FOWA07b: Matthew Haughey

Creating and running communities
Started as a single developer, building something after work hours, thought about turning it into a community, eventually it became a business – Metafilter. Aside from coding issues around building a community app, there’s a lot of stuff that he hadn’t thought about. Hope to provide lessons to help you avoid his mistakes.

Mefi, started off as a multi-user weblog with comments. Didn’t like /.’s interface. Sub-sites for different topics. Advertising.

Couldn’t find a Web 2.0 definition that didn’t include social aspects. Tried to find something that’s considered Web 2.0 that isn’t social but only could think of Gmail, but even that’s going in that direction. Any site you have, any idea you have, can go from good to great if you throw a network at it. Peer communities. But if you’re building an app and never thought of the social aspects of it… there are many apps that you never would think would have a community.

Running communities are tricky, lots of pitfalls. A great app and community is great both for the members and the creators.

Lifespan of a community – steep growth to start with as everyone tries it out and then leaves, so a dip, and then people hear more about it and come back or attract new people you get more growth. Three outcomes – continued huge growth, reach an equilibrium, or a decline as the creators make mistakes and the communities dies. Need to stay away from 2 and 3.

Be a third place. Most normal people have home and work and then… their third place, like a pub or sports team or World or Warcraft. Goal is that people should want to join you and enjoy their time with you, get away from it all.

It’s not as simple as just throwing up some software. A lot of thought has to go into it. Going back to very first principles, it’s about the idea behind it, the more clever and original your idea is, the faster your community is going to grow. If something’s been done a thousand times it takes a long time to gather momentum. Winnow your ideas, run with just the very best ones. If your friends tell you it’s impossible, that’s probably a good idea.

We have 10 – 15 years of prior art on communities, social apps are at least 5 years old, so if you’re going to do a new shared to do list manager, for e.g., then make sure you know the best features from everything else.

Eat your own dog food. Use your own app. If you’re building a community you love to fill some need for yourself, build it to scratch your own itch, be the best user of your app. Users will pick up on this and contribute accordingly. It’s obvious when someone is just trying to fill a niche an make a buck. Make it personal – it if makes your life better it will probably make other people’s lives better.

Reward contributors, e.g. status based on postings. It’s not just about rewarding people for contributing, it’s about the readers at the end. Makes your site more interesting to casual readers. If 10% are actively commenting, 90% of people just lurk.

Flickr was a good examples, used to be just your photos and everyone else’s photos, but now they have interestingness pile to show good stuff.

Moderators: your best contributors. Elevate good contributors to moderator status.

Don’t be too controlling, let things run their course, within reason. Be flexible – MySpace is the example where people use HTML in their name area to do mad things. Allow unintended uses – sometimes it’s the most innovative stuff. Build out based on the edges. Use the weird innovative things at the edges of the community, and when you see something good and interesting make it part of your app. Smart tech companies do that all the time – see what the hackers are doing and then do that.

Run it well. Stick with guidelines over rules, as rules put you in positions where you have to do something that you know is wrong but technically they broke the rules. Provide ideas about how to be a good member of the community.

Keep your emotions out of running the community. You are going to get insults from day one, and if you start making rash decisions that just serve you, it’ll come back to haunt you. Bounce stuff off other moderators. Guidelines should be a living document and tailor them to what your community finds acceptable.

Never surprise people with a new rule out of left field – causes outrage. Facebook’s update stream cause outrage. Understand what the community norms are and don’t do things just because you feel like doing it.

Balancing act between chaos and happiness. Some things can push you over the edge.

Ownership is an issue – the more people invest in your community tool, the more they feel that they own it, they love it so much it’s part of their daily life, and there’s grey area about whether it’s yours or the people who show up? Are you going to make decisions top down or bottom up with voting? Keep this in mind – it’s not always black and white.

Happiness on the part of users is fleeting. Having a day of downtime will kill goodwill. People will think of apps as unreliable if it goes down for a day, even if rarely. Twitter has its issues, goes down for a day and everyone freaks.

Every community will eventually have a revolt, you’ll have a problem moment. You might have a few passionate users who say they will quite in a huff. Shouldn’t be unexpected, and it’s good that it happens. Site’s are run by humans and they make mistakes. Everyone has these problems, but take them as an opportunity to learn.

Customer service. Even if it’s free and everyone can do everything themselves, there will be customer service issues. You’ll spent more time on customer service than coding – this sucks for developers, but plan it from the beginning. Expect it. Even if it’s free, even if it’s a side-project, people will expect responsiveness. Soon as you can, hire people or take volunteers from your users. Best hires ever made because frees him up to do other stuff.

Metrics can ease the workload – if people can flag abuse then you aggregate that then you can go and look at problems, not trawl for problems. Or have a support forums, and if you get a lot more questions in one problem area pay attention to fixing it.

Averting the eventual disaster.

Be transparent. Do support in public forums so people can see every decision made. Be honest. This goes a long way – people will trust you and see you’re responsive. have a place to talk about the site or app. Starting to become more regular. If you don’t, people will complain about the site all over the place and will wonder why you didn’t see their comment on their blog. People assume you’re reading everything, everywhere, so have a dedicated place for it. Gives you a good venue for collaborating on new features – float mock-ups and discuss it with people. Wholesale redesigns cause revolts if you do them overnight.

Always over-explain changes. Make sure you’ve covered every base possible.

When things go really, really wrong. Take the high road and acknowledge you made a mistake. Don’t play the blame game – you’ll lose community goodwill. If you see your community turning, email people, IM people, even if they hate your guts. It takes the wind out of their sales, and gives them a chance to understand your mistakes. They forgive you, and appreciate the honesty.

Legal problems. Anything multi-user will have legal issues. Comes up more often that you think. You make a multi-user apps, and it’s used by the world, so try to understand which laws apply. Find a well-versed internet lawyer. Might take a while to find someone but worth finding someone who understands the internet. Set up a business to give protection to your own personal life. Terms of Service and Privacy Policies – be clear about what you are going to do with people’s contributions. Get a lawyer to help on language with that. Copyright law, e.g. DMCA in the US. If you’re accepting content from people you could put yourself in legal hot water if someone uploads copyrighted materials.

Lawsuit threats are many, but lawsuits are few. Usually gets one veiled threat a month, specially from companies that people are talking about negatively. Digg is probably getting them daily, but it rarely ends up being much.

What’s stopping your site/app from building out a community?
Have a social aspect so people can talk to each other. Good communities can please readers and creators.

Q: Do you see periods where the community dips, then has to restore itself again, almost like a two year cycle? Can a dying community revive itself?
There are ups and downs, and I’ve seen communities take a dive when the community admin doesn’t have time to nurture it.

Q: How do you keep the core community welcoming to new members, rather than getting defensive and closing in?
It’s a natural trait, it happens all the time. The old guard will fight the new, will mock them. Haven’t discovered the magic bullet for fixing that. Can tell people it’s not cool to shout down new members, but I can’t think of any quick tricks I’ve seen people do. Digg does something new by showing you recent joiners and people meet that way.

Q: More focussed your community is, harder it is to find people interested in joining. How do you find those people?
Once those communities get going, people get way into it, and it’s a good thing after the first six months to a year of rockiness. I would find similar, not to the point of spamming comments, but find similar user groups and maybe announce it in a mailing list about that subject, but without stepping on toes. Subtly promote it.

Q; When you are at that initial stage, prior to critical mass, how do you make it look like an attractive place to congregate.
I started out with good seed content, which is important. Started with friends, showed it to then, and half of them started posting, so 10 people started posting. It wasn’t phoney, it was honest, but their was a silent slog on my own for nine months, and eventually people showed up. Seed content is important. Highlight best bits.

Q: Is there a risk of over-nurturing your community?
At some point, yes. At some point I got frustrated and told everyone that the more I spent on the forums, the less time I can code. It got out of control because I wasn’t realising. When Blogger started, didn’t have a pay version because didn’t want to do support, but end up providing support anyway.

Q: Are you earning more money now than when you were a developer?
Eeer, yeah.