TITLE: the growth of programs (Newsgroups: comp.lang.java.programmer,comp.lang.java.advocacy,comp.lang.java.help,comp.lang.eiffel,comp.lang.c++,comp.object,comp.software-eng,comp.sys.unisys, 15 Dec 98) PRESCOTT: Ralph M. Prescott >> Both of these posters have failed to realize the inherent nature >> of the competitive process (or human nature) to continually attempt >> to reach beyond our grasp. As tools become better, problems that >> were previously economically unfeasable now become possible. >> Not guaranteed, but possible. >> >> Bottom line...it isn't EVER going to become simple. MARTIN: Robert C. Martin > Right! In the 60's a 10,000 line program was a big program. It was five > boxes of cards and weighed in at 50 pounds or more. You needed a hand > truck to carry it down the hall to the card reader. In the 70's a 100,000 > line program was a big program. The source program filled up the tape > cartridge on which it was stored. In the 80's, a million line program was > a big program. Source code control systems and compilation systems had a > rough time managing all that source code. Now, in the 90's, programs > aren't respectable until they pass the million line boundary. Windows far > exceeds ten million lines. Most fundemental desktop applications are in > excess of a million lines. I consult for many companies who work on > multi-million line applications. And the trend show not sign of abating. > > In the next decade we will cross the 8th order of magnitude. We'll need a > whole different set of tools, processes, methods, and technologies to deal > with source code on that level. [snip]