The following post is a fleshed-out version of my notes for my talk at the Elgg International Conference on Monday, December 1st, wherein I discussed my attitude to social networks and how they should be built.
My slides are available in Powerpoint or OpenDocument Presentation format.
Let’s take this to first principles. Stating the obvious, what is a social network? Is it a collection of profiles, friends lists and so on, or is it something more fundamental? What does the term even mean?
Social is an adjective that means relating to human society and its members.
A network is an interconnected system of things or people.
Therefore, I’d suggest that we can define a social network as just being an interconnected system of people.
When defined like this, everyone has a social network, regardless of Internet or technology use, and they’re as old as human society. Your friendships, colleagues, professional contacts, fellow students and book group members are all social networks. They’re not necessarily communities - a “community” tends to imply a common geography or set of interests, which isn’t always true to a social network. But while a social network is not always a community, a community is always a social network.
Clearly, social networks are made of people, who are joined through something in common - perhaps an interest, a shared geography, a mutual friend, a workplace, etc. People are complicated; they have individual personalities, quirks and foibles, which make it hard to interact with them in a cookie-cutter way.
Because people are complicated, networks of people are exponentially more complicated. To get the most out of your social networks, you need to be able to embrace everyone’s individuality. Furthermore, they’re not discrete; they may overlap in all kinds of ways. My friends may also be my coworkers, or someone at work may also be a part of my knitting circle. (If I had a knitting circle. Cough.) They have all kinds of different contexts, which may impose requirements on how the members of the network interact with each other. Work colleagues generally need to communicate within an office space, or via methods imposed by management, for example. More formal networks have more restrictions. Personalities may also impose restrictions: some people are bad at talking on the phone, for example.
Of all the tools and methods social networks can talk to each other, the Web is just one. Face to face conversations, telephone calls, SMS messages, faxes, emails, letters and telegrams are all perfectly valid types of communication.
So in short, let’s reclaim a piece of language: a social network is an interconnected system of people, as I’ve suggested above. The websites that foster social networks are simply social networking tools. A social network doesn’t live on the Web, but a website can help its members communicate and share with each other.
With this in mind, what’s the best way to foster a social network using a Web tool?
Joshua Schachter, the creator of Delicious, has this to say:
“If you need scale in order to create value, it’s hard to get scale, because there’s little incentive for the first people to use the product. [...] The system should be useful for user number one.”
In other words, people need to be able to visit your site and see something immediately useful, even when a network has not developed around it. Flickr, first and foremost, is a site for uploading photographs. Delicious is a flexible bookmarking utility. Facebook is the exception to this rule, because it’s a utlity that helps you keep in touch with your existing friends - but because it was initially limited to Harvard students, Mark Zuckerburg et al were able to carefully grow it from a handful of people. The Harvard community was an existing social network, and Zuckerburg simply gave them a tool.
To summarise, this deserves its own paragraph: you cannot install a social networking tool and assume that a network will grow around it. You must either have another purpose, or an existing network of people to plug into it. Either way, it’s also going to take a lot of work: you need to lead by example, and participate heavily every day.
As each tool should focus on one particular network, or at least type of network, I’d argue that the exact feature set should be dictated by the needs of that network. Educational social networks might need some coursework delivery tools; a network for bakers might need a way to share bread recipes. The one common feature in any social network is people; even profiles may not be entirely necessary. (Look at Twitter.)
What they should do, however, is amplify the network effect. The idea of a social networking tool is to make that network communicate more efficiently, so anything that the tool does should make it easier for that network to talk to each other and share information. The tool itself shouldn’t attempt to create the network - although that being said, new network connections may arise through a purpose. Most of us have made new contacts on Flickr or Twitter, for example, because we enjoyed someone’s content.
The final lesson is that, once again, people are individuals, and social networks are complicated. Therefore, the featureset in any tool needs to embrace as much of the full range of personalities and ways of communicating as possible. Tagging was a great invention, because it didn’t try and dictate the terms with which people sorted their content. As Schachter said about Delicious in the above linked article:
“If I went in there and said, Hey, you’re using that tag wrong, people would just tell me [where to go].”
In other words, he was smart enough to leave people to sort their bookmarks however best suited them. There will be inevitable variations in the tags different people use to describe the same resource, but because the network’s personalities are catered for, they’re more likely to continue to use the tool.
This attitude is what led us to develop Elgg, initially for the educational market: a user-centred social networking tool to support educational communities rather than the top-down, rigidly specified software that was common at the time. The features we built into it - extremely granular access controls, cross-site tagging, personalisation and customisation for site admins - drew a lot of attention, and it quickly became apparent that they would be useful in scenarios well beyond education. We spent the next four years developing Elgg into a flexible tool for facilitating social networks.
The latest version - rewritten from the ground up to be even more flexible, while learning from all the feedback and Elgg usage to date - addresses all the aspects of social networks I’ve discussed above, except for one: overlapping networks. That’s what the Open Data Definition is trying to solve - and something we’re coming very close to being able to support. Marc Canter is trying to solve something similar with his Open Mesh, and he’s not alone.
The Web has become a great tool for supporting networks of people, and with the kind of innovation we’ve seen over the last eight years, can only become better. The only remaining question is: what kind of network do you want to build?
