Hi Stefano, Thanks a lot for the feedback. I do agree with you that the so-called "participation ratio" is a little bit misleading. I will consider your suggestion when doing the CY13-Q2 report. From the OpenNebula mailing list I got similar feedbacks, the reason being that the OpenNebula folks can always help the user resolve a problem with less email messages. Yes a neutral name for the parameter would be much better. Also I do agree that this parameter should not be related to the "activeness" of the community members. I will remove such judgements from the CY13-Q2 report.
I have not investigate the reasons for the ratio changes for OpenStack and CloudStack in their early days. This seems to be an interesting question, and I will try to answer that in the CY13-Q2 report. I saw Jay's suggestion on removing review.openstack.org from the git domain analysis. Can you shed some light on how this system works? Is this system shadowing more real code contributors? John 在 2013-4-3，上午3:25，Stefano Maffulli <stef...@openstack.org> 写道： > Thanks John, it's always interesting to see comparisons among different > projects. Although comparisons are hard to make :) > > I was looking at the "Participating Ratio", the ratio between “the number of > posts” and “the number of topics” that in your post you assume it represents > the participation rate of an online community. > > In your post you argue that "the number of replies to a specific topic > represents the attention being received, and the depth of discussion for that > particular topic" basically assuming that a high ratio is always better. I > disagree with this assumption. > > My explanation for decreasing depth of discussions is that the topics > discussed are less controversial and require less amount of messages to > close. For OpenStack, patches and changes to code are not discussed on the > mailing lists anymore but they are debated (sometimes heavily) on > review.openstack.org instead (which your tools don't capture) while afaik > cloudstack and other projects seem to review patches on the -dev mailing > lists. Comparisons across these projects are hard indeed. > > You highlight OpenStack's decreasing 'participating ratio' leaving in the > reader the impression that this must be bad but I don't think that's the > right conclusion (nor the only one). > > I would suggest you to rename "participating ratio" into > "participating-bickering ratio": slightlgy longer but more neutral; lowering > "participating-bickering ratio" is neither good nor bad and it forces reader > to go investigate the causes (and you avoid being mis-quoted by some > blogger/pundit that doesn't have time to dive deeper in the causes). > > Have you investigated and found other reasons for why the > participating-bickering ratio changes so much for OpenStack (especially in > the early days) and Cloudstack? > > Cheers, > stef > > PS just a heads-up: we've added https://ask.openstack.org in March to our > properties, so next quarter please add it as a source to your dataset > > > On 04/02/2013 02:32 AM, Qingye Jiang (John) wrote: >> Hi all, >> >> I am glad to present to you the 6 edition of my quarterly analysis on >> this subject. CY13-Q1 Community Analysis — OpenStack vs OpenNebula vs >> Eucalyptus vs CloudStack is now available for your reading at the >> following URL: >> >> CY13-Q1 Community Analysis — OpenStack vs OpenNebula vs Eucalyptus vs >> CloudStack <http://www.qyjohn.net/?p=3120> >> >> In this report I have added some preliminary analysis on the github >> activities of these 4 projects. >> >> Best regards, >> >> Qingye Jiang (John) >> >> >> _______________________________________________ >> Mailing list: https://launchpad.net/~openstack >> Post to : openstack@lists.launchpad.net >> Unsubscribe : https://launchpad.net/~openstack >> More help : https://help.launchpad.net/ListHelp >> _______________________________________________ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
