Kristian Luoma from Finland pinged me yesterday, curious why I still used Foursquare. When I lived in Finland, we were one of the first people on the ground to use the app and we used to compete on who would retain the mayorship of Helsinki’s Vantaa Airport.
I no longer really care about being the Mayor of someplace but I told Kristian that I still use Foursquare regularly as a personal journal of places I’ve been. I have all my check-ins written to a Google Calendar (I use an IFTTT recipe to do this) so that I can quickly check where I’ve been when needed as a reference.
I also religiously check the recommendations left by others. I find the smaller, explorer-minded crowd on Foursquare more interesting than those on Yelp.
But what I really like about Foursquare is the collective data that you get after logging your location over time. I’ve written about them before (Timemachine, 2010 Infographic, and WeePlaces)
There’s a new one that I missed released for Foursquare day back in April.
I ran across some notes from the Web 2.0 Expo back in April that are still relevant and worth sharing. Today I’ll post on the talk that Jyri gave on Building Sites with Social Objects, tomorrow I’ll post notes from a talk given on iPhone Development Anti-Patterns.
Jyri Engestrom founded Jaiku which was later acquired by Google and is involved in some of their most interesting social networking products including Google Latitude. In his session, he started by giving a quick run down of successive social networks from the past emphasizing that despite media coverage of facebook (and more recently twitter), the game is far from over:
Firefly, grew to 2M users, acquired by Microsoft Six Degrees, grew to 3M users, folded Friendster, grew to 90M users, collapsed under it’s own weight MySpace, tens of millions of users, acquired by Fox Facebook, over 250M users, still growing and independent
The game is not over. We are still talking about a segment of the population. Social Networks have not (yet) replaced e-mail, sms, or the telephone as the lowest common denominator way to get in touch with someone.
Jyri went on to describe how social networks are built and what differentiates a successful social network from others that fail. Most importantly, social networks are about connecting people but there needs to be a catalyst to drive that connection, something with a tangible incentive. Using the metaphor of kids gathering to play together on the beach, Jyri explains that they gather together around a common object, such as a ball. In this same way, people connect around something he calls a “social object.” Think of all the successful social networks and you can see this pattern:
YouTube – video clips digg – links flickr – photos last.fm – music tracks good reads – books slideshare – presentations
In each case, there is a shared object that drives the connection and draws outsiders into a conversation. Bringing up the case of Russell Beattie, Jyri talks about LinkedIn. It’s a useful but only as a profile service, a place to showcase your resume and people don’t connect around profiles. I would argue that LinkedIn has been improperly categorized as a social network when, in fact, it’s more accurately a crowd-sourced recruitment database.
Jyri then went on to list out the steps towards designing a successful social network:
1. Define the Social Object.
2. Define the verbs around the object. For example,
– eBay – buy/sell
– flickr – upload/favorite
– dopplr – add a trip
– upcoming – add/watch an event
The Activity Strea.ms group has been doing some work here and their wiki and mailing list is a great resource. They have catalogued a number of common verbs and are attempting to unify these verbs to enable broader sharing and connectivity across social networks.
3. Promote the sharing of objects with easy to use tools
– ensure all your objects can be adddressed by permalinks which can be emailed
– create embedable widgets so that bloggers can promote attractive galleries of your social objects. Think flickr badges.
4. Turn invitations into gifts. Each invitation sent by your members should have an immediate value attached in the payload of the message. The days of “Register or click here to see more” are long over.
5. Charge publishers, not the spectators. The people that use your platform to further their financial goals should pay to access your audience.
Finally, Jyri spoke about the future. I’ll add my own observations in here as well because I’ve been doing a lot of thinking about this too.
As with the Russell and LinkedIn, the days of connecting just to goose your numbers are over. Twitter is the latest to inherit this behavior and while it may be valuable to have a large number of followers, it makes little sense to follow too many people yourself. As Robert Scoble has realized, the less people you follow, the better the signal.
Future behavior on social networks will be to enhance what Jyri calls, “social peripheral vision” Citing examples such as the head’s up display on World of Warcraft, Jyri tries to imagine a world where your physical world is annotated with data about their interaction your social network.
Today it is still very early in the game. Each of us, with our lifestreams and status updates, are firing off signals like “pulsars into space.” The tools we use to monitor and keep up with are friends are still very primative and much of the talk is about trying to keep up or make sense of it all. In this sense, as we consume a combined lifestream of all our friends, we are not unlike ham radio operators, sifting through the radio bands and sharing notes as we look for a signal, a pattern.
Jyri concluded his talk describing something he called “nodal points” which is he uses to describe a, “pattern or algorithm that pulls information out of data.” It is this pursuit of nodal points that we are seeking. Each social network should have it – it’s where the collective commentary draws a pattern that represents a greater intelligence. Think of flickrs’ Shape files, the clusters of headlines on Techmeme, or the still fallible trending topics on twitter. These are nodal points and each successful social network should have them.
So here’s Jyri’s checklist for a successful social networks.
Define a social object
Define a set of verbs & actions which can be taken on the object.
Aggregate the social objects and annotations by the community to create nodal points.
A recurring theme at this year’s Defrag conference is the concept of a flow app (see Stowe Boyd post from last year) which I loosely define as an aggregator that brings together multiple information streams into a single view. More broadly known as lifestreaming, these applications are still very basic and only being used by a niche audience of the hyper-connected. Brad Feld said that we live in a world stuck in “manual configuration mode” where the pruning and tuning of our filters is too crude. Call it the Facebook news feed or FriendFeed, these apps are great at bringing together information but are starting to grapple with the problem of how best to filter out nuggets useful information from the collective noise.
In a world where the daily newspaper is a day late, bloggers and the twitter masses filter through the flow and make decisions of what pieces to pass on to their audience. Whether we acknowledge it or not, we are stepping in to fill the role of the editor or journalist – we pick and chose from our own personal newswire to decide which bits are worthy of passing on, with or without our own editorial color.
During Dick Hardt’s demo of his sxipper firefox add-in, someone asked what one would consider the feature set of a flow app. Here’s a running list of what I noted. Please add more in the comments.
Flow apps are complimentary to existing apps, they monitor activity that takes place on other applications. I take this to mean that the tools are not up to the level where they could replace the raw information flows.
A good flow app doesn’t demand completion, it doesn’t provide an unread items count.
Default ranking should be on something more like relevance, not straight reverse chronology sort. I would love to see a social sorting algorithm that took into consideration the popularity of an item within your social network (look at subscriptions, in-links, comments, trackbacks) and the freshness of an item.
Real-time scrolling. MyBlogLog’s New with Me pages has updated in real-time since it launched. FriendFeed also added a real-time scroll with a helpful pause button. People love ambient animation.
Provide visual queues for items that match specific contexts. Is there an item that is from someone you’re meeting with today? Is there a restaurant review about someplace close by? Is there an item from an old friend that hasn’t posted in a long time? Highlight those items as “must reads.”
Collapse frequent items from people in your list on media that tends to generate lots of activity. If someone just listened to every track on the Beatles White Album, do you need to know each track? Summarize it intstead (“Chris Law listened to the White Album by The Beatles”).
Fade activity that is over 24 hours old into a graphical representation that lets you easily pick out spikes in activity. You shouldn’t care that you missed everything while you were away for the past week while you were away in the Bahamas but it would be nice to know that a ton of activity came through around a particular tag that broke through the baseline.
Provide a preference page similar to Google Reader’s Trends view which reflects back your activity and allows for manual tuning and override. I love the fact that Trends has a “most obscure” tab for the feeds you subscribe to that have few readers and provides a helpful trashcan icon to clear them from your subscription list.
Just re-discovered Dipity which was a cool tool for creating timelines. They now have jumped onto the lifestreaming bandwagon and provide a way to publish and share a portion of your lifestream in a visual way.