How stuff spreads (4) the bottle-glass trick

Posted by on May 8, 2008 in Uncategorized | 4 Comments


Great new format for a breakfast meeting yesterday – hosted by the redoubtable Steve Moore.

Interesting chats by Jeremy, Kevin and Matt, each attempting to define “what’s next?” in their respective worlds. Lots of great insights and ideas. Wish I’d had a tape running.

That said, the story that stays with me is one Matt told of some current work C4 are doing about how teenagers are using media today.

Seems one girl the researchers were following was hanging out online doing amongst other things a spot of the hi-speed Instant Messaging that only the young can really manage for any length of time.

She had sorted all her contacts into 6-7 or seven groups – schoolfriends, family etc but also “bitches” “wankers” and so on. What was striking though was the way in which she switched contacts between the groups in real time. Even if the members of her different social networks remained mostly consistent over the short term, their roles were in constant flux. And those are just the small set of folk she is in regular contact with regularly…

Something to bear in mind the next time you here talk about how stuff spreads through social networks and how there are some folk through whom information and influence flows. If only it was that simple…

In reality, it’s like watching the great Tommy Cooper do – in his inimitable shambolic stylee – the trick here.

Things are – or seem to be- constant motion…whenever you lift the cover, the thing you expect is not there. Bottle or glass?


  1. Helge Tennø
    May 8, 2008

    If I understand you’re point correctly, then Duncan Watts’ research would be interesting as a comment to this post.
    Duncan is enriching Malcolm Gladwells book: Tipping Point (and traditional advertising “influencers”-theory) by saying that there are influencers, but they are not the same every time, they differ according to the theme and context. I think Duncan describes this as the adhesiveness of an idea in the social and cultural context, rather than regular people spreading good stuff regularly.
    But, the most interesting part of the article ( is Watts’ reference to a social experiment he did with 14.000 participants, logging into different website, listening to, and purchasing music.
    Now some groups of people where just asked to find and buy some music, these groups showed no theme or preference.
    The other groups where asked to rate the songs first and then buy them. In these groups there was a clear show of preference.
    Now the interesting part was that, there were no similarities between the songs favored or loathed in any of the different groups rating the songs. It seemed totally random which songs got to become popular or not – dependent on the taste of the first people starting to rate.
    This would mean that adding a feature to your web service allowing participants to participate would increase the preference of certain songs, it just seems totally random which, and would be very hard to control.
    Just a thought, hoped it matched your original content :o)

  2. david cushman
    May 8, 2008

    Perhaps the roles remain consistent, the people shift? That’s the point i think citizen dave was making here:
    (see comments):

  3. Mark Earls
    May 8, 2008

    Thanks Helge.
    I’ve not just blogged about the Watts work ( but also presented alongside him ( The work I and my colleague have been doing demonstrates the essential point of Duncan’s work: that there is no real structure to how stuff spreads, that there is no “influencer” type. Or indeed individuals who “influence” the rest of us…
    Indeed the point is that this is looking at the game the wrong way round: why do we do insist on looking at things from our perspective as opposed to the folk out there.
    So, dear Dave C – welcome back! – that’s the reason why the roles being the same or not is precisely wrong – it’s the “influenced” rather than the “influencers”, EMULATION rather than influence that matters. More pull-me than push-you…
    Today’s post was intended just to unpick the delusion of the fixed social network. Glad it stimulated you both!

  4. Jordyn
    May 17, 2008

    Making the essential concept of the “influenced” be the irresistible force and the “influencer” be the irresistible object is just so much fun, figuring out how to play with that (or even if it can be played with) has been occupying my mind ever since reading your book.