, I hope and expect that "pride of workmanship" alone pushes the vast majority of VFP developers to test their code to as close to 100% as possible.
>
>Is this opinion backed with actual experience with code coverage on a non-trivial app?

I think I can answer yes to that. I've worked on several non-trivial apps (all on mainframe) and they all got as tested as could possibly be done. That included EVERY compare construct, EVERY BRANCH (these were OS Assembler apps), etc.
It was a part of 'basic training', both in early work courses and in formal university education and at work it was 'reinforced' based on system/QA malfunction reports. Did we do ROIs to evaluate effectiveness - no.