As we get into the middle of January of 2008, there are two questions that has lingered since I dove head first into the world of social media over a year ago: Where are my friends and whom should I be friends with (and who should be friends with me)?

2007 saw an explosion of broad-based social networks, such as Facebook, Ning, Bebo and niche social networks such as Shelfari (books), Dogster (dogs) and Dopplr (travel).

It seemed that every site that launched added a social component, whether it made sense or not.

So, what was the first thing I did every time I signed up for a new network? Checked to see if my friends were on it.

More often than not, this was accomplished by uploading the email addresses that I have in gmail, yahoo or outlook. The basic flaw in this, is that people I am friendly with on Twitter or in Facebook are not always in my gmail address book.

So what is missing from today’s social media framework? Two things: 1) friend/content discovery; and 2) the ability to search through social content.

I want an application that allows me to list all the social networks I am a part of and then does two things:

1) Creates a single list of every friend and the networks they are apart of; and

2) Suggests to me whom I should be friends with, based on trust relationships.

There are several social aggregators that sort of do the first, and semi implementations of the second (Last.fm’s neighbor feature is kinda along these lines).

There are niche plays, like EventVue, that attempt to suggest which conferences I should attend based on my friends conference activities. Which in theory is a great idea, but in practice is limited by budget and focus. Knowing the founders as I do; I know that they are spending a lot of time thinking through these limitations.

SocialThing! Does what other aggregators dont by having a robust friend finder feature, which I why I am excited for its launch (and have been–maybe–a bit hard on its founders to do it right), but as far as I know, it doesnt have much of a friend suggestion engine.

FriendFeed has a “recommended friends” tab, where they list “The people below are popular among your friends, and you might find their feeds interesting.” But dont seem to explain how they select their recommendations.

So, maybe in 2008 discovery not only where my friends are, but WHOM I should be friends with will become a reality.

Then, once I figure out which social networks to be a part of, and who my friends are and should be, I need an easy way to search through all that trusted content to find the information relevant to me.

Enter social aggregator/search applications, such as Lijit.

When I joined Lijit in December of 2007, I felt it was a great bridge between my previous search engine marketing life, and social applications. What I didnt realize was what Lijit could become.

I have watched Mahalo launch and found it interesting (if they could really write unbiased, informational pages for the top 10,000 searches, there is value over Google), but thought it was missing the social element, which they added with Mahalo Social, which makes Mahalo so much better, but in many ways incomplete.

What is still missing from Mahalo is the concept of trust (Imagine using Lijit to search through a Mahalo Guide’s personal blog or Flickr and realizing that they produce great Mahalo pages as well as external content, and have it all display on a Mahalo page? Mahalo’s results would be so much more valuable and trusted, unlike Wikipedia’s).

So what does the killer app I would like to see in 2008 look like?It would have the friend finder feature of SocialThing mixed with the recommendation engine of FriendFeed. It would have the search capability of Lijit to search through all my friends’ trusted content (be it from blogs or any social network) mixed with highly relevant results, such as Mahalo and search specific results–for example, maybe music/concerts from PocketFuzz or HypeMachine or events from Upcoming or Meetup.

This would allow me to know where my friends are; whom I should be friends with (and who should be friends with me) and search through all that content to find the information that matters most to me in context, regardless of where it lives.

So, who wants to buy up SocialThing, FriendFeed, Lijit, Mahalo, PocketFuzz, HypeMachine, Upcoming (really Yahoo!) and other potential content sources to build the killer app of 2008?

UPDATE: Brad Feld wrote about have an option to create a “Friend Hierarchy” which would also be a cool feature of my “killer app.” After all, once I discovery where my friends are, whom I should be friends with (and who should be friends with me), the ability to categorize those friends and search their content becomes highly valuable.

Popularity: 9% [?]

Share and Enjoy:
  • Suggest to Techmeme via Twitter
  • RSS
  • Twitter
  • FriendFeed
  • Digg
  • StumbleUpon
  • Yahoo! Buzz
  • Reddit
  • Sphinn
  • del.icio.us
  • Facebook
  • Mixx
  • email
  • Ping.fm
  • Great post as always.

    My question...what defines a friend? Do you mean friend in the true sense of the word or acquaintance?

    That is what I have been trying to wrap my head around.

    Keep in mind I have several Twitter buddies that I feel are like friends....you included, Micah..even tho I have never met them and may not ever meet them.

    That also brings to mind people I meet thru my blog. I am actually fairly close to many people I meet via my blog. The ones I have connected with in person have moved from 'blog buddy' to 'friend'. The ones I haven't met are still at the 'blog buddy' stage.

    It might just be semantics when it all comes down to it.
  • Harrison,

    I think Spokeo is a good social aggregator to date, and I should have included it in my post along friendfeed and socialthing.

    It doesnt tell me whom I should be friends with, nor can I easily search through my friends content.

    But I have been a user for awhile, and I like getting the updates.
  • Great post Micah. I agree that there's a need for some sort of uber friend recommender. I plan to incorporate some aspect of this into the web app I'm working on, although my focus is more on content discovery than contact discovery.

    I like the distinction Last.fm makes between "friends" and "neighbors". I would like recommendations on forming connections with both of those types of contacts. Connecting to actual friends - the people you already know - should be fairly straight-forward. If the system notices, for example, that you are friends with Bob on Flickr and that Bob and you both have a profile on del.icio.us, it should recommend that you friend Bob on del.icio.us. Neighbors are a little more complicated to identify... recommending people who you have no connection to other than some shared interest (ie. you and a stranger happen to bookmark a lot of the same sites in del.icio.us). Oddly, I'm personally more excited about connecting with neighbors online since I have a lot of alternative methods for connecting with actual friends in the real world. For example, not many of my actual friends use sites like Tumblr. Meeting neighbors with those interests online allows me to connect with people online to discuss a topic that is mostly missing from my personal life.
  • @joe, there is a good conversation going on over at http://blog.adaptiveblue.com around friend hierarchies.
  • Micah, thanks for the tip. I'll check that out.
  • yungbitz
    looking for real friends to hold by the hand
blog comments powered by Disqus