Networks have complex topologies and can even exist in many dimensional space. Their boundaries are indistinct, and they change over time. Given this, how do we look at them?
Right now there buzz about "social graphs." In order to do this it is necessary to assume networks as distinct nodes, that ties are concrete and that the network is looked at from the individual's viewpoint. Technically this is called an "egocentric" viewpoint, not because the individual is egotistical, but "ego" means "self" in Latin and sociologists distinguish "ego" as "self" and "alter" as "other." (We could use "self" instead of "ego," but calling someone "self-centered" is not much of an improvement :).
To me, I wonder how important that is to know? We already have our own view of our networks, Perhaps we do not think of it that way, but we are limited in what we can see by where we are. Most of us won't ever really know what Warren Buffet and Bill Gates talk about with their friends at dinner parties; or what the leader of a village in Pakistan talks about with his friends. We are literally in different worlds.
It is important to know ones own network because people that have more diverse networks are healthier, happier, wealthier and many good things, that aspect of it is valuable. By seeing who our connections connect to we can broaden our search. Of course, all kinds of social media is allowing that. Heck, it is pretty cheap to put an ad on Google if we are want to be seen or to find something. So in my mind, the social graph as we are now thinking of it is more cool than useful. This is because it is an ego centric view of the networks
But there are places that can have a network centric view. LinkedIn has its view of the social networks, Microsoft has its view of IM traffic, phone companies have their networks. All of these provide "network centric" views of the network. That means they can see the whole network at the same time. We can see it from anyone's point of view, what could that mean?
We will see groups of people that are not connected, they could not have a view of each others. We can see groups forming and groups disbanding. We can see who in a group has higher status or is closer to what is going on. We can see the connectors and the hubs. Each community is like a geographic region with its own stories and when we see them, we can ask what is going on. From our own view point they are invisible, or blur together like the Milky Way.
Tuesday, May 27, 2008
Sunday, May 25, 2008
Social Network Marketing Goldmine
As I go around Silicon Valley and talk to people, I don't meet that many that are aware of the goldmine that lives in the social network data that is being collected. Most of us see the interfaces to networks, we see cellphones, computers and other interfaces, but the network inside is out of sight and out of mind.
Several companies, including Microsoft, are making software that helps visualize the networks. They draw pretty pictures of the networks and we can get a sense of how connected we are or the world is in general. These are sometimes called "social graphs," but it is the latest generation of sociograms which were invented in the 1930's by a psychologist named James Morano. There has been wonderful software available for more than 10 years to visualize and analyze networks.
The key here is "analyze." What does that mean? There is a lot to it and over time I will add bits and pieces to this blog, but the point I want to make that is not obvious relates to what I previously wrote about 'roles." We are both who we are inside and who we are expected to be by ourselves and others. It can be difficult to see, but who we are, to some extent (and I think a very large extent) has to do with what others expect of us over the years.
We talk about things like "culture" and have a sense that people who grow up in different environments appreciate different things and have different ways of behaving. It is worth noting the "different environments" portion of this. What is an environment but a set of relationships located in one part of the social world? The key here is location. If we know where someone is located in social space, we can make some sophisticated guesses about who they are. Their culture is a consequence of their location so knowing their location can be a proxy for culture. In fact, perhaps "culture" is just a word we used because we could not see the networks. Now, with cell phone data, IM data, social networking data, email traffic data and so on, we can trace the patterns of connections between people. But, is that enough?
The links in "small world networks" like these can be very useful, but they are not the whole story. We are members of multiple networks. Some large organizations are creating "social networking" software that they can use internally. In some more progressive organizations they allow users to create their own social networks. In the US for all of this is owned by the companies. In principle, it is possible to see individuals as members of multiple social networks, which is the optimum situation for knowing who that person is. But even in flat, small-worlds networks, it is possible to find structural equivalence (or its variant regular equivalence) or people that, because of the patterns of network ties in which they enmeshed look at themselves and are looked at as similar.
Old marketing does segmentation of society by attributes, age, gender, education, race and so on. But, if you think about it, the relationship between age and music taste is a coincidence. Because people listen to music of their age group and they all grow older at the same time, there will be age graded bands of musical taste. But it is the relationships that "caused" the musical taste, not the person's age.
Innovation spreads through structural equivalent sets. People have a tendency to adopt ideas or do thing that people they think of themselves as similar do. Status, prestige and so on are all relative to who we think we are. The radical thing that I am saying is that we only get to make up part of who we think we are. We compare ourselves and are compared, just to know who we are. Networks can show us who is similar, who is different, who associates with each other, who is thought of as influential and much, much more.
Companies these days need to be taking some steps:
1. Collect data in a form that can be easily analyzed by SNA.
2. Create strategies to have people use multiple networks (this goes opposite of the idea of social networks as a broadcast medium)
3. Choose and implement analysis processes.
4. Work to mitigate privacy and other fears.
Several companies, including Microsoft, are making software that helps visualize the networks. They draw pretty pictures of the networks and we can get a sense of how connected we are or the world is in general. These are sometimes called "social graphs," but it is the latest generation of sociograms which were invented in the 1930's by a psychologist named James Morano. There has been wonderful software available for more than 10 years to visualize and analyze networks.
The key here is "analyze." What does that mean? There is a lot to it and over time I will add bits and pieces to this blog, but the point I want to make that is not obvious relates to what I previously wrote about 'roles." We are both who we are inside and who we are expected to be by ourselves and others. It can be difficult to see, but who we are, to some extent (and I think a very large extent) has to do with what others expect of us over the years.
We talk about things like "culture" and have a sense that people who grow up in different environments appreciate different things and have different ways of behaving. It is worth noting the "different environments" portion of this. What is an environment but a set of relationships located in one part of the social world? The key here is location. If we know where someone is located in social space, we can make some sophisticated guesses about who they are. Their culture is a consequence of their location so knowing their location can be a proxy for culture. In fact, perhaps "culture" is just a word we used because we could not see the networks. Now, with cell phone data, IM data, social networking data, email traffic data and so on, we can trace the patterns of connections between people. But, is that enough?
The links in "small world networks" like these can be very useful, but they are not the whole story. We are members of multiple networks. Some large organizations are creating "social networking" software that they can use internally. In some more progressive organizations they allow users to create their own social networks. In the US for all of this is owned by the companies. In principle, it is possible to see individuals as members of multiple social networks, which is the optimum situation for knowing who that person is. But even in flat, small-worlds networks, it is possible to find structural equivalence (or its variant regular equivalence) or people that, because of the patterns of network ties in which they enmeshed look at themselves and are looked at as similar.
Old marketing does segmentation of society by attributes, age, gender, education, race and so on. But, if you think about it, the relationship between age and music taste is a coincidence. Because people listen to music of their age group and they all grow older at the same time, there will be age graded bands of musical taste. But it is the relationships that "caused" the musical taste, not the person's age.
Innovation spreads through structural equivalent sets. People have a tendency to adopt ideas or do thing that people they think of themselves as similar do. Status, prestige and so on are all relative to who we think we are. The radical thing that I am saying is that we only get to make up part of who we think we are. We compare ourselves and are compared, just to know who we are. Networks can show us who is similar, who is different, who associates with each other, who is thought of as influential and much, much more.
Companies these days need to be taking some steps:
1. Collect data in a form that can be easily analyzed by SNA.
2. Create strategies to have people use multiple networks (this goes opposite of the idea of social networks as a broadcast medium)
3. Choose and implement analysis processes.
4. Work to mitigate privacy and other fears.
Tuesday, May 20, 2008
Some papers I have written
I have several papers, some peer reviewed, that are avaialble at:
http://www.steiny.com/writings/
http://www.steiny.com/writings/
Roles and networks
One of the most important aspects of networks is that of roles.
Roles have been widely discussed from a number of points of view over time. One that some people go to right aways is roles as attributes of an individual with a psychoanalytic basis. For instance, roles as archetypes, like "the warrior" and so on. In this theory, to understand ourselves we need to know the roles that are fundamental to our character. The notion of role that we take (network folks) is almost the opposite. We take the position that there are countless roles in society and that they exist whether we occupy them or not.
When you first meet someone, or anytime you are interacting with you, what are they interacting with? Are they interacting with you lips? Your immortal soul? Your mind? Specific neural patterns in your brain?
Depending on your beliefs and the level of analysis, possibly all of the above, but if you meet someone new and you ask yourself what type of a person you are meeting, a story about them will come to mine. As Gladwell points out in "Blink" the decisions we make about others are instant and unconscious. A more practical way of thinking about it is that somehow you create a story about that person and they about you.
So, what are the elements of that story? Where do the stories come from? Could you tell your story to someone else, in other words say it in language? Probably you could. That presupposes that they have a shared understanding of what is possible. The stories don't come from nowhere, they are given to us by the communities in which we are involved. Our mothers, your friends, the social world around us. We are born into stories and explanations.
One thing that "exists" in the world outside of us is "roles." It is easy to see that the authority of, say a police officer, is not because of the individual that has the role, but the social weight of the role. Anyone who is a police officer can arrest people, carry a gun and so on. There are more or less sensible ways of talking to a police officer and a police officer is trained in having the bearing of a police officer.
Suppose someone wants to be an accountant. He or she goes to school and learns to do what accountants do. Though there can be some creativity, the domain is small. And an account needs to have the kind of office an accountant has appointed with the appropriate equipment and furniture. The accountant needs to wear the appropriate clothes, drive the appropriate car and have the appropriate friends. Society has many ways of letting us know when we are not following the rules. If you ask the accountant "what are you?" He or she might answer "an accountant." But, what part of that did he or she have any choice in?
Of course, few of us have a single role. We surely are sons or daughters, we may be parents, golfers, leaders, social icebreakers, and countless other roles. These roles to some extent have to be agreed on by others. We fill the roles, we cannot make them on our own. We can move from role to role, but the roles will exist even if we do not fill them.
Later on I will talk about more subtle aspects of roles. The subtely has to do with multiple networks. The role of a father might be seen differently in different networks. But that will come in a later entry. What I want to emphasize here is that when we interact with each other, we are, to some degree interaction with a socially influenced expectations by both parties. You can think of highly formal diplomatic situation where both parties are the roles and every word spoken is scripted to be the words of a prime minister or president. We hope that a judge will act in his or her role, not from his or her feelings about a situation. It is not difficult to see that roles strongly influence the way we are seen and the way we act.
So the idea is: what if we could know people's roles we could know a lot about how they will act.
Roles have been widely discussed from a number of points of view over time. One that some people go to right aways is roles as attributes of an individual with a psychoanalytic basis. For instance, roles as archetypes, like "the warrior" and so on. In this theory, to understand ourselves we need to know the roles that are fundamental to our character. The notion of role that we take (network folks) is almost the opposite. We take the position that there are countless roles in society and that they exist whether we occupy them or not.
When you first meet someone, or anytime you are interacting with you, what are they interacting with? Are they interacting with you lips? Your immortal soul? Your mind? Specific neural patterns in your brain?
Depending on your beliefs and the level of analysis, possibly all of the above, but if you meet someone new and you ask yourself what type of a person you are meeting, a story about them will come to mine. As Gladwell points out in "Blink" the decisions we make about others are instant and unconscious. A more practical way of thinking about it is that somehow you create a story about that person and they about you.
So, what are the elements of that story? Where do the stories come from? Could you tell your story to someone else, in other words say it in language? Probably you could. That presupposes that they have a shared understanding of what is possible. The stories don't come from nowhere, they are given to us by the communities in which we are involved. Our mothers, your friends, the social world around us. We are born into stories and explanations.
One thing that "exists" in the world outside of us is "roles." It is easy to see that the authority of, say a police officer, is not because of the individual that has the role, but the social weight of the role. Anyone who is a police officer can arrest people, carry a gun and so on. There are more or less sensible ways of talking to a police officer and a police officer is trained in having the bearing of a police officer.
Suppose someone wants to be an accountant. He or she goes to school and learns to do what accountants do. Though there can be some creativity, the domain is small. And an account needs to have the kind of office an accountant has appointed with the appropriate equipment and furniture. The accountant needs to wear the appropriate clothes, drive the appropriate car and have the appropriate friends. Society has many ways of letting us know when we are not following the rules. If you ask the accountant "what are you?" He or she might answer "an accountant." But, what part of that did he or she have any choice in?
Of course, few of us have a single role. We surely are sons or daughters, we may be parents, golfers, leaders, social icebreakers, and countless other roles. These roles to some extent have to be agreed on by others. We fill the roles, we cannot make them on our own. We can move from role to role, but the roles will exist even if we do not fill them.
Later on I will talk about more subtle aspects of roles. The subtely has to do with multiple networks. The role of a father might be seen differently in different networks. But that will come in a later entry. What I want to emphasize here is that when we interact with each other, we are, to some degree interaction with a socially influenced expectations by both parties. You can think of highly formal diplomatic situation where both parties are the roles and every word spoken is scripted to be the words of a prime minister or president. We hope that a judge will act in his or her role, not from his or her feelings about a situation. It is not difficult to see that roles strongly influence the way we are seen and the way we act.
So the idea is: what if we could know people's roles we could know a lot about how they will act.
Tuesday, May 6, 2008
What's this got do do with social networks?
Entrepreneurs the world over are trying to figure out how they are going to make money on "social networks," sites like Facebook and standards like Open Social. I take the view the Economist magazine takes, that social networks will become like air. Think of a few technologies of the past. One is ISPs. In 1995 there were many ISPs and they could even make money but as it became a commodity service people would could not scale disappeared. Web designers are as common as rain these days, but it is a marginal, life-style business, though in the late '90s there were thoughts of mega-design companies. Now Web design has been incorporated into business and personal communication and it is another kind of marketing design. There are special things as new technology like ajax and such open windows for programmers, but it has gone the way of specialized word processing, it is something almost anyone can do to some degree.
What is the outcome a social networking company needs to accomplish to be successful? Several CEOs of well-known social networking companies told me it is "the size of the network.' Given my interest in social networks, I can't help but agree. However, the part of me that is a businessperson thinks "if that makes money." After all, unless those network members represent revenue, they are just a drain on the company. In fact, as the Economist has pointed out, social networks are not like phone networks or the Internet. Maintaining a connection with a person has a cost and after a certain point larger networks are less valuable than smaller ones.
How can social networking sites make revenues? Pretty much the main way people have figure out is advertising. When would that work best? If there is specialized social network for, say travel, then it would make sense to advertise travel related stuff there. For a site like Facebook, where it is used for a variety of purposes, it is more difficult for advertisers to target interest groups. One strategy that people are trying is to trace the diffusion of addin modules, which is based on the theory that there are "influenctials" that can be discovered and marketed to. However, a recent article by Duncan Watts calls into question if that is really what is going on. My opinion is that it is not terribly useful information. Having someone take the time to install and distribute a virtual vampire application is probably and indication of how much free time someone has on his or her hands rather than any special influence the person has.
What is the outcome a social networking company needs to accomplish to be successful? Several CEOs of well-known social networking companies told me it is "the size of the network.' Given my interest in social networks, I can't help but agree. However, the part of me that is a businessperson thinks "if that makes money." After all, unless those network members represent revenue, they are just a drain on the company. In fact, as the Economist has pointed out, social networks are not like phone networks or the Internet. Maintaining a connection with a person has a cost and after a certain point larger networks are less valuable than smaller ones.
How can social networking sites make revenues? Pretty much the main way people have figure out is advertising. When would that work best? If there is specialized social network for, say travel, then it would make sense to advertise travel related stuff there. For a site like Facebook, where it is used for a variety of purposes, it is more difficult for advertisers to target interest groups. One strategy that people are trying is to trace the diffusion of addin modules, which is based on the theory that there are "influenctials" that can be discovered and marketed to. However, a recent article by Duncan Watts calls into question if that is really what is going on. My opinion is that it is not terribly useful information. Having someone take the time to install and distribute a virtual vampire application is probably and indication of how much free time someone has on his or her hands rather than any special influence the person has.
Monday, May 5, 2008
Social network -
In this entry I am going to talk a little bit about the relationship between social networking and social networks as Web 2.0 communications platforms and social networks as networks of humans irrespective of platform.
Social networks as I think of them are the networks of relationships in which we all live. In my experience, people I talk to that think about networks think about them from an "ego centric" perspective. I am not using this term as a pejorative, it is a standard term in networks that means "from my point of view." In the picture on the right, that dot way off in upper half on the extreme left edge would have trouble getting information about the folks in the lower right areas because there are so many others he or she would have to go through to get it.
While that person is an extreme case, only talking to a single other, no matter where a person is in the network he or she best sees the immediate area around and there are distant parts of the network he or she cannot contact directly. This is what I mean by an "ego centric view," the view from our particular place in the network. Of course we, magical beings that we are, are looking down on the whole network and can see all of the relationships. We have a "network centric" view, we can see the whole network. This view gives great and special powers to know more about others that are obvious because we have not been able to look at ourselves this way before. Powerful new tools and ways of thinking about the world have allowed us to better understand and describe how our social relation create us and makes us who we are.
Subscribe to:
Posts (Atom)
