I don’t usually write a lot about work on this, my personal site, but I read Derek’s musings about commenting on newspaper sites with interest, and felt moved to share this, a single slide from my communities and media presentation, which I’ve given to a number of audiences over the last 18 months or so. It’s relevant here, so I hope it bears repeating.

When media organisations think about community development and management, or dealing with potential problems, they need to consider the holy trinity of community management and not just one part of it.

Human (or organisation) solutions include anything which involves throwing more people or organisational muscle at a challenge: policies, procedures, hiring armies of moderators (or external teams to do it for you), extending rotas, writing community into people’s job descriptions and responsibilities. All of these are important, but they’re not necessarily scalable. Human solutions can be a lot like that game you play at the fairground, whack-a-mole, except it’s probably more like whack-a-troll. Employing people to bop away problems isn’t efficient, and ultimately rewards bad behaviour with attention.

Technical solutions are things which allow technology to help with the management of the community. This can include powerful moderation tools and filters which can be applied (like StupidFilter or disemvowellment, but can also include reputation management, recommendation, user hierarchy and/or associated weighting, rate-limiting, peer moderation, pre- and post-moderation and the like. Technology can very powerful in helping to manage a community, but it’s not very effective when used on its own.

Editorial solutions are often overlooked, and are crucial to community development on media sites. They might include journalist (and other staff) participation and highlighting/featuring valuable comments or users, as well as informing the more general editorial approach to community - how does it reinforce or complement content? What is the value or proposition behind user interaction? Why are people there? And what do you expect (or hope) from their interaction? What is the quality expectation? How do active community users reflect your readership, or not? How do you use conversations to address (or reinforce!) power imbalances between you and your readership? Is the role of the journalist to start conversations or to provide platforms for conversations to occur? Thinking about how community participation is influenced by, impacts on and interacts with other kinds of content on the site is key.

The point of this diagram - this whole concept - is simple: you can’t do one of these bits and expect everything to work out great. You have to think about what can and needs to be done in each bucket - quantities and blends vary wildly depending on the kind of content and kind of community - but they’re all important.

This is something I’ve been working hard to make a reality within guardian.co.uk’s community activity over the past 15 months (all three aspects, but with a big focus on organisation and technology, the fruits of which are ripening now) and also in all future community strategy, development and management. It’s definitely starting to make a difference.