[PREV - MELEE]    [TOP]

SOCIAL_NET


                                             August 5, 2010


                                                             KADUSHIN
   "Introduction to Social Network Theory"
    by Charles Kadushin (2004)

  I got interested in Kadushin's paper
  for a number of reasons.

  His application of network theory to
  human relationships looked a lot             BRENNER_6
  like a method I worked out for
  plotting stories when I was a kid.

  And the phrase "social network" is
  of course a big deal in the web world...
  might there be some aspect of Kadushin's
  review that would apply to it?


  One of Kadsushin's first principles
  is "propinquity": you're more likely
  to be connected to people physically          He points out that economists
  near you.                                     tend to use transportation
                                                costs rather than distance,
    Geographic proximity would not              citing "Krugman and Obstfeld"
    seem to be the deciding factor              (2000).
    in the internet era...

    But then, isn't it possible it
    matters even on the web?  The
    "world wide" web could be a source
    of connections from anywhere, but
    if people tend to "*friend*"
    people they talk to in-person,
    your "*friend* network" will tend
    to mirror your off-line network.

       You could define a parameter,
       a ratio of face-to-face "facebook"
       "*friends*" and ones you've never         And there's a third
       met before connecting with on-line.       class of people,
                                                 the ones you met
              Some of those connections          on-line who you
              will be people who have            later arranged
              connections with people            meetings with off-line.
              you *have* met off-line
              first.                                 (face the faceless
                                                      hordes of facebook)
         So right there we've got
         a possible field of study,
         something the marketeers
         are no doubt greatly
         interested in.




  Early on Kadushin emphasises that there are
  different kinds of possible connections,
  something our "social network" sites have
  had trouble grasping.

                                          "Proposition 2 c. Individuals or
     He has a number of propositions      groups with homophilous relations
     about "homophily": e.g. you          are likely to share similar
     tend to have connections to          attitudes."  -- Charles Kadushin,
     other people who are like you.       "Social Network Theory", Feb 17, 2004

     (Depending on the kind of
     connections under discussion,                      Networks
     and what kind of enitity "you"                     within
     are-- his nodes are allowed to                     networks.
     be businesses or other
     collectives).                                          (Are their
                                                             networks
         This subject, by the way, has a lot                 within
         to do with theories of city formation.              humans?)


         Different parameters of
         network "distance" can               Note: sociograms and sociometry--
         be defined:                          terms invented by Jacob L. Moreno
                                              in 1934 (wrote about it in 1953)
         Links can be weighted
         and summed, so that
         multiple first-order
         connections count much
         more than one multi-hop
         connection.                   Stanley Milgram did some
                                       "small world" studies in 1967.


    first order zone = interpersonal environment = graph theory

Kadushin claims that organizations that are
*structurally* similar tend to behave in similar
ways, though there's no direct communication
between them.

   (But there's no reference on that one... and
   I wonder what he means by "structure" in this
   context.  Flat vs deep hierarchy?)

   There's a suggestion about
   individuals defined by their                  He chatters a lot about
   relationships (no monster                     intersubjectivity and so on.
   lives on Decartes' island self).              Seems a little fuzzy, but it's
                                                 worth remembering that there
                                                 can be emergent properties of
                                                 a network that are hard to
      Perhaps there's an unexamined              grasp.  What we're really
      assumption in the notion of a "node":      after here may turn out to be
                                                 fuzzy.
      The node is a hard, atomic monad
      in our network diagrams, but
      nodes can be complex entities
      with internal structure, and
      controversial boundaries.

            Systems set-up for
            individuals sometimes
            end up treating couples
            as individuals when they
            choose to share an                  NODAL_SELF
            account, etc.

                 A really tight connection
                 between individuals might
                 be better thought of as a
                 joint node that needs to be
                 treated as a one entity...

--------
[NEXT - INNER_TUBES]