Post History

A community with common interests is a part of it too.  At Some Other place somebody wanted to combine all the Abrahamic religion sites into one, arguing that they share a lot of biblical books and tags could differentiate religious context.  I argued against this because the three communities view those topics very *differently* and there weren't people who wanted to build a community for all three religions, but there were people who wanted to build communities for each of the three separately.  In addition to different views of the bible, there are whole additional layers -- extra-biblical sources, customs, etc -- that differ among the three.  This community would have been too general; by making everybody wade through a bunch of stuff they didn't care about to find their own subcommunity, that arrangement would have actually driven people away, in addition to not providing any benefit from those subcommunities all being in the same place.  (It also likely would have led to tensions among users with *very* different perspectives.)

*But*, in an alternate universe, I could imagine a "religious studies" community forming, drawing from academia (some universities have departments for this), where there would probably be more cross-cutting questions, fuzzier lines, and fewer people deeply invested in personal religious values.  That community, having basically the same breadth as the one in the previous paragraph, could work just fine -- because the people in that community have that shared interest.

So it's not really about *scope*.  It's about the *people*.  Who are the people coming together to form a community?  What do their boundaries look like?  What forms their identity?  Where do they see themselves going?

Now consider an example from our own network, Software Development.  This topic is pretty broad; on Some Other network its scope would cover a dozen or so different sites.  Some Other network, as you point out, has tended to carve off smaller scopes for sites -- *often*, but not entirely.  I think that's because of scale; Stack Overflow is so huge that people with a more focused interest like databases or quality assurance or web development can feel lost there.  Their questions are on topic on SO, but it's harder for them to find their communities (people).  They can do that more easily on focused sites, and so, over the years, SO has spawned a bunch of specialist sites.

On the other hand, there are communities with a very broad scope, like Writing, that *haven't* spawned children, because even though the scope is huge, the community is still fairly small.  Even in the before times, if SE had split Writing up into sites for fiction, technical writing, and other topics, each of those sites would have had too few people to sustain them.  (In fact SE tried to create a technical-communications site, over the objections of the Writing community because it was 90% scope overlap, and that community failed.)

In my opinion, we should create a separate community here when either the topic really is specialized (it doesn't make sense for it to be part of something broader) or when there is an active community that wants to focus on that area of scope.  Either way, we're looking for enough active people to sustain it.  If there's an active community of people, I don't really care whether the scope is narrow or broad -- they're doing stuff that makes sense for that community.  So long as we don't create confusion or tension on our network of other communities, I don't see a problem.

We do not have the user base of Some Other network.  When we were planning the Software Development community, there were arguments for a general community and arguments for several specialized communities.  I don't currently see enough people to support any specialized communities, and as you point out, more related communities creates a little more friction for people coming in.  So we created the general community, *but* with the expectation that as it grows, sub-communities might want to split off.  We will support that in a way that minimizes disruption to all involved; I think that just by planning for that kind of growth from the start, we're already ahead of Some Other network where the subcommunity would have to start by proposing a brand-new site and people with questions would still find lots of questions on SO even though there's that other site over there.

(What does our plan look like?  I have a ticket to develop this idea and share a draft with the community, which I haven't had time to work on yet, but the broad idea is that we'd spin off a new community but make its content (or some of its content?) available on the parent community too and steer people from the parent to the specialized community when appropriate.  If there's a Javascript community but you go to Software Dev and search you would still find hits (that would guide you to the JS community).  There would be something TBD in the question-asking flow that would lead you there based on tags.  *Something* like that, but we haven't fleshed it out yet.)