Smesh Blog

Social network futures: information wants to roam free

Posted by on April 6th, 2010 in Theory & Research

The Internet tends towards open standards where its infrastructure is concerned, not least because there are so many different interest groups operating on the net who stand to benefit (though of course there are massive tussles over lower-level technology that place restrictions on usage, such as DRM on video, music and games). There follows some musing on what is happening in this regard in the space of social media and where such activities are leading.

Background: niches and walls

[Personal anecdotal content; may not be quite correct] Back in the 1990s, if you were an AOL user, you could only access your email using proprietary AOL tools and were encouraged to browse sites that existed within a walled-garden of AOL-approved content. This worked fine for AOL, as it gave them a lot of control over users, and they could ensure that these users saw AOL’s choice of ads while they were doing things within the online space that AOL defined.

This didn’t last for very long. People wanted to be able to access their AOL email using rapidly improving 3rd party email clients (a ‘client’ typically being a piece of software used to access an online service), such as Netscape mail and Microsoft’s then popular Outlook and Outlook Express. AOL ended up adopting email standards such as POP3 to enable this to happen, otherwise users would have migrated to other, more open email service providers, which many did anyway in the time it took for these changes to happen. Similarly with AOL’s walled garden of sites. Users wanted to be able to browse the web freely, and lots of sites out there wanted to attract users to then make money, either by selling them stuff or showing them ads. People just used their own browsers via their AOL internet connection instead of using the official AOL browser app.

Having limited, controlled, walled content seemed fairly reasonable at the beginning, when email was a niche, specialised service alongside Bulletin Board Systems (BBS), Internet Relay Chat (IRC) and Usenet groups; similarly there weren’t that many Web sites out there.

Knocking down walls for fun and profit

As email and web browsing became well-defined things that everyone wanted to do lots of the time, having walls became problematic and the barriers were pushed down.

Even when a technical protocol for doing something isn’t opened up, people invent technologies to bridge closed things to open things. For example, Instant Messaging has standardised around XMPP. Where some IM networks haven’t embraced this open standard, there are technologies now built into the open standard so that these proprietary IM networks can be accessed via XMPP.

Pressure to be open

Very generally, there’s a process as follows that happens, again and again:

1. Some specific app, whose ultimate popularity is unknown, gets built using existing tools, but isn’t particularly ‘open’ — this happened with HTML and IM for example.
2. If the app is massively adopted, it becomes a ‘thing’ in its own right (perhaps just an accepted way of doing some kind of stuff), with pressure for it to be defined by a set of standards that users and tech providers across the ‘net can tap into.
3. If the creator of the initial ‘thing’ doesn’t keep up, it gets sidelined as the benefits associated with participating in the open form of whatever-it-is increases and open alternatives build up market share.

Facebook’s high walls

Something similar is happening right now with social networks.

Currently, Facebook (FB) is the ultimate social network walled garden; an entire ecosystem in its own right. FB wants to keep all of the substantial revenues associated with having a vast and highly engaged user base, so it has built the walls very high.

Twitter is similar, though the walls are built very low indeed, with stiles placed at strategic intervals to help content move freely across it. Twitter has very accommodating APIs to help 3rd party developers tap into its functionality; this is a common commercial approach to reducing the motivation to build open versions of things; Google use this technique a lot too.

The problem for FB especially (with its very high walls), is that social networking has moved from being a quirky niche application that a few websites provide, to being a fairly well-defined ‘thing in its own right’, understood by a large portion of web users.

Broadly, the social network ‘thing’ goes like this. There are a bunch of people with connections between them of various types. This is the ‘social graph’, based on friendship, family membership, business connections and so on. Individuals create content: they write status messages, add calendar entries, mark sites as favourites. This spreads out to some number of their contacts based on the creator’s preference and privacy settings, and then their connected contacts’ settings.

Open standards for social networks

There is now a lot of momentum behind the scenes attempting to define open standards for these entities: people, connections, events / activities, and so on.

One of the leading contenders is ActivityStreams
(, which Google has a hand in. Google generally benefits more from the flow of information — to which advertising can be attached — than from specific applications, so they tend to embrace open standards. Once information is publicly accessible, Google can index it and so provide search results for that data, with revenue-generating ads alongside. Plus they’re pretty good at building browser-based versions of apps (e.g. gmail for email), which they can serve another layer of revenue-generating ads into, on top of what they do with search results.

There is likely to be rapid adoption of open standards in this area, as pretty much everyone bar FB stands to benefit from it. Users will have more control over and ownership of their information, and commercial outfits will be able to make money from a range of value-added (or just ad-added) services without having to digest arcane and restrictive Terms of Service and deal with ever shifting and breaking APIs.

Right now we mostly have ‘pretend’ integration between different social networks. Apps like TweetDeck pull ‘your’ content from different sources into a single UI, but have use completely different APIs and techniques behind the scenes to do so; users have to set up Twitter to repost Tweets to FB; and so on. These are resource-intensive work-arounds for the lack of common standards.

Impending changes

This shifting landscape will bring an interesting challenge to FB. As other existing and new networks open up, with the weight of the likes of Google behind them, and compelling apps are built that tap into these networks, FB as a destination will start to feel isolated and stagnent if it doesn’t open up too. If they do open up their content, then they won’t be able to monopolise it in the way they do at present.

As incentives for open social networking standards increase, we’re likely to see:

  • Proliferation of 3rd party apps (both desktop and online) for managing and mashing up data across networks
  • More consistent APIs available on existing social networks (as OpenSocial seeks to do)
  • Increasingly defensive Terms of Service from big networks like FB, attempting to guard the commercial value of data that they’re being compelled by the market to expose
  • Ads networks and producers of ads will try harder to tap into social media
  • New ad networks and formats will emerge that have been designed with social media engagement in mind

Longer term / smoking crack

Predicting the future around technology tends to be a waste of time, because technologies and social trends come along and engage with one another, having a transformative rather than linear or additive effect. But here goes.

From my experience at Poke, I see how digital campaigns increasingly involve a degree of integration with social networks. If you can create genuinely compelling and engaging content that plays to the strengths of social networks (sharing, competing, etc.), or is just novel and fun, then you get a lot of return for your investment compared to the cost/benefit of building a standalone microsite and paying for a load of banner ads. People are their own distribution network (peer to peer, pretty much literally), but have to have a motive for distributing things.

If you combine a trend towards more decentralised, open and peer-based content with the rise of mobile computing (from high-end handsets to new devices like the iPad), you end up with a world that looks a lot like original visions for RSS and the semantic web that didn’t or haven’t yet come to be. Content flows to the user from a broad range of sources (friends, news outlets, brands, etc.), can be fairly ‘rich’ thanks to increasing HTML5 support (video, apps that work offline) and back out again.

The deluge problem (too many emails, too many RSS feeds) is to some extent ameliorated by application of the social graph. For example, for many sources of information, I don’t even want to see what they’ve got to say unless someone I trust gives some item a thumbs-up. In contrast, my girlfriend’s merest expression of online whimsy needs to be readily available to me, lest I be deemed oblivious to her life while chatting to her after work.

Er, the end. Which of course it isn’t.