If you disagree with this, contemplate the names used
outside of academia:
If you do not believe that the output data, either as effects or as written/displayed information, is the only use for computers, then we have nothing to discuss and I hope never to encounter you in a professional setting.
The only rational approach for designing a computerized system is to start from the outputs: what is the system output needed to do? Before that is specified to a reasonable detail, any other effort is futile and counterproductive.
At this stage, describing, much less specifying, any algorithms, programming, or methods is destructive to the goal of producing a working system.
At this point, it is absolutely necessary to determine whether the
output can be specified in sufficient detail.
If not, the project is a research project, which has somewhat
different requirements than a production project.
Confusing a research project with a production project inevitably results in confusion and often results in very significant cost and time overruns.
I will describe the different methods and approaches necessary for research versus production projects in a later essay.
©2009 Geoff Steckel All Rights Reserved