Google
 

Papa Roach "I almost told you that I loved you"

multi-viewpoint movie of my jump

Tuesday, February 27, 2007

Reputation systems and anonymity

Reputation systems help to maintain the order in Web world, and many sites utilize this system such as Amazon and eBay. However, there is still the possibility that someone will cheat in reputation systems. According to Dellarocas, one way to prevent cheating is controlled anonymity which means that the information about the reputation is open, but the information about identity is not open. But I don't think anonymity will prevent such cheating.

I think anonymity and reputation are conflicting concepts. In reputation systems, one is discouraged from cheating or antisocial behavior because such actions will cause bad reputation and that reputation will remain open to everyone. It based on the theory that without such a kind of punishment, people will not take appropriate action. However, that theory should be applied to the people who create reputation, in other words, evaluators. It is not guaranteed that they will equally evaluate. If they can evaluate anonymously, they may evaluate someone unfairly bad or good (maybe because of prisoner's dilemma). I often see that Amazon's some customer reviews are unusually low-rated (ex. I sometimes see "1 of 16 people found the following review helpful" on the reviews which I think pretty good). You need only Amazon's account to evaluate customer's review, and no one can see who evaluated it. Such things happen because of anonymity.

To avoid such situation, evaluators should also be exposed to reputation systems. Everyone should be able to see how they evaluated. Then, there will be the reputation of evaluator, and they will be discouraged from inappropriate evaluation. The reputation of evaluator is impossible with anonymity.

Tuesday, February 20, 2007

Tag's potential & problem

What I'm gonna write is very complicated and dfficult for me to express in English. If I write something incomprehensible, I apologize it.

Tag will help files to be categorized. This function will help seach engine because the data which has no letter information, such as picture and movie, can't be categorized without tag. In addition, if all Web pages are categorized by tag in advance, it will make search easy for search engine.

Tag has the potential to enable the Semanric Web. In Semantic Web, tag will categorize information by its meaning, and search engines can search infromation based on the purpose or meaning of the query. Oh, I can't describe it well. Just imagine that you put the query like "What is a picture painted by the Impressionists in 19c?". Semantic Web can answer your question. Search engine will find the file which has the tags "19c", "Impressionist", "picture". Usual Web will only match words in query with text, so such search is impossible. This is amazing.

However, using tag has one problem. Tag is not necessarily visible. Some blog shows it, but it is not always appeared. Therefore, if someone put irrelevant tags on his or her Web page, the readers can't find whether its tag is correct. It is possible to put irrelevant tag on the page to make the page searched for every words and ranked higher. That's why Google does not emphasize tag so much. When Google searched the Web pages just by matching query with contents, they had to struggle with those who attempted to be ranked higher by distributing many irrelevant words in their page. They don't want to go back those days. So to utilize tag for search, it is necessary to prepare countermeasure to avoid such mess.

Tuesday, February 13, 2007

The Advantages of Blog

When a company utilizes a blog for its business, what advantage a blog has? In "Naked Conversation", Scoble mentions that a blog enables two-way communications. Two-way communications are important for business because it is impossible for traditional advertisement to achieve two-way communications. Blog can do both advertisement and two-way communications at the same time. Is there other advantages? Let's think about them.

There are two types of advertisement. One of them is traditional advertisement such as TV commercial and junk-mail. It is "push" style, which means that a company unilaterally pushes its product information to customers. It pushes their advertisement regardless of customer's need. However, such ad started to lose its effect because customers don't want such ad and also technology help them refuses such ad, such as the recorders' commercial skipping function. Therefore, there is increasing need for another type of ads, and it is "pull" style advertisement. Pull style ad supposes that customers voluntarily access ads. Blog is effective as pull style ad because corporate blog will be read only by those who are really interested in that company. It can address to interested customers.

Another awesome thing about blog is good products can be mentioned in customer's blog. It will be a good advertisement because the comment by the users can be more reliable than what a company says. In addition, a blog tends to be read by people who have the similar preferences with the writer. Therefore, it can appeal to new potential customer. A company's good reputation can be spread by word-of-mouth. Moreover, if a blog can get track-back by many blogs, it will be ranked higher in search result.

Tuesday, February 6, 2007

What motivates people to cooperate

Rheingold mentions the cooperation of the masses for public goods. He refers to the development of Internet which is one of the successes of this cooperation. Today, utilizing this cooperation becomes trend. Open source, You Tube, Wikipedia, and Amazon's customer reviews are good examples. However, the success of these companies brought me a question. What motivates people to cooperate?

In fact, I can't find answer on my own. I can come up with several ideas, such as pursuit of self-benefit, intellectual satisfaction, good reputation and admiration, but none of them can convince me.
In many cases, the cooperation relates to their
own benefit. For example, if someone develops a program based on open source, which means the program code opened to everyone for free, it can be for self-interest because that program can bring huge money to him or her.
However, self-interest alone can't explain the case of Wikipedia because posting something on Wiki can't bring any money. In Wiki's case, it may relate to
intellectual satisfaction because if one's post is reflected on the pages without causing any argument, it means that many intelligent people accept his or her posting.
In Amazon's case, the desire for good reputation or admiration may be related. Amazon's customer review is also reviewed by users (You always see "X of X people found the following review helpful" in Amazon, right?). In addition, there is a ranking system called "Top Reviewers". These systems will bring reviewers a sense of admiration.

Yes, I could come up with these ideas, but I can't still be comfortable with the conclusion that these factors motivate them to cooperate. I can't understand why the quality of Wiki can be better than Encyclopedia Britannica (and also why many professors don't like student referring to Wiki!). So if you have any idea, please feel free to comment.