This chapter provides a big picture of viewing online social interactions through the social network analysis approach. Although it was written 10 years ago, most of the claims are still valid and the many of the research questions remain unsolved and continue to interest researchers.
Here I wanted to raise a couple of thoughts upon this reading:
* Wellman attempt to grasp the main characteristics of online social relationship (or i would also like to say the structure of online social interactions) by 6 measures (or dimensions). One of those, "range" seems to me to be much more complicated than others. The author explained the major variables they consider on this dimension construction though. I am thinking if it could be better to separate the concept like "scale" for the sense of how big the community is. This concept would not be as simple as just how many agents in it, but also can include the ideas like how they are geographically distributed and so on. On the other hand, "range" also wanted to capture the feature of "diversity" of the community, which can be many things like people's demographic distributions. Additionally and importantly, every online community has the characteristic as how different roles people play and how the roles are distributed over the participators. This could be very different among participators as well as among different communities for the same person.
*In addition to those rather measurable and relatively stable characteristics of the online communities, I thought some other features can also be interesting and important. For example, like the dynamic and evolution, as well as the community culture formed though the interactions between the community and the way they are connected (e.g., the application, interface, mechanism, and even some unpredicted mutation).

One of the threads I found very interesting in this chapter was the idea that relationships can be formed offline and later taken online or can be formed online and later taken offline (see top of page 10 and the second paragraph of page 14).

I think that this has important implications for not only the initial basis of the relationship formation, but also the nature and quality of the subsequent maintenance of the relationship as well as the parties' views of the computer network itself. For example, whereas a relationship formed offline may have emerged from mere physical co-location, a relationship formed online may have emerged from a shared interest (as mentioned by Wellman). In a relationship formed offline and subsequently maintained and/or enhanced with online contact, the computer network may be viewed simply as a communication tool (much like the telephone). In contrast, in a relationship formed online and subsequently maintained and/or enhanced with offline contact, the computer network may be viewed as much more than just a tool -- maybe even more like a place (for example, akin to the 'bookstore' or 'bar' where we first met).

 --- Beth

Beth St. Jean

Much analysis of whether CMC supports or enables "real" communities is confounded by misconceptions that communities are groups. Wellman looks at communities as social networks. From the social network perspective, groups are just one iteration of a social network. Social networks (communities) can be made of dense and tightly-bound groups, or they can be sparse, loosely-connected individuals. Wellman argues that CMC supports social networks of many different characteristics (density, boundedness, range, exclusivity, etc).

It is interesting that many of the examples Wellman gives of technologies that support different types of social networks are actually the same technology. For exampel, he says that email forward/copy features facilitate dense networks. These features are mechanisms for common knowledge and common awareness. However, email also supports sparse networks. Email allows private interactions and communication with geographically disconnected people.

This paper explains how social network concepts can be used to understand computer mediated communities. First the author makes distinction between dense and sparse networks in both work and community contexts. (However the paper did not define work and community.) With these four categories the author explains the impact of computer mediated com munition. Then the author has listed six network characteristics that can be used to understand online networks. The examples in the paper show how sparse networks are equally important as dense networks in general. In many contexts weak ties play a more important role than strong ties.

The first four parameters (Density, boundedness, range and exclusivity) seem to be a measurable characteristic. However social control and strength of relationship are ambiguous and I don't understand how they can be used as a differentiating factor.

It is also possible to have one dense network that has both computer mediated and direct relationships. An example will be a software development team spread across multiple locations.

Wellman give us a very comprehensive introduction on the topic that
how to apply social network analysis on online communities, including six
spheres: Density, boudedness, range, exclusivity, social control and strength
of relationship. Furthermore, he gives a sketch on the comparison between
online communities and traditional communities, which I think has much
potential to study in the future, However, besides the research shopping list in
the last of this chapter, I have two questions about the analysis of online communities.

First, instead of homogeneity of individuals, people engaged in
online communities, especially those for entertainment/fun purpose, people are extremely
heterogeneous, which make the application of these measurements above more
difficult to apply on them. For instance, as a high school student who actively
participate in one online game forums, he or she would be probably willing to
spend much time on maintaining this website if they are “promoted” as a virtual
manager, and they consider themselves as one member in this virtual hierarchical
system, on the other hand, as a graduate student who have lots of other things
to do, he or she would prefer to be a “free
rider” on this website, for example, reading the groupnews, know-how to go
through one game. It is hard for them to be controlled by this virtual society.
In conclusion, online communities are more complicated and usually belong to
the boundary of these dimidiate categories. Instead of describing one online
community as a whole group, it is better to investigate different users(frequent/infrequent)
in this group.

Another one I am interested in is that how people describe
themselves as a member of one special online communities compared to traditional
communities. How strong of this new identities and how does it affect people’s
interactions with each other. For instance, instead of p2p transaction in Ebay,
what would be different if we move the transaction place to a online fans-club
where people share more common interest and have higher identification with
each other?

Barry takes as his point of departure two major questions:
1. how topology affects social life
2. how technology affects topology

A brief point:
Wellman is unclear as to whether dense = bounded, and sparse = unbounded. He seems to treat them as if they only occur in that configuration. However, interaction online can be dense and unbounded as well as sparse and bounded.

Next, I will briefly comment on his criteria for investigation:

Density:
I think that "dense" and "sparse" is less useful as a binary characteristic than as a continuum along which we can place networks.

Boundedness:
The question of who is "within" or "without" a boundary is all too often ambiguous and highly dependent on how we define the boundaries in the first place. There are always people who are not technically within a boundary but get treated as if they are.

Range:
"Large" and "diverse" on what continuua? by what criteria? on what landscape?

Exclusivity:
How is this different from boundedness? Barry speaks of the "difficulty of coordinating sparse networks," but compared to what? Does this task become easier with better tools (technology?)

Social Control:
I prefer to think of this in terms of "selection pressure." Social control is never control, only influence. Moreover, this selection pressure can be exerted on both behaviors and members, resulting in behavioral changes as well as membership changes.

Strength:
Strength has two facets: frequency and intensity. While we see a lot of analyses focussing on frequency, we see less of them paying attention to intensity. Both are crucial.

--------------------------------------------------------
PHartzog@umich.edu
--------------------------------------------------------
The Universe is made up of stories, not atoms.
--Muriel Rukeyser

I'm really impressed with the comments so far so instead of directly echoing what others have said so eloquently I'd just like to highlight a few things that are interesting to me and my reseach, but may have gone unnoticed in a quicker read:

--On p. 8 he notes, "researchers at different locations rarely co-authored papers before the advent of email." I'm specifically interested in collaboration and coauthoring and writing-specific learning. I'm already anxious to see how much we'll discuss collaborative learning/writing as opposed to mere exchanging of information (via snail-mail, phones, etc.)

--On p. 11 he notes, "because all members of computerized conferences can read all messages--just as when a group talks in a cafe or an open office--groups of people can talk to each other as casually and get to know the friends of their friends." I'm of the opinion that this is *the* thing that makes social networks work: the coauthorship on a public yet solely owned page. You can see this being enacted on Facebook's Wall, blogs' comments pages, or even user reviews on Amazon.

This was an incredibly interesting, yet dense, reading. It would be very interesting to go through his final list of questions and note which ones have been addressed by research that has followed in the 10 years after this publication. 

Wellman talks about how existing characteristics of networks affect interaction among "e-communities" and how these newer interaction methods impact interaction and social relationships. The author frames this discussion around 6 characteristics of social networks and seems to conclude that overall, computer-networks and the practices and behaviors it solicits, positively impact interaction and social networks. However, Wellman did not say that this was always the case. Perhaps I have a somewhat pessimistic perspective. For instance, Wellman does mention that instead of “destroying community”, computer supported social networks are “responding to, resonating with, and extending the types of community that have already become prevalent I the developed Western world. However, I wonder if this is true across all existing networks. It seems to me that there must be cases where instead of serving as a mere response, computer supported social networks are in fact causing isolation (while of course, whether that is a problem in itself is arguable).

This leads largely to another point which was also touched upon – individual behavior. Wellman’s discussion on the relationship between social and computer networks and the 6 characteristics of social networks include observations on both groups and individuals. For instance in discussing density, Wellman explains that computer networks work well for “sparse, unbounded networks” because individuals can decide how much time and effort they want to spend in any given e-community, or how they might want to present themselves in another. This is interesting because Wellman has already advocated going into one extra level of detail, suggesting that “thinking about relationships in social networks instead of groups can allow analysts to take into account the contexts within which relationships operate.” However, considering how much differently each individual may behave seems to suggest that the large emphasis on ego-centered analyses is inevitable, even though, everything else shows that over-arching networks exists.  

@Chris G. - I too hope we discuss Wellman's questions in class.

From Wellman's shopping list: "To what extent are on-line relationships and virtual communities free-standing or integrated into other modes of communication, such as in-person get-togethers, telephone calls, or written communications?"

From what I think is a whole network view, meetup.com may be an example of a hybrid computer and social network because its purpose is to enable the discovery and support of offline groups. For example, "pastoralist myth of community" (p15) believers can search for and join a group of other alarmists - bashing the Internet in a cafe, and later using their meetup.com listserv to organize another in-person meeting.

meetup.com as a whole network has low-density, unbounded, high-range properties because participants can freely enter and exit independently-operated sub-groups. Thus from a global view, I think tie strength may be low unless there are many inter-personal links between otherwise isolated sub-groups.

When Wellman claims that computer networks "extend the types of community that have already become prevalent in the Western world" (p6) - I think he's missing the mark slightly because computer networks may also replace, rather than extend, "real" interaction. For example, distance learning programs sometimes displace college enrollment locally.

It's interesting to already see course concepts from SI 508 (Networks) in a reading for this course. The content itself seems informative enough, though I did think that Wellman's "open question" about the potential for strong ties through soley online interactions is outdated.

Wellman asks, "can relationships be strong if they are only sustained on-line?" Limited social presence and asynchronous interactions are used as examples of why such ties would be weak. I'd argue that those examples would strengthen ties in the context of social interaction. After all, asynchronous communication is better than none at all, and who has time for physical, temporal synchronization? Existing relationships can be sustained through online interaction, and new relationships formed. I think there are enough examples of strong online relationships (WoW guilds, online forum users, bloggers, etc) that we can say that online relationships can be just as strong and meaningful as real life relationships.

Wellman discusses how computer networks can ‘support’ social networks and how the structural properties of these networks affect the social relationships of people.
Since the article has already been summarized so well and various points discussed I’ll just add to the portions which I feel strongly about. First, I’m not entirely convinced with Wellman classification of groups into dense-bounded and sparse-unbounded. I’m like to second Paul because there will often be cases where the network does not fall neatly into the categories he mentions. In fact relating it to another reading this week, some communities of practice depending on the people involved and the way they work will be hard to classify with certainty in one of these two categories
Secondly, the question of the strength of ties and the relation between online and offline interactions has already generated a lot of interesting discussion here! I like the way Wellman connects the strength of ties to the structure of the network. It’s interesting to see how Wellman looks at ties in term so fhte cost that it takes to maintain them and where that cost is coming from – which gives an interesting way to look at a network. Like some other students here I’m inclined to believe – in part due to personal experiences- that strong ties can be created and supported by online communication and relationships can be taken from one medium of communication to the other without affecting the strength of the relationship significantly- but I’m also aware that maybe I’m just being optimistic!

The comments brought up by others so far have been very interesting. I also felt this was a good introductory piece on social networks, that while 10 years out of date, shows that while technology may have changed, the ways in which we communicate and how that communication is structured hasn't changed all that much. (I would be interested in finding what researchers are currently saying about this though).

This was already pointed out, but the biggest issue for me in the reading was the binary distinction between dense, bounded groups, and sparse, unbounded individuals. The author does mention briefly that in reality the distinction is much more complex and multivariate. He then goes on to say that this will be shown through his discussion of the six characteristics - density, boundedness, range, exclusivity, social control, and tie strength. But I don't know if he actually makes a clear distinction throughout his discussion, since there are so many possible examples of these networks. It seems to me mostly that the dense groups are ones where members know each other in person (work groups, small villages, familes), and the sparse, unbounded groups are any of those who are not in person, but communicating via computers.

One last observation of the piece - I am intrigued by the research question of whether we are becoming more isolated socially because of our increasing use of technology to communicate. This seems like it was more of an issue 10 years ago when email and the Internet were gaining popularity, and you hear fewer concerns now. Wellman briefly mentioned this as well which made me think of it. 

On page 9-10, the argument that the "relative lack of social presence online fosters relationships with people who have more diverse social characteristics than might normally be encountered in person" stood out. I've heard this argument before and tend to agree with it, but I'm not so sure that the "relative lack of social presence" continues to be true in many cases.

Before having a conversation with someone (online or offline), I can facebook them, check out their flickr stream, review their music tastes, and see who else they've "friended," and make many inferences (that may or may not be correct) about their social status. If I'm inclined to hang out with people who are similar to me according to one of these axes, I can do that. If I'm inclined to seek out this information, the online social presence can be overwhelming. The same goes for a discussion board or other online group. If people link to their profiles elsewhere, I can make a lot of judgements (that may or may not be correct) about if I think I will fit in -- without ever reading any of the group's content.

This sort of social presence is different than a face-to-face conversation (and is perhaps more optional), but it does not seem to be less. In a lot of ways it is more: not just the volume of current information that might be accessible about someone, but also that I can pull up no only current evidence about their behavior and social status but also quite possibly past evidence if I go back to the start of their photo galleries, last.fm charts, etc.

In this article Wellman shows how social network analysis might be useful for understanding how people relate to each other through computer mediated communication. I imagine at one time this was a rather novel question but now it seems to be more rhetorical than anything else. However this article did raise some interesting issues.

First I felt that the author ignored the role that proximity plays in the formation of social networks. He states “…on-line discussion groups, bulletin board systems, virtual chat rooms, and the like to make meaningful contact around the world with newfound comrades”. While this is true it also ignores the extent to which our social networks are influence by our daily interactions.

Another aspect of this article that struck me as odd was how awkward some of the terminology fit when it was applied to a physical community and an online community. In the example about density the author makes the point that the forward-and-copy features on e-mail can increase network density. But this measure of density hardly seems comparable to face-to-face communication, even though in his analogy they seem to be equivalent.

I have always imagined that Paradise will be a kind of library.

-Jorge Luis Borges

Wellman describes the internet as "an extreme example of an unbounded network (p.8)" based on its seemingly non-existent population boundary and its illusive structure (always changing, therefore difficult to analyze).  At a National conference for community organizers I attended last year (The US Social Forum in Atlanta Georgia), an activist handed me a shot book entitled "The Organic Internet: Organizing History's Largest Social Movement"  authored by a Non-Proft web hosting agency called May First.  The books primary claim is that the unbounded and rapidly changing qualities have led to the internet becoming itself a social movement.  They support this claim by arguing that the internet has fostered cultural changes and created a new language, has alluded attempts to constrain its accessibility, and "has turned communication into an act of social resistance (Organic Internet, 2007, p.19)."

Although I have a number of qualms with the argument, I think it is a rich argument that deserves consideration. It is particularly salient in the context of budding web 2.0 technologies that increase the extent to which individuals with minimal training can produce content rather than just consume media produced by others.

This potential, however, is limited by the degree to which people are exposed to technologies in ways that are relevent to them.  The "techie" activists who attended the US Social Forum attended a small series of sessions that were occupied by like-minded folk who have enough economic stability and privilege to have the luxury of thinking about spending time voluntarily contributing to Free and Open Source Communities

The few who stumbled into the room from other sessions on racial justice or lgbt rights, didn't stay long.  It was immediately apparent that no matter how great the social networking potential available through online communities may be, it is useless if the potential is so muddled in "geek-speak" that it can't be deciphered.

Lisa McLaughlin

I once belonged to a group (or social network?) of central bank webmasters. I was formed by an invitation to a meeting. Webmasters from about 100 central banks attended and relationships formed. After the meeting, the discussions continued via email. Webmasters from regional central banks started organizing smaller meetings in between the large global gathering every few years. Some meetings also started to become less formal. A couple of years back, the group started a wiki on some data exchange standard. Then I came here.

The question is how can my group be classified in the light of Wellman. It is both a social network and computer-mediated. It is relatively sparse compared to other known networks in the central banking community, yet not unbounded. Of course, the group need not be forced into any one type for discussion of the six characteristics--the group itself can be multitype--for the purpose of this discussion.

Wellman talks about how characteristics of computer networks can be analyzed by social network analysis. Because social structures served as the patterned organization of network members and their relationships of social networks, the author frames this discussion around 6 structural variables of it, the density, boundedness, range, exclusivity, and strength of relationships.

On page 10, the argument that the "many computer supported social networks are ... organized by mutual interests rather than by shared neighborhood and work site." I tend to agree with this argument that the lack of on-line clues about socially-status promotes connections between eCommunities. Firstly because, users tend to sharing less about gender, age, race, ethnicity during social network, instead they share much more about the resources each other have and the similar concerns, interest and values.  secondly because some socially-status may do barrier communication on-line. For example, the boss and employee, once the social characteristics of each others known the connection and communication online changed.

But I am not full agree with this point as author points out that online relationship developed on the basis of shared interests rather than differences in social status. Sometimes social status do help build similar interests. For example, college students tend to have the same interests with colleagues rather than kindergarten children. And people are able to get to know the social status of the one he/she communicates after shares key points of interests.

John Blair

Reading this made me think that as technology has developed, providing us more and more tools to accomplish tasks (certainly one of tools primary purposes), it occurs to me that while the development and proliferation of varying ways to communicate grows and enables all kinds of people to participate, are we becoming less social because of it?  We no longer really have to actually interact with a person any longer in many situations if we choose not to.  Wellman begins to address this towards the end of the article, but never explores it in any depth, as it was not the focus of the subject matter being explored.

Wellman is trying to discuss how computer networks support varieties of social networks at work and in the community. Technology keeps changing our life in a dramatic way. Actually I always feel that computer kills conversation. But on the other hand, I won't disagree that it improves communication among people in various ways and that is also one factors that eventually brought up today's common communities. I think Wellman has successfully shown how some of the key concepts of social network may be applied in understanding and analyzing online communities. In addition, the implications for computer supported network of both dense and bounded and sparse and unbounded types of groups that he mentioned seem to impress me and I believe it would help in the future study of online community.

There are 6 characteristics of social network for us to consider: density, boundedness, range, exclusivity, social control and tie strength.