As a postdoc at the Rotman School of Management I have focused my time and energy in the last year on studying some aspects of the Mozilla community. In particular, I am interested in how open source communities develop organizational capabilities that lead to high-quality software. One of my Mozilla sub projects has been my focus on factors that determine how much time is needed to fix a bug. For example, I wrote a Jetpack add-on that will predict the likelihood of a bugfix based on the initial bugreport. Now, I would like to focus more on the content of the initial bug report and in particular on the use of negative and / or strong emotional words. I would like to identify these bugreports to see how the Mozilla community deals with such bug reports and to study the possible negative consequences such as wasted time or the loss of a potential Mozilla contributor.
I wrote the Crowdsourcing Flame Firefox Bugreports Jetpack Addon which you can use to rate bugreports on the “flamyness”. I will collect the data and use it for my research but I will also give feedback to the Mozilla community based on these data. All data is completely anonymous, I cannot identify the person who rates a bugreport.
My goal is to identify words, phrases or expressions that when absent (present) will decrease (increase) the likelihood of a bug fix. This could even result to an automated assessment of an initial bugreport (in Bugzilla) telling the bugreporter to add (remove) words from the bugreport before the report can be filed.
How to use the addon? First, just install the addon and then start browsing Mozilla’s Bugzilla website. You can rate every bugreport once, but different people can rate the same bugreport. If everything worked, then you should see something like this (when visiting Mozilla’s Bugzilla site):
Screenshot of Crowdsourcing Flamy Firefox Bugreports
If you have tips to improve the code, or feature requests or other suggestions….please let me know!
Recent Comments