>But dBASE III was way cool compared to dBASE II ;-) 10 work areas compared to just Primary and Secondary, and that was just the tip of the iceberg.

I too remember that, though briefly - a friend had made a CP/M dongle for his Commodore 64, and we had dBase II on it. I played with it while learning Clipper - in which I wrote exactly one app.

>Setting aside snarky comments, I do remember both versions from the pioneer days. At the time I was a mainframe programmer and it seemed miraculous that you could do all this yourself on a computer you could control, without having to battle your way through forms, your boss, system programmers, and other defenders of the big machine in the locked room. (Right next to the room with boxes and boxes and boxes of greenbar paper. The punch card readers and rows of tape drives were right in the locked room). And actually do something useful! My first app tracked and scored a football pool -- you can't get much more useful than that. It was beyond cool, it was transcendent, regardless of how primitive it seems today.

While I had the (tough?) luck to be both guys - i.e. in charge of the big iron and writing forms using VT100 sequences (remember those, eh?), I remember the big thing about divorcing the table structure from the tables engine (though VAX and RMS had that, although not relational - but it had indexes and records; it was up to your code to split them into fields). The big deal, for me, was that you didn't have to recompile everything and write a conversion routine when you just wanted to add a field.

I remember the first wave of inflation in '88 caught me on the VAX - we were expanding numeric fields, replacing them across the board with BCD packed (comp-3, IIRC) with enough digits to survive several years. No renaming, no new fields, just changing the field type and width. It was still a number. In dbf world, you wouldn't need to change a line of code. In Cobol, four of us worked two weeks to do that.

Never looked back :).