Ed Yourdon, RIP...

Ed Yourdon, RIP...  For geeks of a certain age, Ed Yourdon is a name that conjures up memories of computer programming's first tentative steps toward engineering, rather than as a pure art.  His book Structured Design was a milestone.  It's largely forgotten now, as its essentials have been absorbed into the programming culture as well as into successor design “fads”.  For me that book was revelatory ... it truly introduced me to the idea that one could usefully think about patterns in the way one wrote code.

I had the chance to meet (and talk with) Ed on two occasions, both times at conferences.  Once was in the early '80s, when my then-employer sent me to a conference put on by Roger Oech; the other in the early '90s at Comdex.  At that first conference, I had lunch with him, and we share some memories of the early days of microcomputing.  Even then, both of us were surprised at how quickly microcomputers were ramping up – and neither of us had any clue that those days were but the beginning.

Ed's book sticks in my mind mostly for one thing it accomplished: teaching this young geek why the spaghetti code (the only kind I'd ever written to that point, with jumps sprinkled randomly throughout) was a bad idea, and how it could be better done.  I still remember the conflict I felt then, between the rigor and beauty of structured programming and the need for performance optimizations.  I really didn't understand, then, that the low performance microcomputers of the day would shortly be antiques, and that the cost of a few extra computer instructions would soon be zero for all practical purposes.  It seems downright silly today to assert that it might be necessary to write some unstructured code to get the performance we needed, but back then it was quite a normal thing.  It still happens today, but on increasingly rare occasions...