>>Some have gone longer! There are reindexing utilities - mostley to remove the deletes, clean up the CDXs and do a physical sort on the primary key fields - or for repairs, when a system crash occurs.
>>
>
>The above is what I am talking about.

Yes, but they do it once a blue moon. It's in-significant. Even [one of] the bigger, global financial servers (like Bloomberg), schedule weekly down times. Normal operation of VFP will not hurt the data. It also seems that projects relying more on low level xBase operations (like seek and replace) and less on the very fine packaged 'batch' or macro data engine (select from and update)have fewer problems. The more discrete approaches seem more dependable and take better care of the data.
>
>>No - Luck has nothing to do with it - I practice - and have delivered a lot of mission critical, full cycle (full system) xBase projects. I take a lot of pride in my work and don't settle for half efforts. I am glad that less seasoned developers, that can excuse themselves for being a little sloppy, have clients (that probably really don't need SQL), willing to pay the cash so they don't have to pay attention.
>
>Must be nice to be so good that you can eliminate the inherient problems with FoxPros architecture. And, well YOU CAN'T as your first paragraph indicates.

When I initially responded, I was really PO'ed with an intense debug session. I was feeling insecure and may have appeared arrogant - which - is something that has made therapy a consideration. I was rude. My appologies. But - that still does not belie the fact that some operations do not need SQL. SQL implies more expense than just the licensing.

MicroSoft could scale up VFP's data engine so that xBasers and selectors could do their thing regardless of architecture. VFP's data engine is more advanced than anyone imagined it would be. No telling where it could go. Any limitation we see in that engine can be addressed. They [MS] could ask and recieve an annual license (think of all the VFP apps in the world) if they [MS] could make the architecture transparent to the development environment. Imagine - just writing VFP code, not having to go out of the box.

There are so many priorities just dealing with the desktop, rules and data structures. I guess the choice is to attempt to play at the highest level in a narrow range of diciplines, or relegate to a dilution of expertises, yet have a moderate understanding of a broad range of disciplines. Simply put, some race and others cruse.

Gosh - that was kinda of long winded, wasn't it! Thanks for listening.

Again - my appologies - NOT! just kidding