Session 5: Social knowledge production and services: Peer production in online environments vs. in-person collaboration

For this session I choose discuss on “Peer production in online environments vs. in-person collaboration” as I am impressed by the success of Wikipedia

The peer production of open source software development is very successful. It shows that the networked communication can build individual contributions into collective, synergistic projects without intervention from formal institutions or dependence on conventional expertise. Generating from the Open Source software development, “Linus’s Law” (Raymond, 1998) appears. It states that “given enough eyeballs, all bugs are shallow.” That is any problem is ultimately trivial in software development where, according to this law, the number of people contributing to a project provides a useful indication of its quality. Paul Graham claims that “The method of ensuring quality” in peer production. That is: the good stuff spreads, and the bad gets ignored” (Graham, 2005).

All the laws based on the assumption: more people making more changes only make things better. That is numbers and time work in favor of quality. In many cases they may, but in some they appear not to. Open Source software is for the developers which is easier to work in a networked communication environment, but ordinary users are quite unlike developers. It is difficult to transfer from Open Source software which produced mostly by developers to a project involved contributions from ordinary people like Wikipedia.

Do these laws can be extended to all peer productions ? How we guarantee that the good stuff remains and not the bad remain? here comes the boundary of these laws.

1. Freedom of speech is not the same as the freedom to replace other’s versions of the truth with your own.
2. Though many eyeballs survey a project and many hands update it, work on the system is not necessarily distributed equally. Hot topic, of course attracts more eyeballs. So Wikipedia limits the trust it puts in “improvement” in the quotation above to “widely circulated articles”.
   

With all of these constrains, many huge distributed projects are produced online, Wikipedia is one of the very successful peer productions in online environment. What Wikipedia impressed me most is that: it reports the daily news and update on every topic (see screenshot for example.) As an ordinary user, you can easily grab the new developments in your interested areas and as a developer, you can post your contributions easily.

For the in person cooperation, the group can focus on more specific topic. The group members are less variety than the online communities. They are easier to get agreement on the topic and produce a consistent project while the online peer production may constantly change. What is flawed today may be flawless tomorrow. But the in person cooperation are limited by the frequencies the member meet and the place they are located. Not many people can join the cooperation. It’s difficult to produce complex and large project. The in person cooperation can be improved greatly by online tools, like email etc. With the help of online tools, more people can join the cooperation, not just the people who can meet face to face.

So we can say the best way is the combination of peer production and in person cooperation. But there are some boundary on the varieties of the cooperation people. Beyond the threshold, more cooperation may not bring in high quality result. As more and more people join the cooperation, we face the same problems of the quantitative assessments and the qualitative judgments people tend to make.

Session 4--Social role, capital and trust

Gleave, etc ‘s paper focused on ‘social role’ in online communities. They try to define the ‘social role’ conceptually and operationally. Comparing with their ‘social role’ definition, I like the way they extracting online social role. In their diagram, based on the collective data, extracting social role by watching people with specific action patterns and discovering their relation with the remaining communities. Iteratively refine the ‘social role’ until we get stable conclusion. Then we can predict this new ‘social role’ pattern in other communities. From this paper, in the first time I know how social scientist analysize an online communities.

To extract model from huge collective data which seems unrelated is very difficult. Gleave, etc point out that “Building a catalog of social roles is an important first step towards understanding complex social systems” . It allows us to reduce the huge systems to a relatively small number of roles. There two general ways to do social role analysis: interpretive and structural. Interpretive analyses focus on the collections of e behaviors while Structural analyses mainly deal with metrics of social structure. Both approaches have their own limitations. Gleave, etc combines these two methods together in online community analysis. They obtained very interesting results.

To explore their methods, I joined two online communities. One is http://www.forumshawaii.net/. My user name is yw123. It is Hawaii's Sports Car Club of America's official forum. The homepage attracts me at the first. After I login, all are gone. I got an almost empty forum. I need to give the topic name I am interested. That takes me a while to do it as I don’t even know what kind of topic they have. So I can say it is not well organized and classification. It is very help if it can post the catalog after user login. I do learn a lot about the sport car in this forum. It’s very useful. I give me trust to this forum although I don’t know wither the forum check the information post are correct or not.

Frankly say, I can’t analysize this community like Gleave did on Usenet and Wikipedia. There are so many members and so many posts, I can figure out who plays any role in this net. It has the administrator who I believe play a crucial role in the communities. The members can communicate directly or they can post their questions in the forum. The forum has tree structure. It includes five sub forums , each forum has their own specific topic. Inside each sub forum, there are several interested topic etc. According to member’s contribution, the administrator may assign some member’s priority to members. The members with high priority play important roles than others. Maybe like the administrator of sub-forum. Maybe they have some right to adjust other’s post etc. That all I guess as I just join in and don’t have any priority.

The next step is to analyze the context of participation and the content of behaviors of the actors whose interactions formed those social network structures. Williams, etc investigate the relationship between the Internet and social capital. He points out instead of goods and services, the things being used and created are personal relationships and the benefits that come with them. He focused on the effect of the network although social capital may also have relation with network itself. As the online systems are so different with the traditional society, new concepts “bridging” and “bonding” are introduced. “Bridging” is the connection between social networks. It is broaden and weak. “Bonding” is the connection between strongly tie individuals. It is strong with little diversity. Williams tries to set a measure scale on these two concepts and got some interesting results. Some social actors interact and form a network of individuals, resulting in positive affective bonds. Some interactions are qualitatively different from others, meaning that different types and levels of social capital will result.

According to his opinion, the Hawaii's Sports Car Club of America's official forum I joined is clone to bonding as it has very little diversity. But I feel the connections between members are still too weak for bonding and are too strong for “bridging”. I think it should belong to some area in between. To have some experience on social capital, I joined a very popular net, the Wikipedia: http://www.wikipedia.org/. It’s typical ‘bridging’ site. Anyone can join in and anybody can post. The username I use is also yw123. This is a very famous net. It is very well organized and classified. I use it as a science directory quite often. There is one part I feel it need to improve. It does not check the information people post. That is the information you got from Wikipedia may be wrong. As more and more people use this net, it is a great help if the information is checked before it is allowed to post.

Paolo investigate the trust concept and explore how trust is used and modeled in online systems. He point out trust is considered as the judgment expressed by one user about another user, often directly and explicitly, sometimes indirectly through an evaluation of the artifacts produced by that user or his/her activity on the system. “trust” is the term to indicate different types of social relationships between two users, such as friendship, appreciation, and interest. These trust relationships are used by the systems in order to infer some measure of importance about the different users and influence their visibility on the system. Ellen etc complement his opinion by introduces a new generic model to incentivize cooperation between parties that are engaged in the paradox of a social dilemma. Eryilmaz etc explores the design and evaluation of a trust model to establish trust management in an open source collaborative information repository for an emergency response environment. Nicole B. Ellison etc give an example by examining the relationship between use of Facebook, a popular online social network site, and the formation and maintenance of social capital.

As for the two nets I choose, I give my trust to the Hawaii's Sports Car Club of America's official forum as I know very little on sport car. I choose it because I happened to find it and the homepage attracts me. After I joined in, I realized that it’s a sport car club. I accept all the information I got from it. As for Wikipedia, I use it quite often and trust the information I got from it in %90. It’s so weird that the more I know it, the less trust I give it. But this is what I really feel!

I plan to choose two nets, one is belong to “bonding” , the other is belong to “bridging”. It turns out I only choose a correct “bridging” net. The other net I feel it trapped in between “bridging” and “bonding”. So the topic of My final project is in the social capital area. I plan to further explore the concept of social capital and how it can be measured. There are lots of arguments on this topic. William’s paper is a good beginning on this topic as he mentioned so many references.