On 2 Jan 2009 12:10:57 -0800, [email protected] (Eric Bielefeld) wrote:
>I hope that when people say "Real Programmers don't comment code" that they >are being humorous, or just kidding around. Assemble code for someone like >myself who has done a lot of coding, but done it a long time ago, is hard to >read. Good comments make it at least possible to follow what the author was >doing. I think if I were the boss, and someone wrote a lot of assmbler code >without comments, they would either change their ways, or find a different >job. I believe we all are kidding. That said, unreliable documentation can be dangerous, and in my experience, documentation rarely gets updated adequately over time. I remember running a flow-chart program around 1980 that made absolutely no sense to me. Why have a program read a CoBOL program and create a flow chart? It's much easier to read the CoBOL program itself. Documentation is useful in telling us what the program is SUPPOSED to be doing, and why. It should tell us who made business decisions about the program. It should explain data flow and impacts outside of the job or dialog. Today's code should not be obtuse enough that other programmers have to study to find out what we did. But sometimes they should know why we did it a particular way. (Assembler needs more documentation than "self documenting" languages, but design the documentation so that it is read and maintained). ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
