Re: Schedule, Cost, & Quality: Pick Two

[SteveFJong -at- aol -dot- com]

Tom Murrell <tmurrell -at- columbus -dot- rr -dot- com> wrote: "Pardon me if I don't think the 
ivory tower resembles the world I'm working in."

Tom, I assure you I'm no academician; I'm a working professional. (Well, 
okay, I don't work; I'm a manager now 8^) Neither were Deming and Crosby. 
Their ideas were and are entirely practical. There are two arguments to be 
advanced; the first is that such quality programs can work in the real world. 
Off-topic though it may be, I'll take a shot at that one now.

When Dr. Deming approached Detroit with his ideas for statistical quality 
control in 1950, he was turned away. The automakers were too busy selling 
cars to returning GIs to worry about anything other than production. Instead, 
Deming went to Japan, which had been bombed flat and had a negative national 
net worth. The desperate Japanese, whose postwar products were the worldwide 
symbol of shoddiness, embraced him as a teacher and potential savior. 
Following his methods, the Japanese completely reversed their fortunes. Today 
Japanese products are the worldwide symbol of high quality, and the country's 
highest honor is the Deming prize.

In 1980, when Ford Motor Company came crawling on its belly to Deming for 
help, they advertised that each car was inspected 22 times before shipment 
(that's the QC model in its full fury). Nevertheless, in that year Ford 
recalled more cars than it shipped, and it was near bankruptcy. Under 
Deming's direction, Ford instituted sweeping changes in its quality program, 
and adopted the slogan "Quality is Job One." They didn't do it with more 
inspection, they did it with less. Today Ford is in excellent shape, and in 
quality surveys its cars are as good as--well, Japan's.

Philip Crosby's Zero Defects program, which he instituted as ITT's quality 
manager, saved the company $700 million a year when he left to form his own 
consulting firm. He did it by reducing scrap and rework by about three 
quarters. I highly recommend his book _Quality is Free_ to anyone interested 
in a good general introduction to practical quality.

To try and keep this posting from being deleted, let me add that in 
documentation terms, the QC model is the model of drafts cranked out in haste 
and edited at leisure. Should I hire an editor? How many drafts should I 
write for best results? When will I run out of time?

When I started at my current company, the writer of our flagship product 
turned around drafts as fast as humanly possible, yet the documentation was 
three releases behind the software.  Today, his successor supports twice as 
many products, three times as many documents, and probably five times as many 
pages--yet he turns around the documents (almost) with the software. One big 
reason is that we no longer turn out endless, unplanned drafts. Instead, he 
writes a doc plan, one draft, then a signoff (finished) draft. And this is 
service-bureau software, installed in our own facility, than can change--and 
has--right up to the day before it "goes live."

Quality programs aren't easy, but they work, even in the real world.

          -- Steve

Steven Jong, Documentation Team Manager ("Typo?  What tpyo?")
Lightbridge, Inc., 67 South Bedford St., Burlington, MA 01803 USA
mailto:jong -at- lightbridge -dot- com -dot- nospam          781.359.4902 [voice]
Home Sweet Homepage: http://members.aol.com/SteveFJong