Quantitative community management

*
Accepted Session
Short Form
Intermediate
Scheduled: Wednesday, June 19, 2013 from 4:45 – 5:30pm in B201

Excerpt

In this talk, you will learn the state of the art in community measurement, common mistakes made in surveying, and how to actively use data to improve activity within a project.

Description

In recent years, communities as wide-ranging as Wikihow to Thunderbird have been surveying participants and using this information to improve the experiences of participants. A variety of open source projects are now tracking contributors to identify where people fall away, and to nudge them forward. In this talk, you will learn the state of the art in community measurement, common mistakes made in surveying, and how to actively use data to improve activity within a project.

This talk will cover the following issues in detail:

  • How Wikipedia used A/B testing to improve contribution rejection messages
  • Based on entrance/exit surveys from OpenHatch’s Open Source Comes to Campus program, what do new contributors know?
  • The impact of treating gender as a plain-text field, rather than a drop-down, on the answer rate
  • How Ubuntu’s Developer Advisory Team tracks, contacts, and nudges new contributors
  • How motivations for Thunderbird contributors differ substantially from the FLOSSpols survey
  • How to misread your survey data (and tips on avoiding that)

Upon leaving this talk, you will have a solid background in the current state of data collection within open source communities and how to apply those tools to your own project.

Speaking experience

I've presented at PyCon 2011 and PyCon 2012 about successful efforts to improve open source communities, including work I started in Debian to improve package review, and also on successful gender diversity outreach strategies for programming user groups. As discussed above, my interest in quantitative community management comes from my active participation in open source communities and my work in open source outreach.

To get a sense of my presentation style, I recommend these links:

* http://pyvideo.org/video/415/pycon-2011--get-new-contributors--and-diversity--

* http://pyvideo.org/video/719/diversity-in-practice-how-the-boston-python-user

I've also presented at OSCON 2008, OSCON 2010, OSCON 2012, a few US-based Linux and tech user groups.

Speaker