I’m working with a global open community that is just starting to define their overall governance framework, and the question we are wrestling with is how to seat a governing body (e.g. board or council) with “equitable” representation from the community, where we are defining equity primarily around gender, geography and/or social minoriites.
My question for you: might any of you be able to point me in the direction of any work done on constraint models for elections that help avoid the “all the best-known loudest guys get elected” class of pathologies?
I have looked at CommunityRules and done a bunch of searching, but haven’t found anything that really seems to get at this issue, but I can’t imagine solid work hasn’t been done.
A baseline example of the constraint models I’m seeking is an election/selection model designed to seat representatives equitably across e.g a gender spectrum, i.e, a balance of female/male/non-conforming representatives.
I realize there are many kinds of challenges around any model that seeks to modulate identity traits, but before we give up, I very much want to know what has been tried, successfully or otherwise
Thanks in advance if you can share any links or pointers.
Hi @gunner, I think the AMP Advisory Committee has a good process for this. Unfortunately we don’t have it written down somewhere (yet - a blog post about it is on the todo list) but I can briefly describe, and you can reach out to me or Tobie Langel if you want to know more.
In the case of the AMP AC, the AC are the folks who elect new members - all the electoral persons are known. Once the call for nominations are closed, AC members review the nominations and rank them in order of preference. Each person’s ranking is turned into a ballot, which is processed through multiple methods of STV and Counting. All the results (from the individual ranking, STV and Counting Methods) are put into a spreadsheet to visualize which candidates were most consistently successful, and which candidates were most consistently divisive.
We use this data to reach a consensus decision on the election. It makes it really clear which candidates are obvious yeses, and allows us to consider variables like geography, nationality, gender, and other under-represented groups when things are less clear.
This process doesn’t solve the problem if you don’t have a reasonable number of diverse choices in your call for nominees, but it can help highlight and mitigate issues within the election process.
These Inclusive Leadership principles (originally from Mozilla work, and then put through CHAOSS metrics release process), might be helpful when thinking about building structures that can lead to the type of outcomes you’re asking about. I like these, because the focus on building an inclusive experience for building new leadership, rather than just focusing on numbers. It’s probably now more than you asked for - but I’ve also written about applying metrics for inclusive leadership here.
CHAOSS has a placeholder for board diversity, but we haven’t advanced that particularly well yet…
Also, not leadership-specific, but to get diverse perspectives,I have tried a nomination process - specifically asking those loudest-guys you mention, to nominate people from diverse backgrounds for research. It was reasonably successful, although still gate-keepy to those within circles of influence.
I think Drupal has done some experimenting around this as well, if you haven’t got that on your radar.
Anyway, a topic I am interested in - looking forward to learning along with you!
Thanks so much for all these insights, they are much appreciated, as is the link.
We are trying to follow a similar set of sensibilities that enable everyone in the community to have a vote, which is where we are still trying to strike checks and balances.
Will definitely share out where we land. Thanks for taking time to support the process.
