>>Reliability is a matter of degree, not an absolute. You can't say that SQL Server never fails; it has a non-zero failure rate. DBFs also have a non-zero failure rate. Clearly, neither product is 100% reliable. How then, can one of them be "reliable" and the other "not reliable"?
>
>I never said that, you did. You are drawing your own conclusions but not based on what was stated. It is all about degree, not absolute.

A direct quote from your earlier message: "DBF flat files do not offer security, reliability, or replication." You are clearly treating reliability as an absolute.

>
>>As MS Visual Studio Data Product Manager, making that statement about reliability without qualification is, literally, ridiculous. VFP developers and their current and future customers deserve better.
>
>It said that SQL Server is "more reliable" than DBF files, which is a true and accurate statement. If you are saying that DBF files are as secure and reliable as data in a SQL Serve database, then I would assume your current and future customers (or employer) deserve better. It did not say you can't or shouldn't use DBFs for the main data store. Of course that is a viable option and works great in many cases, and the statement never said not to use DBFs. It is simply about having all the accurate facts in order to determine what is best for you and your applications/customers. I don't have any intent on changing the statement made, and I don't think a bigger deal should be made about it by parsing more into the statement that it deserves.

I agree that, in the general case, SQL Server is a more reliable data store than DBFs and have never said otherwise. Ironically, just having read the VFP FAQ page in question I think the statement you made there is fair.

However, both VFP and SQL Server have been in widespread use for some time. Coming from you, on this technical forum, I think we deserve some hard numbers or formal study results that show just how much more reliable SQL Server is. If you don't have those numbers you can hardly make the statement that SQL Server is more reliable.

It was the quote above, in your earlier message, that caught my eye and to which I object. Treating reliability as an absolute is FUD at worst, and a marketing ploy at best. Using that kind of "logic" we could "reason", for example, that Oracle or DB2 running on Linux/IBM hardware offers "reliability" and SQL Server running on Windows does not.