Forget Dunbar’s Number, Our Future Is in Scoble’s Number

social-graph-graphic

Photo credit: Mark Wallace

I probably don’t know about your latest job project. I don’t know what your kids are up to. I don’t know about that vacation you’ve got coming up. I can’t say what city you’re visiting for business. I have no idea that you’re having a bad day.

But I do know you’ve got a really strong take about where social software helps companies.

Why? Because that’s an area where we have a common interest. I don’t need to know all of you, as Dunbar’s Number posits. I only need to know part of you.

From Wikipedia, here’s what Dunbar’s Number is:

Dunbar’s number is a theoretical cognitive limit to the number of people with whom one can maintain stable social relationships. These are relationships in which an individual knows who each person is, and how each person relates to every other person.

This is a recurring issue in social networks. As in, why do people maintain large numbers of connections that can’t possibly be personal?

I like to break it people down into three types.

Three Types of Social Network Participants

I’m oversimplifying here, but this is a useful way to segment how people view their social network participation:

Close Friends: These folks view social networks as sites for staying up to date on a limited set of close connections. As in, “actual” friends.

Information Seekers: These folks, including me, expand beyond those with whom they have a pre-existing connection. Their interest is a bit of networking, and tapping information in their field.

Power Networkers: These folks amass thousands of connections. In the offline world, they’d have huge rolodexes. They want to connect with as many people as possible. Connections are fundamental to their professions. Think Chris Brogan and Robert Scoble.

The Close Friends users really want just that…updates from and interactions with their actual offline connections. When they post an update, they’ll hear from someone they know. When they read an update, it will be from someone they know. This is what Dunbar’s Number is all about.

Then there are the rest of us.

We Have Dunbar’s Number…How about Scoble’s Number?

If Dunbar’s Number is defined at 150 connections, perhaps we can term the looser connection of thousands as Scoble’s Number. The next model of social connections. Now let me explain what I’m saying here.

I’m not saying we can magically follow thousands of people closely because of social media. We can’t.

I’m not saying that we won’t have close connections that we know much more about. We will.

I am saying that a significant percentage of our online interactions will be with people about whom we know little.

That last point occurs as your connections get larger and larger. I follow 1,600 people on FriendFeed, 1,100 on Twitter. I can say from experience now that I know little about many of the people with whom I have @reply and thread conversations.

And it doesn’t bother me. I get plenty of value from these drive-by interactions.

Here’s how I differentiate interactions between Dunbar’s Number and Scoble’s Number:

scobles-number

In the top graph for Dunbar’s Number, you’re aware of a fuller range of what’s happening in someone’s life. Even if you aren’t actively trying to know about it. This is the stuff of warm friendships. You internalize a lot more information about someone, and they know a lot more about you. You develop short-hand ways of talking, and can call on older experiences to relate to new information and developments.

The bottom graph is for Scoble’s Number. Here, you only intersect socially with someone periodically. This happens when the stars align:

  • Someone is talking about a topic of interest to you
  • You happen to see this topic being discussed

Scoble’s Number is a our new reality. By maintaining a larger number of weaker connections, you can tap a wider range of opinions. People often deride “echo chamber” aspects of social media. Well, if you’re only paying attention to same people over and over, you will have created your own personal echo chamber.

This is not to say that we don’t have a more limited set of people we trust as information filters. Those people are important for keeping on top of things in a more systematic way.

But I tend to think of Scoble’s Number as a rich, chaotic frenzy of interactions that never would have occurred before social media was adopted so heavily. Online bulletin boards have this aspect, in that you “followed” thousands of participants on them. Think of molecules bouncing around, with occasional collisions. It’s these collisions where interesting reactions occur. Where you learn things you didn’t know, and you get perspective from people beyond your immediate circle.

It’s healthy. And given the growing participation in social media, and the low friction for finding and interacting with others, I see the trend as favoring Scoble’s Number.

Over time, some connections will move from being out there in your Scoble’s Number into your more personal Dunbar’s Number.

I’m @bhc3 on Twitter.

Advertisement

Social-Filtered Search

Recently, there was a lot of discussion about running searches on Twitter, using authority as a filter. The idea is to reduce Twitter search results to only those with a minimum number of followers. The idea garnered plenty of discussion. From that discussion, I saw some perspectives that I liked:

Frederic Lardinois: I would love to have the option to see results from my own friends (or those who I have communicated with through @replies) bubble up to the top.

Jeremiah Owyang: Organizing Twitter Search by Authority is the wrong attribute. Instead, focus search by your OWN social connections. People you actually know score higher relevancy. http://www.loiclemeur.com/engl…

Robert Scoble: On both services you should see a bias of tweets made by people you’re actually following. Who you are following is a LOT more important than who is following you.

Those ideas make sense to me, because they reflect the way we seek out information. I do think there’s room for search results beyond only your friends. Here’s what I mean:

social-filtered-search

The idea above can best be described as follows:

I’ll take any quality level of search results for my close connections, but want only the most useful content from distant connections.

The logic behind this is that any quality “deficiencies” in content generated by my close connections can be made up for by reaching and having a conversation with them. That’s not something I’d do with more distant connections.

The chart above has two axes: strength of ties and usefulness signals. Let’s run through those.

Strength of Ties

Harvard professor Andrew McAfee blogged about the strength of ties back in 2007. With an eye toward employees inside companies, he segmented our connections as follows:

strong-weak-potential-ties-mcafee

The segmentation works inside companies, and it also applies in the personal world. For example, on FriendFeed, my Favorites List is akin to Strong Ties. The rest of the hundreds of people I follow are my Weak Ties. Friend-of-a-Friend entries I see are my Potential Ties. And of course there are a lot of people I never see. Those would be the “None” Ties.

The hardest part of this segmentation is that people aren’t likely to take the time to create and update their Strong Ties. Rather, Strong Ties should be tracked via implicit signals. Whose content do you click/rate/comment on/bookmark/share/etc.? Extend this out to email – who do you correspond with the most?

For example, I tried out the social search of Delver. It lets you load in your social networks, from places such as Facebook and FriendFeed, and uses content from those connections as your search index. Innovative idea. What happened though is that when I run a search, I get a deluge of content. My social networks are too big to make the service really useful.

Here’s where apps that handle a large percentage of my clicks and interactions will have an advantage. FriendFeed, with an extensive library of content from my connections, has this quality. Inside the enterprise, workers interact with a limited set of applications. The company’s IT department can set up tracking of interactions to identify implicit Strong Ties.

Bottom line: determining Strong Ties via implicit interactions is scalable and useful.

Signals of Usefulness

I’ve already described these in the paragraphs above:

  • Clicks
  • Ratings
  • Comments
  • Bookmarking
  • Sharing

Implicit data + explicit signals are the most powerful indication of usefulness.

Putting These into Place for Social-Filtered Search

When I say that I’d want to receive search results, even without many signals of usefulness, from my Strong Ties, here’s an example.

  1. I’m planning to run a marathon
  2. What marathon training plan should I use?
  3. I run a search for marathon training.
  4. I see a tweet from one of my Strong Ties: “Just started my marathon training this weekend. 4 miles FTW!”
  5. I @reply my Strong Tie, ask what training program he’s using.
  6. I now can leverage someone else’s work on this subject.

Of course, I’d want to see well-rated marathon training programs too, like Pete Pfitzinger’s Advanced Marathoning. I’d want to see the content from my distant/non-existent connections that had the highest signals of usefulness. Not unlike Google’s algorithm.

But the key here is that I’ll make up for any deficiencies in the utility of content for someone I’m close to by contacting them. A search on ‘marathon training‘ in Twitter shows a lot of results. But I’m not going to reach out to most of these folks, because I don’t know them. I only want those with whom I can have a conversation.

As I said, the ability to track both implicit and explicit activity is key to making this work. Facebook, FriendFeed, Twitter and Enterprise 2.0 all seem like good candidates for this type of search.

*****

See this post on FriendFeed: http://friendfeed.com/search?q=%22Social-Filtered+Search%22&who=everyone

You Know, If Yahoo Could Actually Focus, They’d Be a Helluva Lifestreaming Service

Yahoo announced its latest effort around making itself more social.  And unlike previous efforts (Mash, Yahoo 360), this one has potential.

Yahoo beta released the “Universal Profile”. To be honest, the initial release is underwhelming. Here’s what mine looks like:

The profile includes the picture + bio + interests that we’ve come to expect. What else?

  • Status Update = Facebook status, Twitter
  • Guestbook = Facebook Wall
  • Connections = any social network
  • Updates = Facebook news feed, FriendFeed

A Compleat Social Profile. Well not really. But it has potential.

Interestingly, the Status Update is a 140-character field. Not 125 characters, not 150 characters. 140, like Twitter. That probably says something about future Twitter integration.

Lets focus on that Updates box there. Because that’s where maybe, maybe Yahoo can finally integrate its disparate Web 2.0 properties.

Facebook, FriendFeed Have Been Enjoying Yahoo’s Sites

FriendFeed nailed the idea of pulling updates from different services. 43 of them currently, including three Yahoo properties. Check ’em out on FriendFeed:

Facebook, seeing the possibilities in lifestreaming, added feeds into the news feed. So now friends can see your activities on external sites as well, including Del.icio.us and Upcoming.

Currently, Yahoo’s beta release doesn’t include updates from those services. What is included? Yahoo 360, Yahoo Buzz, Yahoo Avatars (huh?), Yahoo Shine. I tested the update timing. Yahoo Buzz updated quickly onto my profile. I created blogs on Yahoo 360 and Yahoo Shine (yes, the women’s site). Neither blog post ever showed up on my profile.

So the initial profile is a bit underwhelming, the sites that feed into the profile are limited and activities don’t update from the sites that are available. Wow.

OK, not the biggest endorsement so far, but read on…

The Yahoo Advantage

Yahoo has three distinct advantages in the lifestreaming race:

  1. Traffic
  2. A set of sites that lend themselves well to lifestreaming
  3. Email/IM social graphs

Traffic

Yahoo draws the second highest traffic in the United States. For reference, here are traffic stats for Yahoo, Facebook and FriendFeed:

Critical to these lifetreaming social networks is having a sufficient number of users. Twitter has maintained its dominance in microblogging despite the emergence of competitors. Biggest reason? Everybody’s already there.

Yahoo ‘s lifestreaming starts with a critical mass of potential users.

Lifestreaming-Oriented Services

Yahoo has a number of sites that lend themselves well to lifestreaming, as mentioned above. The users of these sites are already putting things into the public domain, a psychological hurdle for many people in terms of social networks. While these sites aren’t available yet, Dan Farber reports that they will be soon.

Two services merit particular call-out here: Yahoo Buzz and Twitter.

Yahoo Buzz is Yahoo’s answer to Digg, Reddit, Mixx and other social news sites. Users submit stories, and other users can vote them up or down. Apparently, Yahoo Buzz delivers huge traffic. TechCrunch reported its single highest traffic day ever thanks to a link on Yahoo Buzz. And ReadWriteWeb reported that Yahoo Buzz had overtaken Digg in terms of visitors.

What I like about the integration of Yahoo Buzz into the Yahoo Universal Profile is that it becomes quite easy to see what users are interested in. It also becomes a great way to find stories via your social network.

Twitter is the best-known microblogging service, and competes quite well with Facebook’s status updates. Twitter is enjoying network effects allowing it to pull away from its competition. The 140-character Yahoo Universal Profile status updates are now available. I’d like to see how Yahoo either integrates Twitter into its experience, or creates a distinct microblogging experience built on Universal Profile status updates.

Email/IM Social Graphs

In the Yahoo Universal Profile, you are provided with a list of 10 connection recommendations. How are these recommendations generated? Check out what Yahoo says:

Now in a lot of ways, this is no different from uploading your email contacts to a social network and inviting people. The difference here is that these contacts are already using Yahoo. And Yahoo has the advantage of exposing lifestreaming via the email and IM that people use. Not as some third party social netowrk where a lot of people won’t bother to show up.

This goes back to the existing user base of Yahoo. If Yahoo can figure out how to put lifestreaming in-the-flow of its users’ daily interactions with the site’s properties, it’s got huge upside potential with this idea.

But Yahoo Is Always So Distracted

The latest distraction is the news that Yahoo will be laying off a lot of people. This follows the Microsoft acquisition attempt drama. And Yahoo has a history of acquiring sites, but not doing much with them. And many people question just what Yahoo is doing with this Universal Profile:

Don’t really see a point in setting up a profile on Yahoo!. I mean, I see the company’s goals here, but I do not see any user benefits. After all it’s just another Facebook, however tied to a dying Search Engine.

There is justifiable skepticism that Yahoo can actually pull off creating a vibrant, useful lifestreaming service. If Yahoo could pull it off, here’s what it gains:

  • Higher page views
  • Longer page views
  • Exposure and growth of its many sites via the lifestreams of a user’s social graph
  • A clear and distinct advantage over Google and Microsoft
  • An ability to lead the technology conversation again

I’m rooting for Yahoo from this corner. I would love to see Yahoo bring lifestreaming and social networks to the mainstream.

*****

See this post on FriendFeed: http://friendfeed.com/search?q=%22You+Know%2C+If+Yahoo+Could+Actually+Focus%2C+They%E2%80%99d+Be+a+Helluva+Lifestreaming+Service%22&who=everyone