A new entrant to the knowledge market
If you want to look up the 2004 Indian Ocean tsunami in an encyclopedia, where do you turn? Not to the Encyclopedia Britannica, which will take a long time to research, draft and approve an entry on the disaster, or even to Encarta, Microsoft's disk-based rival publication.
The only instant reference work is Wikipedia, a free online encyclopedia written by anybody who wants to contribute. If you look now (
www.wikipedia.org), there is a first draft of history, with a simulation of how the tidal wave spread across the ocean, a table of estimated deaths in different countries and links to entries on related subjects.
Wikipedia is a phenomenon. It was started three years ago as an experiment in spontaneous communal publishing and now holds 441,000 articles in English. Its editions in various languages receive 360m page views a month and it is steadily climbing the charts of popular internet destinations. Wikipedia articles often appear close to the top of Google searches on many topics.
The project is the most visible sign of the growth of “wikis” internet sites that are edited by groups rather than individuals (the name comes from the Hawaiian term for “fast”). Wikis are starting to achieve for publishing what the open source movement did for software development: they provide an unruly new form of competition for traditional publishers.
Yet Wikipedia has a credibility problem, as Larry Sanger, one of its co-founders, pointed out in an online article last week. Put simply, why should anything it contains be believed, given that entries can be written by anyone and are not checked by experts? As Jimmy Wales, Wikipedia's other co-founder, cheerfully admits: “Any article could be wrong at any given time”.
Mr Sanger, a philosophy lecturer who is no longer involved in Wikipedia, says the project has two flaws. The first is that, no matter how good the material on it may be, it is not trusted by academics, librarians or teachers because of its lack of peer review the fact-checking process under which articles are written by experts and reviewed by other experts before publication.
The second is the “dominance of difficult people” in the online community that runs Wikipedia. Potential contributors, including experts on topics, are put off by infighting and abusive behaviour among the core group of 400 enthusiasts who do the most editing. The underlying problem, says Mr Sanger, is the project's “anti-elitist” philosophy of not wanting to defer to academic expertise.
This creates obvious weaknesses. Entries are reviewed by others and can be “reverted” restored to the original if an error is spotted. But there is no guarantee that truth will triumph. Robert McHenry, a former editor-in-chief of the Encyclopedia Britannica, argued recently that Wikipedia's entry for Alexander Hamilton is not only flawed but has steadily deteriorated.
It also encourages bias. Wikipedia tries to present a “neutral point of view” but the selection of articles and material in them tilts towards the views of its core group. One example is the entry for McDonald's, which devotes a lot of space to criticisms by environmental and social activists but says little about how the company originally succeeded.
In many ways, however, Wikipedia is surprisingly good. Only a fool would turn to it to research an academic paper or trust all material without checking it. But used with discretion it provides a lot of information on a wide variety of topics. I would trust it, for example, to name the capital city of a US state, or as a guide on information technology matters.
There is method in Wikipedia's madness, too. Allowing anybody to contribute may produce uneven results but it has proved a fecund way of operating. Nupedia, Wikipedia's predecessor, was organised more traditionally, with peer review of all contributions, and it made far slower progress as a result: fewer than 30 Nupedia articles had been completed when it folded in 2003.
Above all, Wikipedia's growth shows that lots of basic information can be exchanged by people who know things because of where they live, their hobbies or their education. This is not specialist research-based knowledge but it does not need to be. A Wikipedia user is often looking for one or two facts or an introduction to a subject not an immersion in it.
Knowledge is like other goods and services: some consumers want the best item available, even at a very high price, while others will be satisfied with the cheap and cheerful. The combination of the internet and Wiki software is a disruptive technology: it allows Wikipedia's enthusiasts to produce something that is extremely cheap (free, in fact) and good enough to satisfy many people.
The remaining question is whether Wikipedia will be able to bridge the gap with the peer-reviewed world. Mr Wales says the Wikimedia Foundation, which oversees Wikipedia, could produce an edition with articles that have been reviewed by experts to enhance its credibility with librarians and teachers. The trick is to do so without alienating its productive, but turbulent, community.
If it achieves this, the publishers of other reference works should watch out. So far, they have been protected from competition by Wikipedia's odd constitution and unwillingness to co-operate with academia. When its enthusiasts learn the meaning of the word “compromise”, the open source onslaught will begin.
知识市场的网上百科
如果想在百科全书上查询2004年印度洋海啸，你会网上去查哪本百科全书？不要去查《大英百科全书》(Encyclopedia Britannica)，因为该书要花很长时间才能研究、草拟，并批准有关这场灾难的一个条目。甚至也不要去查微软(Microsoft)的光盘百科全书Encarta，这是与大英百科全书竞争的产品。
唯一的即时参考书是维基百科全书(Wikipedia)，这是免费的在线百科全书，其作者是任何想为之撰稿的人。如果你现在去查(
www.wikipedia.org)，就会看到这次灾难历史回顾的第一稿，其中包括对海浪如何卷过大洋的模拟图景、一张不同国家估算死亡人数的清单，以及相关主题条目的链接。
维基百科全书是个奇迹。三年前它作为自发公共出版业的一项实验问世，如今已有44.1万篇英文文章。不同语言版本的维基每月共获得3.6亿次点阅，在流行互联网站点排行榜上稳步攀升。在许多话题上，维基的文章经常出现在Google搜索结果的前列。
该项目是“维基”(wikis)网站壮大最明显的迹象。维基是一些由小组而非个人编辑的网站，这个名字来自夏威夷语中表示“快速的”一词。维基正开始对出版业产生影响，就如开放源运动对软件开发所产生的影响：它们对传统出版商构成了一种不受约束的新型竞争。
但正如拉里?桑格(Larry Sanger)近期在一篇网上文章中所指出的，维基百科全书存在一个可信度问题。桑格是维基百科全书的联合创始人之一。简言之，既然任何人都能为它撰写条目，而且这些条目未经专家审核，那么为什么要相信它包含的任何信息呢？维基百科全书的另一位联合创始人杰米?威尔士(Jimmy Wales)也乐于承认：“在任何一个时刻，任何文章都有可能是错的。”
桑格先生是一位哲学讲师，他已不再参与维基百科全书的工作。桑格先生表示，该项目有两处缺陷。首先，不论其中的内容可能有多么精彩，但因为未经同行评审，所以未得到学术界人士、图书管理员或教师的信赖。同行评审是事实查证过程，依照该过程，文章由专家撰写，并在发表前由其他专家评审。
其二，管理维基百科全书的网络圈子由“难缠人士主导”。这个核心群体由400名维基的热心拥护者组成，他们担任大部分编辑工作，但其内讧和粗鲁行为使一些潜在撰稿人(包括各种主题的专家)望而却步。桑格先生表示，根本问题在于该项目的“反精英”哲学，即不愿接受学术界专家的意见。
这产生了明显的缺陷。输入的条目由其他人审核，如果发现一个错误，词条会被“复原”，即回复原状。但还是无法保证真理就能获胜。《大英百科全书》前主编罗伯特?麦克亨利(Robert McHenry)最近声称，维基百科全书的“亚历山大?汉密尔顿”(Alexander Hamilton)词条不仅有错，而且错得越来越厉害。
维基百科全书还鼓励偏见。它试图展示一种“中立的观点”，但文章的选择及文中的材料则倾向于其核心集团的观点。“麦当劳”(McDonald’s)的词条就是个例子。这个词条中有大量篇幅是关于环境和社会活动人士对该公司的批评，但对公司最初是怎样成功的却说得极少。
但在许多方面，维基百科全书好得出人意料。只有傻瓜才会借助它去研究一篇学术论文，或不经核查就相信其中所有的材料。但只要使用得当，维基百科全书能提供题材广泛的大量信息。举例来说，我会用它查出美国某个州的首府，或作为一般指南来研究信息技术课题。
维基百科看似古怪，却自有其合理之处。允许任何人撰文可能会带来不一致的结果，但它已被证明是个多产的运作方式。维基百科的前身Nupedia是用更传统的方式组织编纂的，对所有撰稿进行同行评审，结果是，它的进展比维基百科缓慢得多：当Nupedia在2003年关闭时，其完成的文章不到30篇。
最重要的是，维基百科的成长表明，许多基本信息能在人们之间交流，由于人们生活的地方、兴趣爱好或所受的教育不同，因此知道不同的事情。这些固然不是通过专业研究而得到的知识，但其实也无需如此。维基百科的用户经常是寻找一两个事实或是对一个主题的介绍，而不是要深入探究某个主题。
知识就像其它商品或服务一样：一些消费者想要可提供的最好东西，即便要付出很高的价格，而其他消费者则满足于便宜而让人开心的东西。互联网与“维基”软件的结合是一项引起混乱的科技：它允许维基百科的热心拥护者们生产一些极其廉价(事实上是免费的)、同时又好到能令许多人满意的东西。
剩下的问题就是，维基百科能否填补同行评审的空白。威尔士先生说，负责维基百科的维基百科基金(Wikimedia Foundation)可以制作一个由专家审议的文章组成的版本，用来强化维基百科在图书管理员和教师心目中的可信度。诀窍在于，如何在这么做的同时，不疏远其多产但杂乱的群体。
如果它达到这一目标，那么其它参考书的出版者就要当心了。迄今为止，由于维基百科结构怪异、且不愿与学术界合作，因此一直未对这些出版商构成竞争威胁。但当维基百科的热心拥护者们悟出“妥协”一词的含义后，那么开放源式的攻击就将展开。
