What is old is new...
Back in the 70's we did the same thing - in Assembly or Fortran on the Cyber 74 or the Univac 1108 - and we didn't have any kind of fancy development environment, unless you count getting a chair with arms in the room where we punched cards - ah those were the days, submitting a card deck and going away for hours while we waited our turn for the deck to be run and the results printed or plotted and then waiting for the display light to indicate our job was done.
Also did that on every other computer I worked on - micros, minis, controllers, and on. Why? Shouldn't these fancy IDEs do all that? Well, no, sometimes you were trying to debug a highly sensitive to time code bit, interrupt or driver or whatever. When working on something that is broke, you sometimes need to figure out where it is broken at speed. The old print statement worked well, but sometimes you needed to send the place code off to a port so you could get back to running the real code - a quick write to a port and back to work.
And then you found out that it worked differently with the debugging statements and it was back to the drawing board trying to figure how to measure without changing the result... all hail Heisenberg!!!
Ok, that reference might be obscure for some, Heisenberg's Uncertainty Principle states there is a limit to the simultaneous measurement of position and momentum. That same principle applies to the examination of a working machine or code.

LinkBack URL
About LinkBacks
Reply With Quote


Bookmarks