I heard an interesting report today on the CBC describing a study that found when gas pumps go out of calibration they tend to benefit the retailer. According to the report, every year Canadians pay twenty million dollars for gas that they do not receive because of small pump discrepancies.
This is a great example of how values are (inadvertantly) designed into our machines and devices. Gas pumps, we can say, are biased to favour retailers when they make mistakes. The corollary is that gas pumps tend to discriminate against consumers.
I suspect that engineers who designed the pumps would not describe the bias as problematic, and would not even talk in terms of values to begin with. They would likely say something like “The pump isn’t biased, it’s just out of calibration. Retailers need to calibrate their pumps more often.”
Well, that’s one way of assessing the situation. But it ignores the social context in which the pumps are situated, and the social effects that the pumps engender in virtue of their tendency to go out of calibration in such a way that favours the retailers to the tune of twenty million dollars a year.
This could be otherwise. By adopting a values based design approach (one I am in favour of) the design of the pumps could be approached in such a way that explicitly evaluates the social effects that an uncalibrated pump will have, and allows design decisions to be made that are explicitly based, in part, on the values that we want to have designed into the pumps.
For example, the current state of affairs is such that the pumps go out of calibration in such a way that tends to favour the retailers. Customers think they have purchased a litre of gasoline, when in fact they have received slightly less. This means that the consumer is left carrying the cost of uncalibrated pumps rather than the retailer.
However, it is the retailer who is responsible for regular calibrations and who is offering the gas at the particular price per litre. The fact that uncalibrated pumps favour the retailer creates a disincentive on the part of retailers: why worry too much about calibration if it isn’t costing them anything? Retailers will likely not adopt a process that increases the reliability of the pumps because they have little at stake, and the cost of calibration is not worth it. Consumers are left having to lobby (or sue) the retailers for changes to the way pumps are calibrated, a long and costly process borne asymmetrically by the consumer.
There is another way of preempting the problem, which involves making a slightly different design decision with regard to gas pumps. If pumps were designed in such a way that an uncalibrated pump tended to favour the consumer, that is in such a way that consumers filling up at an uncalibrated pump got slightly more gas than they paid for, most of the problems would be solved. The retailers would have an added incentive to calibrate pumps regularly, or face losing money. Consumers would also have reason to trust the retailers, knowing that there is less of a chance that they will be ripped off even if pumps go uncalibrated. Most importantly, the costs of uncalibrated pumps would land were they ought to land, in the laps of retailers.
Values based design offers a way to evaluate design decisions in a way that matches our social expectations. In this case a values based design approach could result in a better gas pump.