Steve O’Grady and team from analyst firm Redmonk put on a very different day-long event (Monktoberfest, the “Developer Conference about Social, Tech and Beer”), at a very different venue (the Portland, Maine Library), with a very different lunch and dinner (lobster, beer, duck, beer and beer), a non-typical list of topics, delivered with an unusual balance of tech talk and beer. And while this event was delightfully unusual, perhaps most unusual were the compelling number of “ah hah” moments. Something in the Maine water? You can read about the conference and its “reason for being” by reading Steve’s blog.
“I can lead you to beer, but can’t make you think.”
This was my first random thought of the morning. And I surrendered to thinking with each and every speaker. They were all thought-provoking, and all for different reasons. For example, Matt LeMay got me thinking when he said: “Who you are is probably more interesting than who you think you should be. Being a person is smarter than building your personal brand.”
Given the theme of “social,” we discussed Agile (we look down on those who aren’t), considered Six Sigma and ITILv3 bad words, and with Donnie Berkholz’s talk, “Assholes are Killing Your Project,” we discussed just how ruinous a “not nice person” can be to a project, especially within open source communities where it’s difficult to rid projects of a “not nice person.”
Prior to this discussion, I was unaware of the metric “TCA,” or “total cost of an asshole.” The takeaway is that “soft skills” are under-valued when we think about the success of certain projects over others, as is the importance of soft skills in general in the broader open source community. Consider: what if one of your primary committers is hostile to other frequent committers? Or someone new, a genuinely interested committer, is affronted by someone identified as a key contributor? How do you monitor behavior? How can you share with the guilty party that their behavior is ruining the project? Will they change?
This, of course, got me thinking about Ohloh data (full disclosure: Black Duck manages Ohloh), which Donnie was nice enough to quote in his presentation. It’s all well and good to look at the number of committers and other metadata, and Donnie was the first to point to trends that may have been caused by the people behind these projects. If an otherwise exciting and growing project suddenly shows declining commits and committers, perhaps it’s time to evaluate the project from a “soft skills” standpoint. Who is among the core team, and what are their dynamics? Is metadata all “hard data” or should you go deeper? Because it’s becoming clearer that interpersonal dynamics are impacting the enablement of open source.
A closing thought from Zack Urlocker’s discussion, “Social and Distributed Development”: regardless of whether developing internal, commercial, or open source code, developers ought to publish it in a fashion that it could be open source. As opposed to a corporate entity, if a developer’s name is on the code, the concept of pride of authorship should not be underestimated.