Since I took exception with the construction industry on my previous posting, I thought it would be fair to take a stab at my second favorite counterpoint to software quality problems: the auto industry.
To make the point, I resorted to the news of October 27th, where Ford Motor Company announced the end of the line for its (once) beloved Ford Taurus. After almost 7 million units sold, the automaker called it quits on one of the least known cases of success turned failure in modern industry.
The Taurus history begins at the end of 1985, with a rounded profile that blasted the boxy gaz-guzzling competition. After six years in production and facing mounting competition from increasingly better Japanese cars, the new cosmetic modifications to the 1992 model fueled a 5-year streak as the best selling car in USA.
And then tragedy. On the trail of that enormous success, Ford introduced the 2nd generation of the car, on 1996. In the less endearing terms used by Ford's engineering teams, it became known as "the car that could not be built". Its impossible combination of curved oval shapes colluded with the free falling quality control standards in the US auto industry to deliver the most severe blow to the company fortunes in decades.
The complicated assembly process forced the prices to go up and the body panels to go sideways (in opposite directions). On 1997, the car lost the top honors as the nations' best seller and forced Ford into the death spiral of pushing the cars at a loss. The idea was to keep the assembly plants running and bet on the sales of large trucks and SUVs. Meanwhile, its Japanese competitors worked overtime to meet the demand for their own best sellers. The sales chart hides the loss of favor amongst US consumers, and incorporates fleet-rental sales that accounted for the bulk of the Taurus sales towards the end of its life.
What happened during that 11 year span? Essentially Ford executives missed a market transition to where customers had overtaken product control from the companies. They decided to innovate from a shell, transplanting its oval emblem to everything in its best-selling model, from door handles to dashboard shapes. Deciding for the customers had worked in the past and lifted the company from a financial tail spin on 1986. Ten years later, the same recipe (go rounder than the competition) resumed the tail spin.
You can read more in this insightful book by Mary Walton: "Car: A Drama of the American Workplace".
Oh yes, what does it have to do with technology companies?
I think we all know. Of course, 20 years is an eternity for a technology product, but specialists in the industry all agree with the combination of increased complexity and lost quality as the source of Ford's impending demise. Insult added to injury, the Taurus '96s design was widely perceived as ugly by the press and the market. That ill-received design was somewhat rectified 4 years later, but its competitors countered it with even better all-new models packing engineering years ahead of what the cash-strapped Ford engineering teams could muster.
Know thy wall, mind your trenches. geekish alert. A few years back, a colleague introduced the notion of DevOps to an internal large au...
"Underlying most arguments against the free market is a lack of belief in freedom itself." Milton Friedman A few months ago...
I have been meaning to document a bit more my experience as former Bluemix SRE and Operations Engineering Lead during 2014 and 2015, which ...