As a computer consultant I get to walk into and clean up a lot of messes. Sometimes these are messes created by months of misunderstanding or simple neglect: logical misarrangement or the organic detritus of any natural process over time. Sometimes these messes are caused by a single catastrophic mistake on the part of the operator, what in the airline industry would be called “controlled flight into terrain.” But sometimes these messes are both systematic and willful. The stories behind these methods usually begin with a single sentence: “I have this person I know who knows all about computers.”
Every job has its stereotypes and hazards. Kindergarten teachers often become preciously pedantic. Military men tend to be gruff. Young lawyers tend to be speculatively argumentative. The novice mistake that computer people make is to be in love with their own systems, and assuming that their favorite systems are the right systems for all people at all times. The nephew who flies in and sets up a Windows computer for grandmother is not grandmother, and almost never makes the icons large enough or the screen contrast sharp and bright enough for grandma to see. The person with clear ideas about workflows will rarely take the time to understand the workflow of another, especially a single other. Like the ancient sculptor Pygmalion, most engineers are so love in with their systems that they ignore people, and particularly the feelings and needs of “civilians.”
There is a great temptation in our silver-bullet society to imagine that a technique or corporate training, some late-night infomercial gizmo or gadget, can magically solve our problems and somehow save our lives. This is not the way that people or the real world works, and three of my favorite computer-consulting stories center around a basic understanding of that fact.
The three best technical consultations I ever did all ended with the suggestion that people not use a computer. The first was a charismatic older man who sold hand-made sporting equipment to fly-fishermen from a semi-rural location. After purchasing a brand-new iMac and setting up with an old printer, one of the first things he wanted to do was to laser print envelopes. For various reasons related to the program and particular model of printer he was using, each envelope would end up taking at least two minutes, so I suggested that he appreciate his unique niche and instead hand-address each envelope in his charming, slightly palsied, older-man hand. We then went on to other things which the computer could do faster, and his outgoing mail continued to reflect his distinctive, unique market niche.
The second was a woman who ran a small shop selling craft supplies. She had fewer than a thousand items in her entire store, and a good day for her was a gross of $300 or more. She contacted me about the idea of configuring a point-of-sale computerized cash register with bar-code scanner and inventory control. For a shop of less than 1,000 square feet, with one owner and one employee, with gross sales of less than $2000 per month. I suggested instead that she save the $1500 or so plus training time it would take to insert such a system, and showed her a simpler card-based system that helped her and her one employee to track inventory more slowly but with much more care: the sort of system that was developed in the 1920’s, and which has served thousands of small shopkeepers. She looked at me as if I had suggested she cut off and eat her right arm, found someone else who sold her a system, and was out of business within two years.
The third was a sales and service organization who contacted me to spec out a computer database and network server. Meeting with them I saw that they already had a good system, and asked for more details about what did not work with the current system. After a time it came out that the central issue was not the lack of a networked computer database but a consistent inability to track and deliver phone messages. I suggested that they invest in a few stacks of duplicate phone message pads (which use NCR paper to make two copies of each message) and place one at each phone. With a simple clip on each person’s physical inbox, they established clear rules for taking and delivering phone messages well, each and every time. Although I went on to help them with other aspects of their network, these $5 spiral pads and one clean procedure probably saved them over $1500-$3000 in software development and staff training costs. I went on to help them with other, more appropriate problems, and to this day I am proud that I never wrote a spec for that project.
The best breakfast is the one you’ll eat, not the one that some lab tech or a nutritionist in a hospital somewhere has calculated is perfect. People are different and they understand the things they understand. A person’s comfort zone is where they are most comfortable, and the good engineer or manager’s skill is being able to understand where a person is, honor that and gradually train them toward something that better serves them or meets their need.
History is littered with Maginot Lines, sprinkled with Titanics and punctuated by Hindenburgs. One can design all the most beautiful castles or systems or skyscrapers in the world, but they are only so much sculpture if people are not happy to live in them. “There are no systems so perfect,” Ghandi supposedly said, “that people do not need to be good.” An over-engineered system that does not accomodate or properly train people is only so much performance art.