Long time listener, first time caller
I'm working on a community that will exist alongside an 'app store' for public services (each app is a Discourse plugin). It's called Pavilion. 'Public services' means legal and government services.
The platform, and the community discussion, will be categorized based on where people live (i.e. cities, towns, villages), as public services are location based. The platform design reflects this. You can see an (out of date) version of this platform on my old dev instance.
When I say 'working', I'm currently finishing off a few different Discourse plugins that will constitute the 'community platform'. Basically the platform is Discourse + about 20-odd plugins I've built (they're all open-source and currently available in my github (angusmcleod) and Pavilion's (PavilionDev) After finishing work on v0.1 of the platform, I'm going to spend a few months building up to launch.
I'm working on some custom structures for user-driven community (category) creation. The point of these structures is to try and ensure each new community on the platform has enough support to be viable before it launches, and to get 'buy in' from early adopters.
First, any user can start a 'petition' for a new community to be created. Petitions are started when a user successfully fills out a multi-page form (new plugin I recently made). The form ensures the petition is valid (e.g. non-duplicative, refers to a real place, etc). A successfully created petition turns into a topic in a 'petitions' category which has the Feature Voting Plugin enabled, allowing other users to vote on it. I'm currently considering what, if any, other criteria petitions for new communities should be determined by.
If a petition is successful, a new community (category) is created, however it is not yet 'live'. It has a single topic (which the category url automatically redirects to), where stakeholders can discuss the 'constituting' of the community. In order for a community to go live ('constitute'):
- a minimum number of users have to select it as their 'home' community. The minimum is set by a percentage of the population of the real community (e.g. 0.01%).
- the first moderator needs to be elected (via a poll in the constituting topic). I say 'first' as I'm thinking moderators will have elected terms.
- the community needs at least one 'app' (plugin implementing an online public service for that community)
- the community needs a designated 'builder' (someone technical).
Once a community is 'constituted' it finally goes live, allowing users to post topics as they normally would in a discourse category (albeit posting is significantly different, given the Discovery compose experience which involves 'topic type' selection.
So basically I've turned 'community creation' into a user-driven two step process: petitioning and constituting. One of my biggest concerns about this process is that it could be too long and people will disengage. I'm currently thinking a fair bit about what kind of engagement there should be with a user prior to the constituting of their 'home' community.
The process for launching the platform itself will involve a fair bit of content marketing and build up, i.e. the CHIP process. I'm working on a few essays / blog posts about government, law and technology.
I'm checking in here now as my thoughts are turning more and more to 'how to build communities', and you folks are the experts!