I run 300 forums, and I had a startup for them. The lofty goal of the startup was to "cure loneliness" but I never told the users that.

My path to doing this was to encourage interest based communication, in a way that would allow them to take it offline. How? Well to make it interest + geography, i.e. horology nerds in Cumbria, cyclists in London, petrol heads in Cornwall, audiophiles in Suffolk.

By binding together an interest, and a geography, it naturally encourages trade, showing off, having a beer... and coming together. But not in a way that ever was to "cure loneliness" (which few people admit to feeling, and frankly most people find lonely people unattractive people from the perspective of wishing to hang out with someone like that).

By providing a space to share a passion, the forums brought people together.

I had to make our own software to achieve this, i.e. a forum with events built-in by default, etc. But it worked.

Except I couldn't get it funded beyond the crowd-funded angel round :shrug . But still, 250k monthly users, hundreds of marriages, tens of thousands of friendships, and still people meeting weekly for beer and chatting with others.

And like all love, when the lust is gone what remains is something special... these forums are evolving now, from the original interests, into just a group of friends for life.

If I were lonely today, I would say... identify a hobby or interest you could get into, start looking for entry things to dive in, and then look for groups on MeetUp, or forums online, that will help you find other people.

I feel like it would be appropriate to share some of these forums in this thread, I think the immediate need outweighs anyone calling it out as self-promotion.

Not all of them are applicable, the platform has also been used for forums that connect old-age pensioners homes (it's easy to use), and support networks for victims of domestic sexual and violent abuse in South London.

However the largest example is the London cyclist one https://www.lfgss.com . I started with cycling and the plan was to saturate one sport/recreational interest before branching out. So I run a lot of the cycle forums in the UK, from the Rapha one, through to Islington http://forum.islington.cc/ , Brixton https://forum.brixtoncycles.cc/ tiny clubs like Grupetto https://csgrupetto.microco.sm/

One of the newest is a sustainability forum for a group with a core around Cambridge but spread over the UK https://onething.is/

Some of the forums are private, and I'm fine with that as they can permission it the way they want and pull members from elsewhere as they see fit, i.e. the petrolheads in Cornwall.

Each forum is like a startup itself, they all have to find their first 10 users, then to get to 100 users, and to be aware of the 1% rule of internet culture https://en.wikipedia.org/wiki/1%25_rule_(Internet_culture)

But I help nurture and guide, and show them how to do it.

Without funding I've slowed everything down though. It's funded by donations from users across sites today, but expansion risks cost cliffs as I'm just below the threshold for a few services I would need to pay for (auth0 free tier) and don't have time to engineer my way out. With neither time or money spare, it's stagnated.

But if someone wants a startup idea... this does work. Not wildly profitable, but it works.

How did it make money? I'd barely scratched the surface, but conversation around interest contains a lot of intent to purchase, and there are classifieds and meetups. So it's affiliates, a challenge against eBay (people into an interest have emotional attachment to the items and would rather sell to others into the interest), and ticketing for events, etc. And because it's all a platform, one instance, the traditional highs costs associated with running a forum were avoided. Lower costs, and actual revenues :)

I love what I do for work now, but I still believe this was an idea that is good for society.

Any plans to open source / open core this software? I've always been interested in forum software (my first real web project was a Reddit clone in Flask and the complexities and edgecases abound - moderation tools, rate limiting, scaling write-load, finding the minimum front-end that doesn't make users puke).

The only thing that isn't is the very old puppet script that managed the deploy as it was rather bespoke for our setup.

To my own regret I focused far too much on making it an effective platform rather than an easy install, so that bit might feel gnarly but at least the errors are sane and guide you.

The Go API and PostgreSQL schemas: https://github.com/microcosm-cc/microcosm

The Django frontend (the client is nothing but a thin client over the API) https://github.com/microcosm-cc/microweb

The Bootstrap derived theme for the Django fronted https://github.com/microcosm-cc/microweb-bootstrap

And then other miscellaneous things:

The Go HTML sanitizer for user generated content https://github.com/microcosm-cc/bluemonday

Our legal policies for forums on this platform (expensive to produce, but perfectly fitting a forum platform with minimal exposure for the platform owner/admin and minimal but some liability for a forum owner) https://github.com/microcosm-cc/legal

There is also a newer thing, I was (/am?) intending to replace the Django layer with a Go frontend and templating, and then moving the API into this, such that the forum could become a single binary install and thus gain a new lease of life: https://github.com/buro9/microcosm . Once in a while I chip away on that.