Hi All I’m sorry I haven’t been that active on this recently, this has been caused by a few things happening that meant that I needed to focus my time and effort elsewhere.
One of the things that my time has been focussed on is on assignment for my MBA where I have used Kibble as my research tool. To give you some background. My paper was focussed on the transmission of culture and values in open source and I wanted to create a baseline to be able to measure cultural indicators. So how can Kibble help? (I hear you ask :-), so let me explain a little. Kibble includes the following: - Pony Factor – which is an indicator of the diversity of key project contributors. So thinking of the Apache culture and its values, we would be looking to see the Pony Factor grow over time as a project community grows and accepts new conributors. There is also a meta Pony Factor which tries to measure the diversity of the companies contributing. - Sentient / Mood Analysis – which indicates the mood of the mailing list communications. - Key Phrase Extraction (KPE) – which pulls out important words or phrases that summarise the main topics or ideas that are being discussed on the mailing list. - Contributor Retention – this is divided into two parts; one is the length of time contributors have been in a community and two; a breakdown of active, retained, people who have left a community and also those that have returned after a breakdown * Methodology * What I’ve done is this: Apache culture was created as a result of the Apache Webserver project. So I used this Kibble data for this project to create a cultural baseline based on the above indicators. I then took two sets of Apache projects (one group that have been ASF Top Level projects for over 5 years and one group that have been Apache Top Level projects for less than 5 years) and measured their indicators in Kibble. I then compared them both to the Apache webserver cultural baseline. My results were interesting and the most significant thing I can mention is that the +1 indicator, which is something culturally unique to Apache as a consensus indicator came out as part of the KPE analysis is all 3 groups.(So some cultural transmission is happening!) It also seemed to indicate that the older projects were better at some of the Apache cultural aspects e.g the recognition of merit, where the younger projects were amazingly successful at community growth. I’ll load my paper maybe onto the wiki for people to look at (and probably critique :-) before I share it more widely within Apache. Thanks Sharan
