Erik van Bekkum: SNA

Visualization of engineering community of practice

Sun, 09 Jul 2006 22:23:54 GMT

I just finished a round of reporting for a global engineering community. Attached is part of the report, which is the visualization of engineering community of practice (Netdraw, >300 nodes). In subsequent posts I will try to post the evolution of this community by introducing the 'time' factor in the network data, years 2001 - 2006. The labels have been removed.

Friendster acquires patent for social networking

Sun, 09 Jul 2006 14:13:14 GMT

Since Red Herring published the article "Friendster Wins Patent" last Friday, 7th of July, it only took 48 hours for more than 26,000 blogs on social networks to pick the news up. Most of the sites report the same news item: "as the generic patent offers opportunities to the once-beleaguered site to pursue licenses from similar social networks and competitors that offer the same services based on the idea of connecting people within a certain number of degrees of separation, Friendster president Kent Lindstrom tells Red Herring that “it is still too early to say” although Friendster “will do what we can to protect our intellectual property”. Check the patent out here. The problem debated around is of course that the foundation of this patent is the "small world theory" by Stanley Milgram.

It is argued that with the patent, Friendster can put itself up for sale, significantly increasing its value. This story - without a doubt - will be continued.

Social network Forum event wiki

Sun, 09 Jul 2006 13:21:03 GMT

The 2nd Social network Forum, which was organized by the Leeds University Business School and which was held from the 30th of June until the 1st of July, 2006, has finalized. Summaries of the presentations can be found online in the forum wiki, but the slides and proceedings have not been posted (yet) online.

Networks uncovered using Ontology Network Analysis

Mon, 17 Oct 2005 20:24:56 GMT

Ontocopi lets you infer the informal relations that define a community of practice from the presence of more formal relations. For instance, if A and B have no formal relation but they have both authored papers with C (formal relation), they might share interests (informal relation). Because Ontocopi works in this way, we cannot claim without qualification that it identifies communities of practice. Significant informal relations might have little or no connection to the formal ones. Here, we refer to the networks uncovered by Ontocopi as COPs and to informal social networks as communities of practice. We work under the assumption that COPs are sometimes decent proxies for communities of practice.

More background information for reading here (PDF paper)

Organizational Network Analysis and communities

Mon, 10 Oct 2005 21:56:32 GMT

A common objective for any CoP program is to encourage information flow, knowledge reuse, and learning among employees.This informational focus derives from early scholarship on the situated nature of learning and problem solving in communities.However, from a purely practical perspective, substantial efficiency and effectiveness benefits result from communities that promote effective knowledge creation and transfer. Unfortunately, in new communities, we typically see information flow and learning networks that are constrained by formal structure,homophilly, and to some degree personality or interests of those involved.These social forces create silos and a wide dispersion of connectivity that undermine knowledge transfer and performance benefits of communities.

Rob Cross in Assessing and Improving Communities of Practice with Organizational Network Analysis made by his team of the Network Round Table at the University of Virginia; the application of Organizational Network Analysis (ONA) on communties of practice at fifteen participating organizations of the NRT. There's a couple of more resources there too.

Social networks and country-to-country transfer: dense and weak ties in the diffusion of knowledge

Sun, 02 Oct 2005 12:17:27 GMT

This article from Marie-Laure Djelic investigates the social network dimension in processes of cross-national transfer. The empirical focus is the conscious attempt to appropriate, in France after 1945, the American model of the large firm. Structural conditions—internal crisis and geopolitical dependence—created the context in which country-to-country transfer could take place. Our findings also show, however, that the transfer itself required the activation of concrete mechanisms and, there, social networks proved key. Our evidence shows in fact the tight and reciprocal interaction, the co-construction, as it were, of social networks on the one hand and processes of institutionalization on the other. Building upon our empirical findings, we propose furthermore that successful cross-national transfer hinges on a particular kind of network structure. In the story recounted here diffusion across national borders called for the smooth and successful articulation of two types of social networks—a cross-national “weak ties” network and national “strong ties” ones. In the end, this article accords with the current calls for cross-fertilization of institutional theory and social network theory. And we argue that both approaches are useful and complementary when dealing with country-to-country transfers.

Download the article in PDF here.

Yahoo! patents (parts of) social networks: Method and system for customizing views of information associated with a social network user

Wed, 17 Aug 2005 14:50:03 GMT

US Patent Office, application 20050177385:

A method, apparatus, and system are directed towards managing a view of a social network user’s personal information based, in part, on user-defined criteria. The user-defined criteria may be applied towards a user’s relationship with each prospective viewer. The user-defined criteria may include degrees of separation between members of the social network, a relationship to the prospective viewer, as well as criteria based, in part, on activities, such as dating, employment, hobbies, and the like. The user-defined criteria may also be based on a group membership, a strength of a relationship, and the like. Such user-defined relationship criteria may then be mapped against various categories of information associated with social network user to provide customized views of the social network user.

I caught this one through Ojo. What's next?

Social Network Analysis at BHP

Thu, 28 Jul 2005 23:24:17 GMT

There are very few papers on the usage of SNA for knowledge networking in the mining / industrial minerals industry. Some (undocumented) work has been done with Borax, but the most obvious one is a 2000 report from Laurence Lock Lee. In his paper, "Knowledge Sharing Metrics for Large Organisations" which can be downloaded here he provides some insight in the structure and creation of the BHP GMN (Global Maintenance Network), founded back in 1997.The Global Maintenance Network is formalized, and very much structured (facilitated, too), and according to the paper not a community of practice, which is largely unstructured.

The rationale for a facilitated network over a pure community of practice is the business need to meet performance targets and deadlines, which are less of a driver for communities of practice.

BHP has started a very simple yet (seemingly) effective social network mapping of the networks, using sociometric parameters. The fact that it was limited was not a problem, because it still enabled the assessment of the impacts of business unit membership (formal structure) and geographic location on network activity patterns. The paper itself reveals some of the more obvious findings, and some paths forward to overcome the hurdles of data collection for SNA through surveys.

Intervision and communities of practice

Wed, 27 Jul 2005 09:16:44 GMT

Yesterday we participated again in an informal interdepartamental (governmental) KM community members, where the central theme was intervision. Intervision is a problem-solving technique where employees in a team or group can request their colleagues to collectively think about problems. It is analytical more than solution-oriented, because the participants in the intervision are not requested to bring solutions to the table but ask questions about the context, background and approach of the problem.

Intervision is mostly based on self-reflection and collective capability development. It is very practical.

In many senses intervision looks like work that is common practice in healthy (problem-solving) communities of practice - bring people together with a common interest or within a knowledge domain, focus on the communication between the members, address the reflection and support between the participants, the (management) dilemma of making intervision part of the strategy and HRM, issues with roles / responsibilities and confidentiality, the participation of management - or not, etc, scope and boundaries of the personal and collective, etc.

In the past I have seen several problems with just intervision as a technique within the knowledge management domain;

Intervision is very much problem-oriented which limits it applications for (for example) knowledge stewarding. This also makes it hard to link up intervision groups to other groups, to form larger networks of people.
Intervision is not typically not generative;
Intervision requires the definition of an upfront structure by a leader. This makes it sometimes difficult (depending on the dominance) to distribute leadership;
Intervision is aimed at the improvement and renewal of existing processes, but does not address innovation because the domain of participants is limited to those within a (set) group of experts or even team members.
Intervision requires some sort of guidance (as argued by Jeroen Hendriksen, author of Intervision)

It can be argues nevertheless, that these weaknessess can be turned into positive qualities when the intervision approach is aligned with the strategy and implementation of communities of practice (bringing out some of the qualities and addressing most of the points above) and social network analysis (to support the point of group / team boundaries and overcome the limitation that intervision is more renewal than innovative. For example, SNA could show multiple intervision-group membership or informal intervision groups, similar to work done for communities of practice). The guidance aspect then remains an open-ended question, but it can very well be combined with a higher-level support that is typically also available for institutionalized communities.

Weak ties in communities of practice

Wed, 13 Jul 2005 22:26:57 GMT

Today I had to lead a very interesting and long (free and unfinished) discussion about the importance of strong ties and weak ties in a networked organization, especially in the context of communities of practice. "Communities" so it was argued don't build social capital because they mostly comprise of people with strong links. We used the following definition of social capital

Social Capital = The process and conditions of social networking among people and organizations that lead to accomplishing a goal of mutual social benefit, usually characterized by trust, cooperation, involvement in the community, and sharing.

Weak links are important because of their bridging function to other networks, their capability to leverag the social capital of these networks and bring in new ideas at a lower cost (these last words were put aside on a different note paper, as it lead to separate discussions about the cost of maintaining relationships between strong / weak ties, which is high for strong ties and low for weak ties). The question is whether weak ties can, when they bring in these qualities to the community, substantially contribute further to the capability development of the community participants with strong ties.

Suggestion & starting point for the next session: It may be, when strong ties attract weak ties as the community develops.

Corporate SNA in CIO magazine

Wed, 13 Jul 2005 20:04:02 GMT

In CIO magazine, Susannah Patton writes an introductionary article about the use of social network analysis (SNA) in the corporate world. It's fun for reading and brings forward some of the real challanges that corporate SNA face. Her point, based on an interview with Masterfoods USA:

Companies that have been frustrated by traditional knowledge management efforts, such as Mars, are increasingly looking for ways to find out how knowledge flows through their organizations. Looking at the company org chart, it turns out, often doesn't tell the real story about who holds influence, who gives the best advice and how employees are sharing information critical for success

Read the whole article

Introduction to UCINET and social network methods

Sun, 19 Jun 2005 11:46:48 GMT

Mark Riddle and Robert Hanneman from UCR have published their introductionary textbook on social network methods and graph theory online. The book can be download for free (68 MB) or seen in an HTML version here.

For those getting started with network analysis and especially UCINET as a tool, the introduction can be very helpful. The textbook is very much along the line of the most common usage and operations of UCINET

http://faculty.ucr.edu/~hanneman/nettext/

The navigational aspects of social networks

Mon, 13 Jun 2005 06:14:17 GMT

A very theoretical but interesting paper to read on the navigational aspects of social networks is the 2003 paper from Aaron Clauset and (now Dr) Cristopher Moore, written in their time at the Computer Science Department of the University of New Mexico. It's only a couple of pages and it can be downloaded here.

When they discuss rewiring in a social network they come to these conclusions:

This process has a naturalinterpretation: the topology of a social network is constantly being modified by its members, who update their personal connections as they explore and navigate the network. If a member becomes frustrated because the journey to a destination takes too long, they can be expected to change their connections to make similar journeys more quickly in the future.

It would be an interesting paper or study to follow up on their suggestions:

Our discussion contemplates a “social space” consisting of a finite-dimensional lattice, an obviously poor model for the complex social spaces we routinely navigate. An interesting study would be an analogous rewiring process for networks whose underlying structure is hierarchical, involves multiple group affiliations or is otherwise structured, as in the peer-to-peer network Freenet with the modifications described in.

Chris Moore is now a doctor at the Complexity Research of that same univerisity. Other papers that he wrote that people interested in social networks, complexity and communities of practice may wish to have a look at, are:

Finding community structure in very large networks (PDF)
Finding local community structure in networks (PDF)

I have added these three documents to my library, they're definitely "good to keep".

Social Business Networks and linking up people

Tue, 18 May 2004 12:45:56 GMT

Social Business Network platform Ecademy has positioned their feature to find and meet 'people similar to you' as I recently discovered. As I wrote in the blog on the 28th of April (here) Ecademy uses keywords in the profiles to describe who you are and what you want. Then, the 'find people like me' button can search other profiles and return those who have things in common. The more keywords in common, the higher the score.

It itself not an uncommon social software feature, and to my surprise they have added an automated banner on the homepage which selects (presumably at random) someone from the results list that 'has a profile like me'. The idea is good, yet I have questions about the value of this feature for me. Take this example.

Just now I was recommended Tom Smith because he has things in common with me. When I started to look at his profile, I found that there were just two keywords (of 50) that he shared with me (incidentally, "social software" and "virtual communities"). The other 48 did not resonate very much with me. Going back to the homepage and refreshing a couple of times, other names popped up and I checked their profiles likewise. Very little in common - it seems that

Ecademy is trying to use the power of volume in their attempt to create networks of people rather than quality (or perceived quality I should say). Tom Smith was not even on the list of people that 'look like me' when I forced to show a list - meaning he has less than 17% in common (...) which appears to be very much true.

The quality of this selection process can be perceived as quite shaky because Ecademy does not know which keywords are more important to me than others. I would contact Tom for his 'virtual communities' keyword but not if we shared 'tapas' in common. It's just that these keywords represent who I am and what I want, yet the engine does not discriminate between the two. Splitting the list would be much better. Ryze started off with that quite well (I have...) and (I want..) but did not implement an engine to link people with that. Perhaps it can come still, Social Business Network platforms are on the move and they go very fast.. they just have to stay away from the volume..

Social Learning, Social Networks and Social Knowledge

Mon, 10 May 2004 21:14:53 GMT

At the end of last week there was an interesting moment when in a community of practice Beverly trainer rightfully corrected John Maloney from KM Cluster about his misrepresentation of the term 'social network as a lingua franca for communities of practice (~practicioners). I am glad that she did, because there is an increasing confusion and even misuse of the right terminology and understanding about many concepts. Everything that is "social network"-ed seems to fall victim of this trend, so her correction was reason for me to quote her response. She said:

Am I right that "Social Networks" and "Social Network Analysis" as part of the current discourse of Knowledge Mangagement (with roots in sociometrics and graph theory?) Whereas "Social Learning Theory" is part of the discourse of socio-cultural theories of learning (Vygotsky, Bandura, Lave) and "Social Knowledge" is part of the discourse of sociology of knowledge (Kuhn, Mannheim, Berger, Luckman).

Since she phrased it very well, it's worth having a better look at what she is saying; Social Network Analysis. Social Network analysis is a sociometric, which means that it combines aspects of sociology with mathematical representation using structural models to represent human interactions. [Science Week 2001]: In social network analysis, discrete mathematics and statistics are combined with the emerging epistemology of complex systems to explore processes and phenomena as diverse as the diffusion of information through an organization, the adoption of innovations in society, and the spread of infectious disease in a population. Researchers working on social network analysis draw upon many disciplines: sociology, anthropology, psychology, geography, mathematics, statistics, and computer science.

She continues

Perhaps I misunderstood, but I'd be uneasy if you were claiming that there is A lingua franca which should be used by THE CoP community. I see the landscape more as overlapping, complementary and competing discourses between different (CoP) communities. And in fact, Wenger's research agenda looks very much to me like a search(ing) for a transdisciplinary discourse, and not an affiliation to any one discourse community.

Wenger also comes from the social learning theory side. He and Lave wrote Situated Learning: Legitimate peripheral participation (1991). Because Lave and Wenger put learning in a participative context through communities of practice, they state that "theacquisition of knowledge by individuals so much as a process of social participation. The nature of the situation impacts significantly on the process. More of Wengers work in his new research, where he takes his vision one step further.

Using profile in social business networks (Ecademy)

Wed, 28 Apr 2004 05:54:50 GMT

Though I like the way social business network platforms such as Ecademy use the 'personal profile' to place yourself in the network, there's some considerations I have.

First I need to admit that unlike a community, where your identity is also built through participation and quality of conversation, there is a lot of ongoing work that needs to be done in the profile. You almost start off with a framework and incrementally build a decent profile that matches your business needs and describes your ambitions (whether personal or business).

For this, most of them seem to use keyworks. "Describe yourself in 50 keywords" is a very good exercise to bring down "You" - your identity - down to a couple of elements. I tried to stay away from the business keywords only and introduce personal ones as well. To do this I started off with 15 personal keywords and 15 business keywords. Then I let my associates put in four new ones of each category and take one or two out from the ones I came up with (as long as they are motivated..). I did the same with some friends who were able to describe me better than I myself. The result is that I have a list of words that could describe me from my personal perspective (who I think I am, or want to become) and from the perspective of people that know me (who people think I am, or have been).

But how does one emphasize on one keyword? I would really like to make sure that some things are more important than others with these keywords. As long as I cannot do that, the value of "seach people like me" feature (which looks for people with likewise keywords) is heavily inflated. I am not sure that the description inside the profile will make a difference - so I have included the Ryze'izm "What I have" and "What I need" to make sure that people are aware of what I am looking for and what I can offer, complementary to the keywords.

HP: How to search a social network, Finding Communities in Linear Time

Tue, 20 Apr 2004 18:43:34 GMT

Implicit Structure and the Dynamics of Blogspace and more papers from HP Information Dynamics Lab.

It's always like that: looking for one thing you find many others.

Full paper behind Blog Epidemic Analyzer (for Anjo and Rogier :) - Implicit Structure and the Dynamics of Blogspace by Eytan Adar, Li Zhang, Lada A. Adamic, and Rajan M. Lukose

And other papers from HP Information Dynamics Lab, especially those with titles that I found interesting:

[Mathemagenic]

Analysis of Social Software

Fri, 09 Apr 2004 18:07:10 GMT

Alicia L. Cervini has published on her website a thesis called "Network Connections: An Analysis of Social Software that Turns Online Introductions into Offline Interactions " which takes a look at the modern social software platforms. There is an extensive analysis in the capabilities of these platforms with a touch of social networking theory, but the conclusions are too little founded on the theory of social networking to make a lasting impression. Though it'll be outdated soon, if you are interested now in these platforms and their role in the SN world, check it out at:

http://stage.itp.tsoa.nyu.edu/~alc287/thesis/

Creativity, innovators and social networks

Mon, 05 Apr 2004 22:49:17 GMT

MonkeyMagic reported on Ton Zijlstra's and Lilia's Blogwalk (sorry, circumstances lead me that I could not be there); a perception of direction, trust and creativity

Creativity In lots of ways I think Ton makes good sense. Certainly in terms of picking out those diamond signals from the noise. Ton suggests that

"Knowledge workers ... need to be exposed to as much background noise as possible, to open up as much opportunities to respond as possible."

... and it rings true. It certainly seems to be true in the creative sphere. Amateurs, for example, can and do make great leaps of intuition. (Perhaps because their filters are less restrictive, they get more signals?). Equally artists have a long and fruitful history of opening their doors of perception.

I think that the 'exposure to noise' translates back to the multi-community membership properties that creative people and innovators have; they tend to have access to many networks and know how to translate noise to signals, depending on the direction and purpose of the network. Some other starting point was on the CoP theme on KB last year (with contributions from Dr. Patricia Wolf).

Social Network Analysis (see here for more information on that) is a must when it comes to supporting the conversation with revolutionist. In the article above there are many references to that, but Rob Cross has a nice sample for this application as well (pointer by Patti).

Visualization and management of (personal) social networks

Thu, 01 Jan 2004 13:40:47 GMT

While looking at Wallop's design team, I noticed that one of the designers in the MS Research group, Steven Drucker, has written some interesting social-network related papers from 1999 on the The Social Life of Small Graphical Chat Spaces (PDF), together with Marc A. Smith and Shelly D. Farnham.

This data [..] illustrates the usage patterns of graphical chat systems, and highlights the ways physical proxemics are translated into social interactions in online environments.

The other paper is Visualizations of Collaborative Information for End-Users which illustrates some of Microsoft's early work with social network analysis (?) that, judging from the images and stories about the Wallop project, served its purpose well.

While network interaction media are increasingly popular, there are a number of problems facing their users. The interaction context, or information about the kind of space, group and activity taking place, is often missing or ambiguous in the spaces created by these systems.

Steven's colleague, Shelly Farnham has done quite some work in the social network arena as well and her 2002 paper on "Visualizing Discourse Architectures with Automatically Generated Person-Centric Social Networks" was one of the basis of a report that I read from Essex University: The Generation Gap: Managing technology-mediated personal social networks (Chimera Working Paper, Smith / Rogers / Brady in a PDF document)

Our research has shown that an overwhelming need for all groups is better support for the management of social contact and received content arising out of the mix of online communications people engage in. Furthermore, this should be able to transform what is normally perceived as time-consuming and onerous tasks into ones that are viewed as being more enjoyable pottering kinds of activity.

Social networks deconstructed

Mon, 29 Dec 2003 13:09:11 GMT

Having been idle for the last two weeks, today I found that Jon Lebkowsky, one of the co-founding members of the Social Software Alliance (SSA) is writing two interesting pieces in his blog about social networks

Aesthetics of Social Networks elaborates on the idea that technology must be used to leverage the benefits of the social dimension (social networks) - not as a consequence of its mere existence - but from the design & build stage yet. His ideas for now (though still in early forming and rudimentary stage) are the usage of:

visual aspects (network maps) and
the concept of a social aesthetic (harmonious group-forming).

In another post from the 17th of December, Jon mentiones that Mitch Ratcliffe notes that social network sites aren't as effective when relationships are overstated

At the same time, what is interesting about the social networking technology is that it forces requests into a semi-public venue, making us think about them more explicitly. I'm not sure that does much in terms of making successful connections today, though it certainly makes the "pain" social networking software hopes to address more explicit.

Created connectedness though social networks

Thu, 27 Nov 2003 18:16:04 GMT

Jon Husband in his blog points to an interesting article that appeared in the NY times yesterday on Decoding the New Cues in Online Society, about the power and improper or misues of social network tools - as a semi-interview with Danah Boyd (see her blog, too).

But friendship develops in social contexts, Ms. Boyd says; it doesn't just flow through the pipes of a network. "Just because you're friends with somebody doesn't mean their friends are similar in the type of context you are with your friends," she said. Unless the social networking sites adapt to how people need to use them, she said, the sites will not succeed

Social Networking in Business 2.0

Tue, 25 Nov 2003 13:09:09 GMT

Stowe Boyd talks in Darwin Magazine Online about a social networking article he read in the November edition of Business 2.0. Though there is not much news in his observations of the growing attention to the social network marketplace, he outlines three (in his opinion) critical barriers to its success:

Critical mass (more is better)
Buy in and pay back (WIIFM)
Here versus There (SN integrated in business technology, not separately sold)

Stowe's article once again tells me that the market for social networks is very hot - given the amount of attention it has been having in mainstream business magazines since the summer.

RIAA: Using SNA to make the network weaker

Tue, 19 Aug 2003 10:47:04 GMT

In his blog GimJim wrote an article on the 27th of June about Social Network Analysis and the power of the nodes with high inbetweenness - in a scale free network such as the peer-to-peer networks of Kazaa.

I believe the RIAA is applying social network analysis (SNA) in its current battle against P2P systems. This piece in Time dated 4 August 2003 is telling:

[The RIAA] has declined to say whom it is targeting or how many more subpoenas it plans to issue... "We're focused on the supply side," RIAA president Cary Sherman says. "If you can get at the 10% of people who are offering 90% of the files, that makes a significant dent."
However, that comment has revealed their game plan. The P2P network that is Kazaa presumably reflects the underlying scale-free nature of the Internet as a whole. Now consider this PhysicsWeb article from 20 July 2000:

Barabasi and co-workers studied the effect that removing random nodes from a scale-free network had on the ability of the remaining nodes to communicate with each other, and the degree to which the network became fragmented. They found that the network's performance remained constant, even after they had removed as many of 5% of the nodes, and that it was resistant to fragmentation. But when the team simulated an intelligent attack by targeting the highly connected nodes, it was a different story: the network became fragmented very quickly, and with 5% of the nodes missing, its ability to communicate was halved.
By breaking the Kazaa P2P network into isolated islands, they greatly reduce the utility of the network. College campuses with student networks are potentially superhubs, and have thus been targeted as well, though via college administrations. I therefore conclude that the RIAA strategy is solidly grounded in recent discoveries about scale-free networks.

More on scale free networks blogged here.

Patti Anklam in the AOK Star-series dialogue on SNA

Fri, 15 Aug 2003 08:52:44 GMT

Charles Savage is interviewing Social Network Analysis expert Patti Anklam, in the AOK Star-series dialogue.

Charles: "Patti, as you are into networking analysis and social networking, I have a question: do people really like to reveal their networks?"

Patti: "I remember how John Galloway developed Netmap to do just this, but found that, in general, people preferred to keep their social capital to themselves, rather than reveal them. On the other hand, with the development of Social Software and Blogging, social networking seems to be taking on a life of its own."

[..]

Patti's interesting ! opening piece can be found at www.kwork.org/Stars/anklam/anklam.html