A quote from a longtime maemo.org member combined with a brief passage in the book I’m currently reading got me thinking lately about measuring social interactions and expectations.
The maemo.org member quote was in reference to my upcoming presentation on corporate-community engagement, and I’ll repost it:
What exactly does “success” mean for a community? That it’s still existing? That it’s growing? Is the fact that a community exists at all a success?
It’s hard to determine how the corporations are responsible for community success when that success is a lot harder to define than dollars and cents.
Hold that thought for a bit.
The book is “The Drunkard’s Walk: How Randomness Rules Our Lives” by Leonard Mlodinow and is a refreshing ramble mostly about probability and statistics– along with common mistakes we make by not applying them properly, if at all. The entire work is fascinating to an armchair analyst like me, but I want to focus on one particular portion.
Adolpe Quételet was a multifaceted man who counted mathematics and sociology among his many skills and interests. In recognizing that normal distributions apply to human physical characteristics, Quételet wondered if he could similarly quantify sociological elements. He had identified a mythical but prototypical person which he dubbed l’homme moyen (“the average man”), and decided he would dedicate his life to identifying a means of using this construct as a baseline for analyzing humankind. He was sure that, like aspects of physics, human impulses and engagements could somehow be enumerated, measured and catalogued.
This struck a nerve. Certain that there was a connection, I was instantly taken back to zerojay’s observation and decided to follow Quételet’s footsteps to see how far he had gotten and what he had concluded.
Mlodinow’s book does say Quételet was ultimately unsuccessful in developing social physics, but I wanted to know more than a few pages could cover. This excerpt from the Wikipedia article linked previously is a good start:
[Quételet] was keenly aware of the overwhelming complexity of social phenomena, and the many variables that needed measurement. His goal was to understand the statistical laws underlying such phenomena as crime rates, marriage rates or suicide rates. He wanted to explain the values of these variables by other social factors. These ideas were rather controversial among other scientists at the time who held that it contradicted a concept of freedom of choice.
Detractors and eventual disappointments aside, the statistician did demonstrate correlations between factors such as age and gender with certain crimes and pave the way for modern criminal profiling. What he found much more difficult was extrapolating the techniques out to more typical behaviors… and I’m sure he would have echoed the sentiments of zerojay.
However, zerojay’s conundrum could still use a resolution. My goal with the presentation is to identify ways to measure successful engagement efforts between for-profit corporations and open-source developers… a relationship that starts off contentious due to fundamental philosophical differences. Free or Open Source Software (FOSS) advocates can be distrustful of corporate motives, and many corporations stumble when trying to enter their world.
Anyone who has witnessed the dynamics between Nokia and maemo.org knows well that there exists a tremendous amount of energy there, some well-tapped and channeled but much of it still diffused four years after the internet tablets debuted. One could argue that a degree of success is shown in the recent merger between the developer site and the former internettablettalk.com, but then others claim it’s a capitulation to Nokia and thus a failure.
Nobody said this was going to be completely easy.
Still, I think some measures of success should be fairly straightforward. Maybe we can get consensus on the following… especially since some are already in place:
- Community-elected representatives
- Corporate support without corporate governance
- Steady corporate progress toward openness
- Corporate sponsorship of conference attendees without strings
- Clear majority member acceptance of community standards
- Clearly-defined information conduits
All of these are quantifiable although a few are admittedly soft. Still, we have to start somewhere. We could also set some expectation-based measures, such as how many new members join, how many new application packages are released, etc. Whatever is deemed appropriate.
Adolphe Quételet’s failure does not need to be ours. It may not be possible to craft some sort of social physics, but we can identify some expectations, analyze the dynamics and develop a plan for achieving and sustaining some clear goals.
If my suggestions aren’t acceptable, how about some from the community? Especially the “average member”. 😉