A list of preliminary questions to be answered through the Edgesense testing on Edgeryders, slightly focusing on building last year’s event Lote4, timeline July - December 2014 (work on progress). Please add your questions and comments.
How subcommunities develop and grow in Edgeryders?
who are the central non-moderator members? (potential moderator role in the community)
how central members add specialization to conversations? what is the nature of the non-moderator interactions between sub communities? this includes the ratio and done by whom
what is the role of moderators in decreasing the modularity, across time?
(Noemi) My question would be: what, if any, is the influence of community members in the evolution of sub-communities? If you remove them from the graph, do sub-communities still exist?
- what is the sustainability of conversations in sub communities over time ?
(Noemi): Sustainability in terms of what? That conversations become more and more specialized?
(hazem) plus how much of it being faded away and how much new conversations started , in other words what’s the ratio of new edges being built to the old edges already exists?
this would be just a relative value between subcommunities as the edge value represents “unique relationships” as explained by alberto
// the timeline of posts and comments follows the same curve except for a small incident in 2014 where the posts by community members exceeded again the posts by moderators … so when was that exactly? ( probably while proposing sessions for lote4 )
going back to the Spot the Future project in spring 2014, what is the ratio of interaction of the futurespotters subcommunity with the network during and after the project?
excluding the moderators, what is the average degree ( in + out ) of the different nodes
Questions to answer during Masters of Networks event in March, comparing and computing data from more communities.
Can we predict from early behavior which users have a chance to become vary active community members aka “tribe elders”?
What converts a lurker into an active user?
Imagining a mechanism that causes the user to become active, and then building a test for the reality to conform to that model.
Is there a typical time window from user creation during which either the new user becomes active (for some value of “active” or is lost forever?
Example of explanation we’re looking for: “85% of users that do not write a comment within two weeks of creating their user never contribute anything”.
// (hazem) from 16 march till now there was 2 small increases in the no of nodes per users (could be the effect of different activities that requires more engagement from different users lote and futurespotters )
Community health and other questions
Can we maybe break down or build a model for a minimally healthy community? Are there reliable tests for a debate’s good health? Why is debate fruitful and creative in some contexts, sterile and conflictual in others?
NB: Active communities are not the same as healthy ones I think.
- Can we develop a metric as to the optimal proportion of content in a healthy community? Are healthier communities those with a greater share of contributors content than moderators content, or viceversa?
- Do people in more active communities talk to more people? Or do they talk more to the same average number of people?
- can edgesense be developed to predict different scenarios of how the community could grow ? ( trying to keep the same modularity , how many moderators are needed if one new average size sub community appeared …etc )