CCK11: Network Anonymity

I’m familiar with network analysis, especially in social sciences research, so most of the concepts from Week 2 of CCK2011 made sense. Different words (nodes versus vertices) are used in different fields to describe the same basic ideas, and once I got a handle on that I was fine with the readings and discussion. In some contexts, though, I’ve noticed that writers tend to discuss networks as if they are static, whereas I see them as incredibly dynamic and constantly mutating. The challenge is to work this into your analytic tools, so that the diagram of your network changes as the network fluctuates and re-forms. (Something akin to Hans Rosling’s reworked bubble charts, that show change over time [obviously he’s not talking about networks; I’m just interested in the illustration of time in a chart].)

Research companies I write and edit for have discovered important insights by using network analysis (e.g., how health knowledge can purposefully filter through a particular community). Where real people in physical proximity are concerned, I get it.

Where machines and online personas are concerned, however, I am somewhat unsure.

Here’s one example: we don’t let my preteen stepdaughter “friend” anyone on Facebook that she doesn’t know in “real life.” Still, I get a little cranky when she announces to her 122 “friends” that we’re going out of town for the weekend and that we bought a new flat screen television. Who else are her friends are connected to that I don’t know? How many other people use the same physical computer as her friend? There are nodes in this network of which I am completely unaware (though clearly I “sense” their presence).

How to account for those anonymous nodes and their influence on the network and the other nodes within it? I’ll put aside for the moment the question of Those Nodes That Mean Me Deliberate Harm—that’s sort of an extreme example. If knowledge is connections, then it seems that a catalyst of learning (as vjansen describes in a course discussion thread) could be the activation of an anonymous node (you could take this to the Rupert Sheldrake level, too, or talk about the family constellations work of Bert Hellinger). For instance, my husband is an eighth-grade science teacher, and his stories about his students viscerally tamp down my enthusiasm for the Amazing, Uplifting, Astounding Potential of Technology in Education a bit because of the pressing issues he deals with on a regular basis that interfere with students’ learning (say, unplanned pregnancies at age 13).

Another example might be the anonymous machine nodes that collect personal information in order to sell me products I might be interested in based on my past purchases (“you might also like…”). They’re mostly hidden to me but certainly their influence is there: on Amazon, I see one book and not a different one in this section, and I might read it and then recommend it to a friend or colleague in my network. That hidden (even to me) node affects my colleague directly and potentially her colleagues.

George Siemens writes 

In a networked world, the very manner of information that we acquire is worth exploring. The need to evaluate the worthiness of learning something is a meta-skill that is applied before learning itself begins. …  The ability to synthesize and recognize connections and patterns is a valuable skill. (Siemens 2005)

What does connectivism do with these anonymous network nodes? Does evaluating something’s “worthiness” rest on the premise that all of our connections are known to us and the network?  Couple this with the idea that those networks are in constant flux, and analyzing how learning happens gets even trickier.

7 thoughts on “CCK11: Network Anonymity

  1. Theodore A, Hoppe says:

    Re: “What does connectivism do with these anonymous network nodes?”
    “Unknown, (anonymous) to whom?”, is the relationship to the your question there I will be exploring. The research of Nicholas Christikas and James Fowler, authors of the book, “Connected” demonstrate that the identity is irrelevant for the most part. Starting with Milgram’s concept of the ‘Six Degrees of Separation,’ where only half of the participants may be directly known to one another, Christakis and Fowler describe their theory of the ‘Three Degrees of Influence.’ This is where we have a direct influence on our friends, who are on course know to us, but and also have an influence on our friends friends, who we may or may not know. Additionally, they provide evidence that we also have influence on the friends of our friend’s friends, many of whom we will not know.
    An example of this is illustrated by how obesity can travel through a social network. In the TED talk, The Hidden Influence of Social Networks, Christakis demonstrates this point with a graph and explains: “So we did some mathematics to study the size of these clusters. This here shows, on the Y-axis, the increase in the probability that a person is obese, given that a social contact of theirs is obese. And on the X-axis, the degrees of separation between the two people. And on the far left, you see the purple [bar]. It says that, if your friends are obese, your risk of obesity is 45 percent higher. And the next bar over, the [red bar], says that, if your friend’s friends are obese, your risk of obesity is 25 percent higher. And then the next [bar] over says that, if your friend’s friend’s friend, someone you probably don’t even know, is obese, your risk of obesity is 10 percent higher. And it’s only when you get to your friend’s friend’s friend’s friends, that there’s no longer a relationship between that person’s body size and your own body size.”

    He allow takes into account the idea of a “network in flux” as you state by adding: “In addition, because obesity is not a unicentric epidemic, there’s not a “patient zero” of the obesity epidemic — if we find that guy, there was a spread of obesity out from him. It’s a multicentric epidemic. Lots of people are doing things at the same time.”

    I hope this helps. You may contact me at with any question

  2. leahgrrl says:

    The question I was trying to get at was specifically about what the theory of connectivism, which is based on the presupposed existence of networks, has to do with the question of hidden or anonymous nodes. I’m familiar with some of the research; I do quite a lot of reading in a variety of academic fields outside of my own. The question of hidden nodes comes up in neurology, in computer science, genetics, and others fields studying complex networks.

    Mathematically, researchers in these areas adjust their formulas to control for the effects of nodes that they don’t know about. If A and B have a relationship–like Stephen’s chart–and we make some sort of simple probability equation of the likelihood that a node “fires” as a result of that relationship, it seems to me that we’re missing something. We can’t entirely attribute the effects of A-B relationship to that relationship; there’s always the lingering question about what other potential “causes” exist.

    Your example of obesity in a network above, is exactly what I am thinking about in relation to learning. That is, the existence of this network effect is not what I’m querying. I’m wondering whether connectivism as a theory can account for the effects of these hidden nodes. It’s not that connectivist theorists need to know the “identity” of these “people” (and some are not people)–they need to figure out how to account for them if they posit that knowledge/learning is a network function. I think my fairly simple examples above didn’t quite get at the question I was asking.

    Thanks for your response! I will enjoy looking at the TED talk.


  3. jaapsoft2 says:

    Fascinating questions.
    I wonder if your questions will become even more complicated when both information nodes and social nodes are part of networks. A book or video or other information source is not primarily a social node.
    On your question about learning theories that can be used to explain the working of ‘hidden nodes’ this: (social-)constructivism cannot account for this effects and influences, it does not have a view on networks. Connectivism shows us the networks and nodes and because of that we can talk about effects of (hidden) nodes.
    May be this is a way to an answer? ‘Hidden nodes’ are kind of sources of information. This information is not hidden when I notice it. If my friends friends friends share information and this information does influence me. Than the node is hidden, but I could read the information.
    regards Jaap

  4. […] reply on comments in […]

  5. jaapsoft2 says:

    I did try to make a picture of hidden nodes in a network, published <a href=""<weblog

  6. […] kind of harkens back to a blog post I wrote previously, about hidden nodes. In many equations, a random error or hidden variable is included to account for stuff that the […]

Leave a Reply

Your email address will not be published. Required fields are marked *