An alternative to the hackathon

by Nick Charney

---

By now most are familiar with the concept of an open data hackathon: jam people and data together into a small space for a short time and ask them to write applications, liberate data, create visualizations and/or publish analyses (often by providing a monetary reward as an incentive). This has been the model for a long time now and at the height of its hype its merits were debated extensively online. I'm not really interested in re-hashing the debate, but if you want some context you can go read the following from a few years back (and everything those pieces link to):

• Open government hackathons matter by Mark Headd
• Everyone jumped on the app contest bandwagon. Now what? by Alex Howard
• On Hackathons and Solutionism by David Sasaki
• To App Contest or Not App Contest by David Eaves

What I am interested in is fleshing out why we need an alternative approach, how that approach could be structured and what it could mean to the big picture.

Hackathons produce apps, not solutions to wicked problems

A bus tracking app is great at helping me leave the office just at the right minute to catch my bus, but it's completely ineffective at solving the gridlock that hamstrings the bus as soon as I board it.

A health tracking app may help me eat better and exercise more but does little to address the looming healthcare crisis in this country (See: Impossible Conversations: A Review of Chronic Condition by Jeffrey Simpson)

In short, apps address downstream problems facing individuals that are generally easy to deal with. Ought we not be digging in a little deeper, thinking upstream, and investing our effort into solving problems at the macro rather than treating their symptoms at the micro?

Refocusing on wicked problems

Maybe we ought to engage our best and brightest from academic institutions, the public and private sectors and create teams of cross functional experts to gather, crunch and analyse the data.

Maybe we shouldn't ask them to build a minimum viable product in a weekend but rather give them the time required for them to conduct a thoughtful analysis of the status quo, a desired future state, and options on how to get us there (See: Enduring Ideas: The Three Horizons of Growth from McKinsey).

Maybe then we can sit down, have the hard conversations, weigh the evidence and options and chart a coherent way forward.

The wicked impact of solving (or even mitigating) wicked problems

I have no evidence to support this claim so feel free to take me task on it, but I get the sense that solving (or even mitigating) a wicked problem in society will always yield better returns to society at the macro than any app ever could.

I could go down a long tangent on how I think the focus on apps (to the detriment of research) is a by-product of our shortened shadow of the future, our culture of immediacy and the omnipresence of attention economy, but I'll spare you the diatribe for the moment.

What I will say is that I think we need to spend less time scrolling up with our thumbs and more time leaning in with our shoulders.

The Value of People

by Kent Aitken

---

Editors have figured out that there's more going on in documents than people can adequately, reliably process all at once. The standard practice is to take multiple passes, each time looking through a different lens. I was taught to take three cracks at important documents:

1. one for logic, coherence, and ideas
2. one for conciseness and the removal of extra words that add no meaning
3. one for spelling, grammar, and punctuation

The field of business administration teaches something similar, providing frameworks and guidance for how to think about problems: SWOT analyses, risk matrices, etc. My first manager looked at every new problem through the below framework, thinking about the needs and constraints from each lens:

The decision-making lens I tend to advocate for is the more humanistic one: starting from a position of trust, looking for positive-sum outcomes, and appreciating soft and long-term benefits from actions. But in actuality, there is a different lens I would make mandatory for every decision, ever. It would be taking a moment to go completely cold, rational, and economical, and to think of the people within organizations as dollar figures. Just for a moment.


If You Like It Then You Shoulda Put a Value On It

Statscan put yearly salaries for professional/knowledge worker jobs around $62,000 in 2014. Between benefits and pensions I think I can conservatively put the total compensation cost at $80,000 as a working estimate*. 

Some examples in which this lens may be enlightening:

• Consider a productivity-enhancing piece of technology with a price tag of $400. If you get a 0.5% productivity increase, it pays for itself in a year. If it lasts for a few years, you can get away with a 0.16% productivity boost, or, a few minutes per week. To say nothing of the fact that an open mind towards such investments may keep employees around longer, when losing such an employee costs $16,000 in lost productivity (which is the lowest figure I've encountered for turnover costs).

• If you're responsible for convening a working team or meeting, try occasionally working out the labour hours involved. Ten people for two hours is half a week's work: at $80,000/year, about $750. This isn't to say you shouldn't do it; rather, that you should consider what level of planning and communication on your part respects that investment.

• I've used this example before, but the IRS provoked a scandal for $50M in conference spending between 2010 and 2012. But divided over the three years and approximately 100,000 staff, it's $166/employee. The real question is how much their employees are worth, and whether that spending made them at least a half a percent or so more effective. Triangulating a few different sources (average HR costs for companies relative to budget**, my compensation cost estimate), the low end of the range is likely ~$4,000,000,000 worth of annual investment in people. The IRS has a ~$11,800,000,000 budget, and $50M is a pittance of their people cost.

• Lastly, consider a manager with half a dozen staff. We rarely consider the idea that they're responsible for almost half a million dollars in annual investment. And that $500,000 machine's performance can easily vary by ~20%, depending on how it is managed.

One Lens

It is easy to view human resources as sunk costs: past expenditures that should have no bearing on present decision making. It is not so. We should be building decision frameworks that encourage us to consider our stewardship role for the people behind organizations - at least for a moment, as one lens of several. Then taking other passes over problems, with other lenses. Editors give documents this level of respect.

TL;DR: discussions about people and spending need denominators.

*This assumes that the average employee creates at least their compensation cost in value to the organization. In actuality, in any sustainably profitable business the average value created must be greater than their compensation cost, and so these examples would be shorting employee value.

**For the Canadian public service, it's 38.1%.