Web 2.0

The Tipping Point

Tue, 21 Nov 2006 09:18:00 GMT

I finally finished reading The Tipping Point by Malcolm Gladwell. This was one of the most well-written books I’ve ever read, and I already can’t wait to read his next book, Blink. The Tipping Point is an attempt to understand how certain phenomenon happens rapidly and dramatically once a critical mass or “tipping point” is reached, almost like the spread of an epidemic. For example, how do you understand the sudden rise of teen smoking, or the multiplying effect of word of mouth? The Tipping Point describes three principles that are critical to understanding the Tipping Point phenomenon.

The first principle, the Law of Selected Few, says that not everybody is capable of starting an epidemic. On the contrary, only a few exceptional people are capable of starting one. These special people are classified as the Connectors, Mavens, and Salesmen. Connectors are those who are gifted in bringing people together. Mavens are those who are extremely knowledgeable in certain subject matters. Salesmen are those with exceptional persuasion skills. These people can start the kind of epidemics that most people cannot because of their unique personality, knowledge, and credibility.

The second principle is the Stickiness Factor, which says that to be effective information must to be conveyed in ways that make it memorable and “sticky”.

The third principle, which I think is the most powerful, is the Power of Contexts. This is also known as the Broken Window theory, which says that broken windows (context) cause increase in the number of crimes. This theory implies that the right context causes people to behave differently than they normally do. This is scary to think about, and it even suggests that some attributes of a person, both positive and negative, may not be as absolute as we’re used to thinking.

In Afterword (which was written after the original Tipping Point was published back in 2000), though it sounds contrary to the arrival of the Information Age, the author argues that we’re about to enter the age of the word of mouth. As the amount of information increases and the size of the technological networks of people (via phone, fax, and email) expand, people develop immunity, that is, people will learn to use it selectively. Increasingly people will rely on the Selected Few, in particular Mavens in the Information Age, to deal with the complexity of the modern world.

The Tipping Point - ティッピング・ポイント

Tue, 21 Nov 2006 09:17:00 GMT

Malcolm Gladwell 氏による The Tipping Point を読んだ。この本は、あるアイデアや口コミが一気に広がる劇的瞬間（Tipping Point）を感染に例えながら分析・解明する。著者によると「劇的瞬間」が起こるにあたって三つの法則が挙げられるという。

第一は「少数者の法則」。「感染」の裏には特別な能力を持つ少数者が存在するというのだ。その能力とは多くの人とつながりを持つ Connector 、専門的な知識を持つ Maven 、そして並外れた説得力を持つ Salesman の三種類に分類される。こうした一握りの例外的な人たちの働きによって口コミなどは「感染」されるという。

第二は「粘りの要素」。情報の表現や伝達の方法を工夫して覚えやすくすれば、当然その情報は「感染」し易くなる。

第三は「背景の力」。これは壊れた窓の存在が犯罪を増加させるという「壊れた窓」の法則と似ている。つまり、状況は人間の行動を変える鍵であるという考え方だ。これを一歩進めると例えば人間の性格など不変であると考えがちな特性も実は外部環境に決定されているかもしれないということになる。「感染」の広まりにはこの「状況」が大きな鍵となる。

「少数者の法則」、「粘りの要素」、そして「背景の力」の三要素が満たされた時、アイデアや口コミなどは一気に広まる。もしその情報の表現の仕方が少し違ったり、または違う人によって情報が伝えられたり、または状況が少し違ったりした場合、その情報は「感染」しなかったかも知れないという。

結論で著者は情報社会においての Maven の重要さに触れている。情報量が増え、テクノロジーによる電話やEメール等のネットワークが広がるにつれ、我々は情報を選択せざるを得なくなる。そのプロセスに大きな役割を果たすのが「少数者」、特に Maven だと言う。

Why the Future of Business is Selling More of Less

Mon, 09 Oct 2006 08:47:00 GMT

Chris Anderson's The Long Tail is an interesting reading to understand Web 2.0 from the perspective of the Long Tail phenomenon. Though I've heard his talk in the past, it was great to finally read his book and get a deeper understanding of the concept of the Long Tail. I'll discuss some of the ideas in the book that I found particularly interesting.

First, the Long Tail model creates probabilistic systems. I believe this is a key to understanding Long Tail systems. One implication is that one cannot rely on such a system to provide a definitive answer. One can, however, rely on the collective accuracy of answers that are provided. Wikipedia is a prime example. Developed by the Long Tail of professional and amateur writers, Wikipedia may not be able to provide as definitive an answer as traditional encyclopedias on a single topic. Collectively, however, Wikipedia provides broader coverage of topics as well as superior overall quality. And as with other probabilistic systems, I believe it is fair to say that Wikipedia is more useful than traditional encyclopedias in most cases.

Another idea I found interesting is the observation that the Head economy is driven by money, while the Tail economy is driven by other factors, such as reputation. Typically the Head is comprised of professionals while the Tail is comprised of amateurs. This is particularly apparent in the world of blogging, where selected bloggers are successful in monetizing on their effort while the rest are driven by passion and reputation. This is an interesting characteristic of the Long Tail.

Another important notion to understand the Long Tail is that as the collective size of the Tail increases, aggregators become more important to sort out vast amount of information. Google, Amazon.com, eBay, and iTunes are such examples. Some aggregators have sophisticated mechanisms, such as Netflix’s recommendation system, to expose Tail information to larger audience. These mechanisms are designed to combat the paradox of choice by providing relevant information to the audience that assists them with their decision making.

There are many other interesting ideas that are presented in the book, such as the implication of the Long Tail on both business and cultural diversity, and I highly recommend it if you haven’t read it yet.

ロングテール：ビジネスはどのように変化するのか

Mon, 09 Oct 2006 08:42:00 GMT

Chris Anderson 氏のベストセラー「ロングテール」は新しいビジネスの形、そして Web 2.0 の新しい見方を提案する。彼のプレゼンは以前見たことがあるが、今回本を実際に読んでさらに多くの事を学ぶことができた。特に面白いと感じたアイデアについて少し書いてみた。

まず、ロングテール現象から生まれるシステムは「確率的」であるという特徴は、ロングテールを理解する上で重要な要素だ。これはロングテール現象からイエス・ノーという答えを出すのは難しいが、全体的に見れば確率の高い答えを得ることができるという事だ。例えばプロとアマチュアというロングテールによって書かれた Wikipedia と今までの百科事典を比べてみると、一つのエントリーだけを見れば百科事典の方が質が高いかもしれないが、全体的に見た場合 Wikipedia の方が質の面でも情報量の面でも上回っている。これはロングテール現象が生むシステムの根本的な特徴の一つだといえる。

もう一つの特徴は、Head が経済的な要素によって動かされるのに対して Tail は非経済的な要素（例えば個人の世評・評価など）によって動かされるという事。そのため Head ではプロが活躍するが、Tail はアマチュアの舞台であることが多い。これが最も顕著に現れている例の一つはブログだ。経済的な利益を得ることができるのはごく一部の Head のブロガーだけで、大部分のTail では殆どの場合経済的な利益でなく個人の知識や情熱が原動力となっている。

ロングテール現象を理解する上でもう一つ重要なのが、Aggregator だ。彼らは膨大な Tail の情報を分かりやすくまとめ、そしてユーザーの必要とする情報を探し易くする役割を担う。Google や Amazon 、eBay や iTunes などは Aggregator のいい例だ。サーチ、カテゴリー、ランキング、ユーザーによるコメント、Recommendation システムなどは膨大な情報を整理するのに頻繁に使われるテクニックだ。

ビジネスだけでなく、ロングテールは例えば生活・文化に関連する現象を理解するのにも役立つ。是非お勧めする一冊だ。

The future of weather.com in Web 2.0 style

Wed, 20 Sep 2006 20:40:00 GMT

I listened to this podcast on IPv6 a while ago. It talks about a lot of things, but in particular talks about an interesting experiment to generate a micro weather map by assigning IPv6 addresses to the wipers of taxi cabs. The idea is to correlate the location of each cab with the movement of the wipers to figure out the areas that are raining. I think that's a pretty interesting experiment.

My coworker told me a somewhat related story. Apparently one can buy a device that can measure certain weather conditions (temperature, etc), install it at home, and broadcast the result over the Internet. This becomes very interesting if enough houses install the device, since we'll then be able to generate a real-time weather map that are more accurate and more micro in scope than the traditional weather map we get today.

It's interesting to think of these examples as Web 2.0 style experiments, because they take advantage of "collective intelligence" in a fundamental way. This allows them to provide better services than a single provider can provide.

Web 2.0 による次世代の天気図

Wed, 20 Sep 2006 20:39:00 GMT

伊藤洋一のビジネストレンドで IPv6 についての Podcast を聞いた。面白いと思ったのは、タクシーを使って天気図を作るという実験の話。ワイパーに IPv6 のアドレスを割り当て、タクシーのロケーションとワイパーの動きのデータを利用してリアルタイムの天気図を作るというアイデアだ。これなら細かい、しかも実用的な天気図を描くことができる。

似たような話を同僚から聞いたことがある。彼によると、天候に関するデータを測定する器具を個人で簡単に買えるそうだ。そしてそれを家の外に設置して、インターネットでデータを配信することもできる。一人ではあまり役に立たないかもしれないが、もし多くの人がこの器具を設置したらワイパーの実験と同様リアルタイムの天気図が描けるわけである。

こうした実験はすごく Web 2.0 的だと思う。一つの組織に頼るのではなく、皆から少しずつデータをもらって大きなサービスを作り上げるという考え方は、Web 2.0 の基本的な考え方の一つだ。

Experimenting with Amazon aStore

Tue, 19 Sep 2006 02:16:00 GMT

I finally had a chance to play with aStore, Amazon's personalized online store. This is an interesting experiment that takes full advantage of Amazon's incredible selection of products. What I liked the most about aStore is that it can be created without any programming or technical knowledge. Take a look at what I came up with:

Keiji Store

Since aStore is still in beta, there are many aspects that can be improved, including:

The number of personal category is limitted to one, and the total number of item is limitted to only nine (but come to think of it, this may be a good strategy to increase the quality of selection)
Item cannot be rearranged
Personal comment is limitted to 128 characters

Despite these shortcomings, aStore can potentially become bigger by incorporating community-building features and other Web 2.0 elements.

In addition to aStore, Amazon recently has provided other interesting web services.

アマゾン aStore を作ってみた

Tue, 19 Sep 2006 02:13:00 GMT

最近アマゾンでオリジナルのオンラインストアを作れる事を知ったので、時間を作って試してみた。品物の豊富なアマゾンならではの機能だ。気に入ったのは技術的な事を知らなくても簡単にオンラインストアを作ることができる事。

Keiji Store

しかしまだベータ版なので、使い勝手や機能の面ではまだまだ改善の余地がある。例えば：

自分のカテゴリーは一つしか作ることができず、アイテムの数も限られている（ただ数が限られているために載せるアイテムをしっかり選ぶので、考えようによってはプラスかも）
アイテムの順序を変えることもできない
自分のコメントも 128 文字しか入れることができない

等など。しかし将来的にはストアに対するフィードバックやアイテムへのコメントなどコミュニティー機能を加えればかなり大きくなるのではないかと思う。

アマゾンはこの他にも色々なウェブサービスを公開するなど、最近面白い事を試している。

The World is Flat

Wed, 30 Aug 2006 08:06:00 GMT

I saw a great Webcast presentation by Thomas Friedman, the author of The World Is Flat. It offered interesting opinions and perspectives that were not covered in the book. There were also interesting questions from the audiences during Q&A.

As an interesting example of the Flat World, Friedman talked about an interesting experiment conducted by McDonald's. In the new experimental system, drive-through orders are not taken not by the workers in the store, but rather by someone else in a different location. The order is then routed back to the store where it will be fulfill and handed to the customer. The benefit of the new system? Lower rate of errors as a result of the separation of responsibilities. This kind of system was difficult to implement before the Flat World, but not in the new Flat World.

Friedman also talked about the shift of power from countries to companies to individuals. He gave a great example of how Southwest Airlines put this concept into action. About a year ago Southwest implemented a new system in which passengers can print their own boarding passes at 12:01am on the day of the departure. As a frequent Southwest flyer, I thought this was convenient, but didn't quite understand why Southwest did this. I always though Southwest's A-B-C seating system encouraged passengers to arrive to the airport early, which increases the likelihood of on-time departures. By allowing passengers to print their own boarding passes, I felt that Southwest passengers no longer have any incentive to arrive early. Friedman pointed out, however, that Southwest essentially turned its customers to its employees, ticket agents. This was a great example of the shift of power from a company to individuals. This is a win for customers as we don't need to waste time in the airport, and it is also a win for Southwest as it can lower its operation cost. As long as it doesn't cause any negative side-effects, such as more frequent flight delays, this system is a win-win for both Southwest and its customers. Judging from the fact that Southwest now allows boarding passes to be printed 24 hours before the flight (as opposed to 12:01am, which can be inconvenient), I think it is reasonable to believe that the system must be working well.

There were many interesting questions from the audiences during the Q&A session. During his presentation Friedman mentioned that Bill Gates had some comments, so there was a question about what those comments were. According to Friedman, Bill Gates thought that the book put too much emphasis on Open Source, which is understandable given that it threatens Microsoft's software business. Perhaps I'm biased, but I actually share Bill's viewpoint. Bill Gates also talked about the other half of the world that is not flat. Obviously Bill is very active on his foundation work, and I'm sure he's well aware of the issue.

There was an interesting question about education, a topic which Friedman discussed in depth in the book. Given that U.S. is not well-prepared for the Flat World, an audience asked what we can do in this field. Friedman offered two answers. The first is the importance of parenting. He talked about an interesting story of a school teacher in a PTA conference. The first family, an Asian family, asked the teacher to increase the amount of homework so that their child will study more. The second family, a European family, asked the teacher to put more emphasis on science education. The third family, from U.S., asked the teacher to reduce the amount of homework because their child is busy with soccer practice, piano lesson, and also, their child need time to chill :) In the Flat World, people in U.S. compete with people from countries with more discipline on education. The second point Friedman mentioned was the importance of the "ability to learn how to learn". He pointed out that because of the fast pace of change in the Flat World, the ability to self-learn new skills is more important than ever.

Friedman then talked a bit about U.S. immigration policy. Though U.S. is a country which benefited greatly from open immigration policies, the situation has changed since 9/11. Friedman fears that closing the country from immigration means no exchange of ideas, no collaboration, and no development of friendship. He is concerned about closing off what he calls "America's DNA" in the Flat World. At one point he mentioned that Ph. D students should receive diplomas with Green Cards stapled, which drew applause from the audience :)

I'll quickly recap Friedman's 10 Flatteners:

1. Windows: it became a platform on which individuals can author information
2. Netscape: it popularized the Internet as a platform to exchange information, and the dot-com era that followed caused over-investment of $1 trillion in building fiber optic network (which of course became the foundation of the Flat Wolrd)
3. Workflow: offered a way for applications to talk with each other

Based on those Flatteners came:

4. Outsourcing
5. Off-shoring
6. Uploading (Open Source, blogging, podcasting): this is the Architecture of Participation in Web 2.0
7. Supply-chaining: Friedman pointed out that "Wal-Mart is one of the largest companies in the world, but it doesn't make anything"
8. Insourcing: UPS insources Toshiba's technical support, Nike's online store, and Papa John’ logistics
9. Web search: individuals have more power to find information than ever before
10. Wireless and Mobility

The Future of Programming: Less is More

Tue, 29 Aug 2006 20:23:00 GMT

I found an interesting article on the future of programming. In my opinion it is biased toward dynamic languages (naturally so, as the article is based on a conversation with the creator of Ruby on Reils). Though I don't believe that object-oriented productivity languages like C# and Java will become less relevant any time soon, I do believe that there are niche spaces where dynamic languages can thrive.

In my view dynamic languages work great for writing small applications and plug-ins. The latter is particularly interesting as more Web applications make available their APIs. Today Google Maps and Virtual Earth are two of the most popular applications with public APIs. In the future we'll see more of them, and dynamic languages will be the glue that bind them together.

What is Web 2.0?

Tue, 22 Aug 2006 21:37:00 GMT

We see and hear the term "Web 2.0" being used (and misused) so often these days. Web 2.0 is used to describe everything from next-generation of software services to a bullxxxx generator :) To answer the question of "What is Web 2.0?", Tim O'Reilly, who introduced the term, wrote up an excellent article on the subject. Below is my summary of this article.

I think the most fundamental concept behind Web 2.0 is the notion of "harnessing the collective intelligence". This means that Web 2.0 services get better as more people use them. This is sometimes known as the network effects, where the usefulness of the service increases exponentially rather than linearly as the number of users increase. To enable harnessing collective intelligence, Web 2.0 apps are built on the architecture of participation. A simple example is eBay. Once eBay was successful in building a critical mass of buyers and sellers, it automatically drew more and more buyers and sellers. This is network effect.

Another fundamental notion of Web 2.0 is that "data is the next Intel Inside". The observation here is that all successful Web 2.0 services are backed by specialized database, from Google's web crawl to Amazon.com's database of products and community to Wikipedia's knowledge base. This makes sense from a business point of view, because specialized database serves as a long-term sustainable strength of these services.

Other aspects of Web 2.0, such as the "Long Tail", the "perpetual beta", and rich user experience, are also interesting traits of Web 2.0 apps.