No False Users

I’m always struck in technical meetings how quickly people dream up imaginary people. People with very specific needs that they didn’t know they had. A recent meeting I was in suggested that if streetlights and hospital shift patterns were connected to the Internet of Things, we could potentially make sure that nurses can get home safely at given times, by increasing lighting at the end of shifts. Or that by comparing bus times with air pollution data, we can start to think about where buses are idling and reduce respiratory disease. But of course, they’re just possibilities! We don’t know yet! Think of the potential!

And sure. All those things are possible. But they’re fantasies. And it’s OK to start with a fantasy – decades of science fiction have guided science and engineering. Everything starts with an idea, at some level of application. But those ideas rapidly get blown wildly out of proportion. The problem is that by creating these stories and allowing them to persist, they get repeated ad nauseum as post hoc, ego propter hoc justifications.

User stories¹ are meant to be non-fiction. We should not be in the business of giving any more airtime to fictional user stories than we need to, given how easy it is to gather them. The cart should not lead the horse. I’m sure that if you asked medical staff their top 20 desires, the lighting on the way home wouldn’t even factor, and that streetlights are part of a carefully orchestrated city engineering process. And I’m sure that if one really wanted to reduce air pollution, having a networked grid of air quality sensors would give useful information, but do absolutely nothing to tackle the problem of air pollution in cities. And in the vacuum of applications for these ideas, I suspect these “straw users” will have already been referred to half a dozen times as hypothetical benefits².

Adam Smith famously described in The Wealth of Nations the obvious progression of how humans moved from a barter system, to coinage, to a bookkeeping system. This version of economics is widely accepted as the obvious - if not inevitable - backdrop to modern society. Except, he totally made it up. There is no anthropological evidence of a society where barter existed before other forms of currency, anywhere in the world. David Graeber (2011)³ explains:

For centuries now, explorers have been trying to find this fabled land of barter - none with success. Adam Smith set his story in aboriginal North America (others preferred Africa or the Pacific). In Smith’s time, at least it could be said that reliable information on Native American economic systems was unavailable in Scottish libraries. But by mid-century, Lewis Henry Morgan’s descriptions of the Six Nations of the Iroquois, among others, were widely published - and they made clear that the main economic institution among the Iroquois nations were longhouses where most goods were stockpiled and then allocated by women’s councils, and no one ever traded arrowheads for slabs of meat. Economists simply ignored this information.

Stanley Jevons, for example, who in 1871 wrote what has come to be considered the classic book on the origins of money, took his examples straight from Smith, with Indians swapping venison for elk and beaver hides, and made no use of actual descriptions of Indian life that made it clear that Smith had simply made this up. Around that same time, missionaries, adventurers, and colonial administrators were fanning out across the world, many bringing copies of Smith’s book with them, expecting to find the land of barter. None ever did. They discovered an almost endless variety of economic systems. But to this day, no one has been able to locate a part of the world where the ordinary mode of economic transaction between neighbors takes the form of “I’ll give you twenty chickens for that cow”.

There’s a moral somewhere in here about the power of persuasive storytelling. John le Carré comments in many interviews that it is his job to make characters believable, not truthful. And much like a good piece of misdirection from a spy, Smith’s fairy tales about fictional civilisations have made us believe something fundamental about human behaviour that isn’t true. The great revelation here of course is that fundamentally people share, and support each other: not something very palatable to colonial Britain’s Whiggish⁴ history.

Clearly, something about Smith’s tale was so believable and so persuasive that it has fundamentally changed the way we think about money. I’m not suggesting that anyone is doing this by making stories about products - but I do think that the stories dreamt up on the spot like this have a habit of sprouting wings and taking flight. And we should be extremely careful to not release our personal fictions masquerading as technical specifications into the world.

There’s an underlying, unspoken assumption with technology projects that “if you build it, they will come”, much like the Whigs’ belief that we simply march forwards towards greater enlightenment. As Maslow famously remarked: when you’re holding a hammer, everything starts to look like a nail. But we must be careful to not allow these ideas to persist without testing them straight away. There’s a lot of power in stories, and we shouldn’t be using them to justify the enormous expense and time commitment that most technology projects command. The irony is we live in a world with unprecedented potential for gathering data: asking a few nurses what they think about it would take minutes on something like Twitter or Facebook. And by doing so we can put the cart back behind the horse, and make technology solve people’s problems, rather than inventing problems to justify technology.

Making solutions to problems no-one has is a waste of everyone’s time and our planet’s dwindling resources. Innovation shouldn’t mean disengaging from society and has no built-in moral “goodness” - unchecked, it simply will replicate and support the injustice and inequality already in the world. Imaginary scenarios are a fine place to start, but user stories should be non-fiction, and we need to be careful to separate the two.