Engineered With Care

As mentioned briefly in an SSIT Blog entry (Technologists who give a damn) I've been using "Zen and the Art of Motorcycle Maintenance" in classes, both while teaching IT at SNHU, and more recently with life long learning programs. The book addresses many concepts, some mundane and some esoteric. A key reason for using it at the University is the focus the book has on Quality -- the need for mindful care in your work (whatever your work, but the author uses motorcycles as his example and metaphor.) A second, related theme is how to approach debugging a problem, fixing things, repair activity in a methodical way -- something that is not well covered (IMHO) in typical engineering/IT programs. But for this example I want to focus on the need for "Care", an example from common experience, and the factors that are working against Care-filled Technology.

Why Care?

There is a famous mis-quote "If you build a better mouse trap the world will beat a path to your door." While the "Mouse trap" is the obvious mis-quote, the real message is also missing. Ralph Waldo Emerson who is the source misquoted, created an entry in his journal on "common fame" that is the catalyst for this quote ... but that entry addresses your reputation that results from a quality product or service. Technologists who craft well made products develop a reputation, and that should be an aspect of personal satisfaction as well as career opportunity (although as I point out below, that may not be the case.)

So reason #1 is self satisfaction -- do you really want to think of yourself as creating second rate work?

#2 Well crafted/engineered products are a delight to use. We have so few examples that it is actually difficult in our technology dominated world to point these out. While I'm not sure it fully qualifies, the original Apple MacIntosh computer may be close. It was design

ed to be simple to assemble (color coordinated cables/connections), and simple to use. It looked nice and had very little spurious stuff. It comes close to the concept of "elegant" which occurs at times in the engineering world when something is really done right. It is not surprising that the inside of the cabinet contains the signatures of the design team that created it as part of the molded plastic. I compare this with recent personal computers I have where it is impossible to distinguish between the headphone and microphone jack, and an hour or two is needed to clean up the pre-installed ad-ware that comes on the machine.

#2.5 Poorly crafted products produce frustration and ill-will towards technology. This may be the most important social consideration. Technology is pervasive -- and people are not delighted. The irritation created by the ill-ease of use, lack of reliability, non-robust, non-intuitive and just plain shoddy workmanship are all too real. This drives away customers/users. This drives away students and prospective STEM workers and creators. The person-hours lost due to poorly designed/engineered products amounts to billions of dollars of lost productivity. Driving away creative, innovative young people because they see the dysfunctional and often ugly aspects of technology is perhaps a greater economic loss. For communities that distrust or reject modernization -- this becomes a repeating demonstration of the degeneracy of modern culture.

#3 Poorly engineered products can kill. Every product recall or liability suit reflects the tip-of-the-iceberg in negative impacts. While exploding gas tanks and x-ray overdoses were a reference in the last century, today we face the risk of 3rd party hijacking of cars or even commercial airplanes as a result of poorly engineered systems where fail-safe, and malicious abuse were not part of the design considerations.

A Simple Example:

Millions of web sites ask for your address. For those of us living in the U.S. in states starting with the letter "N" for example, often find that instead of entering two letter codes we have memorized, we are forced to use pull down menus, scrolling past other states to just get this in the field. How many key strokes should it take? NONE! Observe the way the post office sorts mail ... they do not start with the street address, they start with the country then the postal code. There is the real trick -- ask for country and postal code then you can display back for the user the city and state/province ... and with sufficient digits, even the street address. If the city is wrong, the user will immediately correct the postal code -- quality begets quality ... and of course it is the postal code that is most critical in any actual use of the address. This idea is not new, it was captured in IEEE Std. 2001 (approved 2002) on well engineered web sites which is also ISO standard 23026 (approved 2006) although that standard has not been widely used, and the emerging update may no longer include this guidance.

There are hundreds of millions of web sites (if Google's count is any indication) that ask for this information. Most if not all do it wrong. This frustrates users, and discourages them from the commercial engagement most of these sites seek, and from use of the internet, computer and technology on a broader basis. Saving a few seconds for each transaction at each of these web sites is a potential productivity benefit for both users and web entities that approaches the billion dollar amount I attribute to poorly engineered products.

[I was disrupted from entering this commentary for about a half hour trying to get to an online IEEE webinar using WebEx technology working, but was unable to get connected following all of the information given ... and I had just spent an hour a few weeks ago solving the problems needed to get it working -- in theory. And I'm only a computer expert, so what can we expect for the general public? The feeling of frustration is something that lingers well after the events have passed -- say duck, can I use that hammer when you are done with it?]

And why we won't fix this problem:

First - too many technologists/engineers don't care. "It occurred to me there is no manual that deals with the real business of motorcycle maintenance, the most important aspect of all. Caring about what you are doing is considered either unimportant or taken for granted." (Zen and the Art of Motorcycle Maintenance, p27)

Second - Employers do not care, and do not provide the time needed to do the job right. Delivery trumps quality too much of the time. I understand one major software company re-builds a major product regularly, and the QA static analysis shows a half billion detected errors. Most of these are "history" -- they are there alright, but the folks that know those portions of the code are no longer there, and until it triggers some major problem, these will be carried forward as long as the product survives.

Third - We aren't expected to do it right. Products are created against some set of requirements. Few call for comprehensive consideration of potential failures, security flaws, etc. Most, at best, require fixing the problems identified in the last liability suits related to similar products. Even when the product specs are clear, "operational down to 53 degrees", management and users will violate this requirement and launch anyway (Challenger, 1986) While it is clear that there will always be unexpected situations -- where possible products should fail safe ... and even this is not likely to be a stated requirement. A 2004 air traffic control failure was apparently a result of software that did not handle a specific error condition properly, a condition that would surface about every month of continuous operation.

Fourth - We are paid to do it with lower quality. We have a 1972 Maytag washing machine -- it still works fine thankyou. My 2004 Prius has 250,000 miles on it -- seems to be running strong, still getting 50 MPG on a good day. We retired a Windows XP system that worked, well as good as any Windows system, because updates (ergo security fixes) would no longer be available -- not because the hardware stopped working. You can't sell replacements if the originals keep working. Gross National Product, Leading Economic Indicators (like new housing starts), Corporate Sales Growth, and probably your salary, are based on selling products to folks who already have one -- poor Maytag hasn't gotten our money again for almost fifty years ... that is not what stockholders want to hear.

What what are crafted products like?

They probably don't have a disclaimer in their warranty statement (do you ever really read these things, much less know what they mean?) that says "No warranty to buyer from seller is express or implied. Seller specifically disclaims the implied warranties of merchantability and fitness for a particular purpose." -- Because it is quite fit for a particular purpose, and has a reputation for being excellent for that purpose.

They probably last "forever" ... and in many cases will be modular so that they can easily be repaired or can be upgraded. It may be that we move towards a "rental" model for quality products, providing an income stream for the supplier, and continuous operation for the buyer.

Consumers will be delighted. They will spread the word to others, and paths will be beaten to the suppliers doors. They will be aware of the total cost of ownership, and abjure conspicuous consumption.

Employees are going to be strongly dedicated to their job and employer. This runs counter of the apparent trend towards a "non-employee workforce" where everyone is free lance. They are highly motivated since they are given the opportunity for real mastery of their trades, and associated with a top quality result -- even a stronger motivation if the product/service is of clear benefit so they know they are doing "good" while producing "goods". Employee owned companies are perhaps a more likely candidate for this future than non-associated stockholders.

Our concepts of "full time employment" may need to change. Companies will probably need less people hours to produce quality products that last significantly longer. If the company didn't "care" they would cut the workforce, but then they wouldn't be playing in this game in the first place, so cutting hours while retaining reasonable salaries and benefits is a possibility.

Our metrics of "success" for corporations, and Economic Indicators will need to change. We can expect fewer "housing starts" (since houses will last a long time, this will be tied to birth rates and immigration.) Corporations are likely to find a stable revenue rate ... or grow by developing truly new products.

We will see at least a gender balanced influx of students and employees in most fields, with young folks knowing they can contribute to making good things, rather than strictly making money.

Less energy and other resources will be consumed since less items are being produced. With more continuing use, there will be less need for re-use, recycle and repurposing -- and less land fills. This may be a key aspect of sustainability.

What will drive corporations to build new and better? Will there be competition? Will capitalism survive?

Adam Smith was not an engineer. Ron Avitzur is one -- I suggest you read his story of creating the Apple Graphing Calculator. The short version is this: Ron and associate Greg snuck into Apple after being laid off to create a new product that was the "killer ap" for a new Apple computer (in 1994). No pay, no legal relationship -- why? "Mostly, Greg and I felt that creating quality educational software was a public service. We were doing it to help kids learn math. ...We were hackers, creating something for the sheer joy of making it work." This is the model of the Open Source community among other examples as well. If the folks with this dedication to creating quality products have the resources to live, the opportunity to create, and a supportive community then maybe Care-Filled Technology will catch on.