With regards to the F120 - I'll guess point well taken, since I'm not familiar with it, but I have looked at some of the "exotics". For the majority of 8052 cores shipped, we're talking about rearrainging window dressing. Some flash memory here, a few extra I/O pins there, lower power consumption, faster clocks, more effecient instruction execution, etc... But the same set of instructions. The 8052 survives because its useful, well understood, multi-sourced, and the problems it solves have not substantially changed either. It might be interesting to compare the numbers of plain old 8052's shipped vs. say the 251's, or something more recent like an embedded microSPARC. But it would just be academic. We could also discuss why 5v TTL compatable logic is still being made...

Intel's x86 arch is a mess. It's always been a mess. In my dreams I picture what the world would be like if IBM had used the 68000 in the original IBM PC. Then I wake up... :(

ISP - At it's simplest, 4 bits of wire, and DB25 shell, a couple resistors, and some free software. Uses chips that I have to mail order, minimum order $25 plus shipping. Plain old parallel programmed pulls are less than a buck each at the local "chip junk yard". As a hobbist, I can prototype with them faster and cheaper, provided I can make use of them. To do that, I'd need a programmer. Were I to take a prototype to production... ISP for sure.

The point is... The guy wants to build a tool. That's a cool thing! He'll learn something and enrich and further himself. Someone else may pick it up and do the same, even if it's just "hey this old way adds an extra produciton step, and costs too much $$". It doesn't help you, but the next tool he creates and gives away may be different.

Mostly, I just think you're being too hard on him...