OT: Language for the ages - test-drb@ccmp.vtda.org

17 Oct 2005

At 12:00 -0500 10/17/05, woodelf wrote:
...
 Are we tallking real gates as in TTL,  or FPGA style
design? Forth does
have several advantages
over say fortran since they removed the Sense Switch requirment of
fortran.My worry is that you
may not be able to high-level threshold logic and memory anymore for
industral problems. 
	Either TTL, or in 20 years, whatever the then-current version 
of FPGA or ASIC is. (To tell the truth, I had in mind re-creating the 
Harris RTX2010, but that's just me.) The point is, creating from 
scratch and then verifying a system (hardware, OS, software) to run 
FORTH reliably would be far easier than the same task for FORTRAN or 
JAVA, and in 20 years, that may be the path you'd choose to take.
	I am depressed at how difficult it has become for any one 
person to really understand either all of the hardware or all of the 
software (let alone both) in any commodity system, whether controller 
or desktop. Two more decades in that direction, and who'll know 
*what* is "under the hood"? Viruses, mal-ware in distribution CD's, 
Pentium floating-point multiply bugs, cache sync errors, my personal 
favorite - a compiler that forgot to save the contents of its 
floating-point registers when it got an interrupt in the middle of a 
calculation - and so on .... That stuff will all affect "emulators" 
too. So if your application is mission-critical, you may want/need to 
re-build the platform system from scratch. Better a simple set of 
system requirements (a la FORTH) if that's where you end up going.
-- 
					- Mark
			210-522-6025, temporary cell 240-375-2995