Following excerpts are not as structured as a serious composition should be, merely combining the selected sentences I took from the book and translated back to English.

• Chapter 1: Emergence of Social Web Era
• Chapter 2: Framework for Social Web Development
• Chapter 3: Sincere Conversation
• Chapter 4: Bringing people to sign up

<< Previous Chapter 5: Keeping the users constantly satisfied

Selected excerpts from Chapter 6: Building collective intelligence

Sometimes, an individual must be sacrificed to provide better service to the mass.

Complex System is the mutually linked one, cannot prove its features or functionality when it’s disintegrated.
But when it’s integrated as one system, it shows more than one actual or potential features or functionality.

Adaptability is the capability to change flexibly throughout time.

It’s important for the designers and the developers to flexibly change the target users for keeping the system as healthy as possible.

The purpose of this kind of social system is to concentrate individual activities and bring out or create the most qualified and largest number of contents. The result of this process is called the collective intelligence.
The designer or developer can draw insightful ideas and information from this collective intelligence generated by many people.
In other words, the larger number of people can make the more qualified and accurate decisions, than the smaller number of people.

For this to be possible, activities of the users must be recorded and the innate or implicit patterns must be brought out using these steps:

1. The initial behavior like submitting a new content. The fate of it is determined by the system rules and the other users.
2. Exposure. The algorithm or the rules of the system based on the subject or suitability to the system. The exposure changes to display the good ones among different contents.
3. Feedback. The users of the system can change the exposure of the content by affecting the algorithm through positive or negative interaction like commenting to the content.

When your content is exposed with other similar contents, it would require only little effort to promote it.

Too much content of different quality causes confusion to the users. The system must be able to control the submission to encourage quality content exposures over others.
Entry barrier is the mean to gain superiority in the market, likewise adopting or abolishing the entry barrier in the web service is critically considered for keeping the service as healthy as possible.
These are three types of barriers:

1. Unofficial barrier: The special design or advertisement only work implicitly on special group of people.
2. Official barrier: Requiring signing up for an account, installing an application.
3. Extreme barrier: Invitation only, not open to the public.

Digg uses these kinds of barriers to ensure the quality of content submitted.
It is important to evaluate the suitability and the originality of the content.

Normally, people consider the well exposed content to be valuable.

• Exposure in the main first area
• Frequency of exposure affects the value
• Head position in the first page
• Higher rank compared with other contents

The purpose of most content is to draw a lot of visitor traffic, to the point of server error!

The ranking depicts the relevancy of contents to the users. Each user applies different combination for viewing conditions to the list of contents.

• Chronological order
• Popularity gained during certain range of time
• Level of participation
• Collaborative filtering
• Relevancy
• Identity of the content provider
• Providing different perspective to look at one’s own content

The ranking is important to promote better user experience. It allows the user to recognize the value of content by comparing it with others.
The value of news service is freshness, of search engine is relevancy, and of social network is relationship.

Interface of the service is the world where users dwell and play. What’s displayed through interface determines the user experience. People often rely on the decisions made by others to make their own. Unless, the interface of the service purposely not showing information about what others think of the content, people’s evaluation on the content is heavily affected by how positively of negatively the content is examined beforehand by others. Social interface changes user’s decision.

The implicit and explicit feedback to the content are combined together to construct the overall reputation of the content.
The implicit feedback is generated from the user activities like visiting the page, bookmarking, clicking the download icon, and purchasing.
The explicit feedback is like reviewing, commenting, expressing the user preference.
It is very important to make the mechanism for feedback to be so easy to use, like Digg button on a page.

Leverage point is the small thing of the service which can affect the large part of it. Opening this leverage point to the users to participate in building the quality of the web service is necessary for actively growing the social network service.

Next Chapter 7: Encouraging to share >>

The article talks about the importance of prioritizing search result, or stream of posts, not based on chronological order, but based on social relevancy, or social network of followers of oneself and people he or she is following.

What we are about to get is a Social Relevancy Rank. Whenever you search streams of activity, the results will be ordered not chronologically but by how relevant each is to you based on your social graph. That is, people who matter more to you will bubble up.

The author brought up improvement ideas for mainly Twitter, or also for any SNS. Main points are:

1. The result of search query, or the front loaded posts should come initially from the relations to myself. But it limits the result, because not everyone has opinion or information about the search query.

This sounds awesome, but there is a problem. “Wilco” works well as a query because the band has just released a new album, but many other queries would return no results. Simply put, your friends on Facebook and people you follow on Twitter can’t possibly have an opinion on every topic you may be interested in. This is a problem of sparse data: trusted opinions are scarce.

2. Since the trusted result from the level-one relations is limited, expanding the network to include the relations of the relations is necessary. This expansion can also be based on likeness of people even outside of the relations.

Another step could be to include people with similar tastes, so-called taste neighbors. This approach is common among vertical social networks such as Last.fm, Flixster, and Goodreads. These networks have ideas about which people, other than your friends, are like you. However, this is a costly calculation and takes time.

3. Using the number of followers as a measure, a few of the relations can be titled as opinion leaders or the influencers among the crowd. Give higher order to their opinions in the search results or the stream of posts.

In the absence of any other metric, someone who is followed by hundreds of thousands of users is likely more relevant to you than someone you don’t know at all. Using number of followers as a weight might be a good way to order the rest of the activity stream.

Social Relevancy Rank by Alex Iskold

I think this relevancy rank can improve what Twitter, or stream based publishing systems have been good at. For their currently chronologically ordered search results of stream of posts to be more useful or more easily to be found and processed, the additional sorting mechanism based on this relevancy ranking is crucial. Depends on the level of openness of the service enabled by APIs, the application of this system may come much quicker and may be developed to become superior service than the original.