> #2, that I can't test or help improve something until it's delivered

We (myself and Hernan Canan from Columbia initially, and later Stefano D'Amico from Italy) were constantly producing things that could be tested. Once we got the base object framework designed, we were adding functions at the rate of 2 to 5 per day. We finished up with over 180 completed, and another 40 in the works. We have the full variable support. I added a feature called thisCode which allowed you to pass a reference to the program satck and local stack frames (local variables and parameters). We supported all VFP data types plus we added extended dates and datetimes, along with arbitrary precision floating point and integer support using the bignums library. It's the same library I used on the spreadsheet-like bignums DLL I posted recently.

We had many SYS() functions working, plus new ones. Several commands were working.

The whole framework was there. It just required labor to get it done, and testing. Stefano was an absolute blessing. He came in as an excellent developer and came up to speed in a few days and was adding functions at a rapid pace as well. I really thought we were going to make it. And had I not gotten sick, we would've. I'm still paying for the medical bills, by the way. Because they didn't find anything wrong with me, my insurance wouldn't pay for it.

I swung for the fences and hit the ball with a big sounding crack. But, the ball hit a bird mid-flight, and dropped somewhere near the shortstop. He was able to snag it and throw me out at second as I was just getting up to full-speed. My great heartache.