In order to make more users join a community, there must be actual content present
Casual users don't join some random community no matter how interested they are in the topic unless given an actual reason to. Reasons in this case include already posted high quality content or the ability to ask questions to domain experts on the topic. Or maybe just an impression that this seems to be a nice place to hang out.

This means that when we start a new community, it is not sufficient to have some 10 or so people saying "yeah this sounds fun". We need several committed enthusiasts who constantly push the site forward: defining scopes, starting meta discussions, posting actual content, remaining active over time.

Regarding luring users over from Stack Exchange specifically, it is not sufficient to simply offer better features or a better organization. Most people only care about the actual site content. It many cases they will expect content of equal or greater quality than that of the site they are currently using, so they will in many cases have very high standards.

Imported posts from SE has been a hot potato discussed on the various community metas. I think it is good that we have the option to import, but the Codidact sites cannot just rely on old imported content - there must be something new content unique to the particular community to make it interesting. You'll want casual users to read a post and think "This was a nice post - hey I better keep an eye on this particular community or I might be missing out some good stuff."

Some concrete suggestions for how to improve the situation:

• Put all the ghost towns in the freezer.
  The definition of a ghost town being a site where there are not enough enthusiast users that are continuously providing content for that community. And "the freezer" would be some less prominent place where embryos of working communities may exist until they can guarantee sufficient site activity. See Stack Exchange Area 51 for inspiration, it seems to be a model which works well.

  This may sound harsh but the network as whole is suffering from these ghost towns. I don't think they should have been launched in the first place. In many cases there was just one or two prominent users pushing for a certain community and when they stop pushing for the site, for whatever reason, it dies. Roughly half of our communities are such ghost towns.

• Ensure that there is high quality, unique content upon launch
  Upon launching a new community, we should already have prepared detailed and high quality Q&A to "seed the site" with - not only to serve as examples of what kind of questions to expect, but to be interesting enough in itself to draw people there.

  This means at the very least some 10-20 prepared Q&A posts that should be added to the community within the first week after launch. Give it a flying start. If there isn't enough such content prepared, this in turn suggests that there may not be enough enthusiast users to get the community up and running. If you have some 10-20 enthusiasts users and everyone tries to ask a single high quality question, post a quality answer or post a self-answered Q&A within the first week(s) after launch, then that might do a big difference.

• Quality moderation early on
  A lot of us want to see a new community grow and thrive. So when some community is launched, Codidact users from other communities who aren't quite sure about what that community is about "stumble in" and post questions. I do this myself too, even if I'm just mildly interested in the topic of the new community.

  But as a result, we do end up with a whole lot of naive, low quality questions, borderline on-topic. These questions drive people with actual interest and expertise in the topic away from that site. If the first impression of the site is a collection of down-voted, low quality posts, people are not likely going to stick around.

  To prevent this, we may have to crack down on such "friendly" users stumbling in to "help". It's important that we get scope and moderators in place as quickly as possibly. Anyone signing up for moderator duty need to be enthusiasts of the topic. Not necessarily experts, but at least with enough knowledge to remove low quality and off-topic posts.

posted about 3 years ago

CC BY-SA 4.0

Lundin‭

30 57 412 51