In most instances, it would be extremely difficult (if not impossible) to build the average app without interrupts.
but we agree, that is exactly what I state.

from Jack:
"promoting the use of polled code wherever possible. Though it is important to recognize their limitations, interrupt-Luddism is an awful mistake."

Jack is expressing my thoughts exatly.

You should note that throughout the article there is no mentioning od "testing" only "debugging" as to what might become more difficult when using interrupts.

I amde a program a very long while ago (on a Lockheed SUE) that used NO interrupts (not a real time thingy) and someone calculated/estimated that there was over 10 million possible combinations of getting from start to finish. If that is "easy" to "test" please tell me how>

The whole issue is that the word "testing" has crept into the discussion and again
TESTING IS WORTHLESS AS A QUALITY DETERMINATOR
there is no way you can ensure error free functionality by testing

Can we please get away from discussing that worthless subject.

A correctly designed project is, in my opinion, more likely to function glitch free, if interrupts are used that if polling is used.

Erik