【双语阅读】A PLAN TO SCAN 谷歌为何要“读”万卷书？

chijiaoshunv

By Richard Waters in San Francisco 2009-08-19


A PLAN TO SCAN

A1960 sociological study of female Finnish students or an 1894 handbook on how to play cricket are probably at the top of no one's poolside reading list this year.

Long out of print, such works are more likely to be gathering dust in attics, languishing forgotten at the backs of people's bookshelves or, as in the case of these two volumes, mouldering in the Harvard and Wisconsin university libraries respectively. Of the estimated 40m different books held by US libraries, well over half are unlikely ever to find their way back into a publisher's favour.

That makes an effort by Google, to burrow deep into the leading US research libraries to make digital copies of all the works it can lay its hands on, seem both ambitious and quixotic. The project, begun nearly five years ago, has also started scanning out-of-copyright works from libraries in other countries. A digital archive of all extant books - even ones in which few people are these days likely to show much interest - is carrying the internet company's mission to "organise the world's information" to the extreme.

Yet this mountain of fading literary oddments is now at the centre of a fierce debate in the book world that is about to come to a head.

After facing copyright lawsuits in the US over the digitisation project, Google reached a settlement last year that seemed to have something for just about everyone: publishers and authors, because it gives them a chance to make money from longforgotten works; public and university libraries, as it provides them with a way to leap beyond their dead-tree stacks into the digital age; and readers, to whom it brings access to millions of works that would otherwise have remained out of reach.

But this agreement with the US book industry, which awaits court approval, has stirred up the sort of passions that always attach to books, those most cultural of manufactured objects. In particular, the deal has provoked the fear that a more centralised industry will arise as publishing turns digital, upending checks and balances put in place over decades.

"The book world has done really well out of decentralisation - anyone who has ideas, or access to a printing press, can take part," says James Grimmelmann, associate professor at New York Law School, a leading critic of the settlement. Giving Google too much power over old, out-of-print works, he adds, could set the stage for its dominance of the broader digital book market: "Control over the past will translate into control over the future of books."

The US Department of Justice has taken such concerns seriously enough to launch an investigation into the competitive implications of the settlement: it is due to submit its views to the court considering the case in the middle of next month. Before that, the European Commission has called its own hearing on the issue, to consider the impact on Europe's book industry and authors' rights.

The main focus of the settlement falls on out-of-print books that are still in copyright. These works probably account for 60 per cent or so of all books in the US, making them a massive - if heavily underused - intellectual resource.

While Google's initial go-it-alone approach to digitising these works brought angst and lawsuits, the accord has turned it into an ally of the American book world. Unless copyright owners opt out of the plan, a Book Rights Registry to be run by representatives of the publishers and authors will have the power to license digital rights for all out-of-print books in the US to Google.

Google will then make parts of these works available through its search service, sell subscriptions to the entire database to university libraries and others - every library in the US will be offered a single free terminal to tap into the treasure trove - and sell access to full versions of individual works hosted on its computers. It will keep 37 per cent of the money from these sales, passing the rest to the registry to be paid out to copyright holders.

The undertaking is set to cost "hundreds of millions of dollars", says Dan Clancy, head of the Google Books effort. Yet there is little business in old books: second-hand volumes are estimated to account for less than $1bn of the $25bn US books market. The scale of the ambition makes it the sort of thing that only a Google would contemplate - or be able to afford.

David Balto, a former justice department lawyer, argues that any antitrust concerns are dwarfed by the benefits the settlement will bring. "What Google is doing is incredible - from a competition policy perspective, you don't want to punish people who are risking millions of dollars doing things like this that haven't been done before," he says.

Even the settlement's critics admit that it will bring immediate and substantial benefits, making millions of books widely available in the US for the first time. Yet its potential long-term impact on the shape of the digital book market has guaranteed that the settlement will attract close regulatory scrutiny, whatever its immediate attractions.

Critics fear that two aspects in particular could hand Google too much power, while also leaving a coterie of publishers and authors with disproportionate sway over setting prices for digital works, to the detriment of readers.

The first concerns the exclusive right that Google would have to distribute digital books whose copyright holders cannot be traced. These so-called "orphan works" may make up a large portion of all out-of-print tomes: Paul Courant, head of the University of Michigan library, estimates that they amount to 1m-2.5m of his collection of 8m volumes.

Congress has failed in its own efforts to free up these works so they can be sold without the risk of claims later from the copyright owners. It is a peculiarity of class action law in the US, though, that private legal action can achieve a result that has eluded Congress: since Google and the new books registry would be free to sell works whose owners did not actively opt out of the court-approved settlement, they would assume a right not available to anyone else.

But even if Google is left as sole distributor of orphan works, do the benefits outweigh antitrust worries? "Google is certainly going to be in a position of power in out-of-print books - but out-of-print books aren't exactly hot commercial properties," says Mr Courant. Balanced against that are the benefits to readers: "Being able to use these orphan works is much, much better than nothing."

Opponents say this understates the potential value to Google in the long run. Having the world's most comprehensive collection could make it the default first choice for book buyers, overshadowing Amazon.com's claim to be the world's biggest bookstore. "You're much more likely to turn to Google first because they'll have many more titles," says the law school's Mr Grimmelmann.

The international dimension to the debate over orphan works has also started to resonate, particularly in France, where a lawsuit against Google brought by local publishers is due to be heard next month.

Under the Berne convention, a long-standing international copyright agreement, copyright owners do not have to register in every country in order to protect their rights. The opt-out provision of the US settlement appears to fly in the face of that agreement by pre-empting the rights of anyone who does not come forward.

The publicity surrounding the case, and the creation of a single registry to administer rights, should encourage more rights holders outside the US to come forward, says Mr Clancy at Google. But with some European publishers already suspicious of having their rights circumscribed by American litigation, a visceral opposition has been building - particularly since the benefits from the settlement will accrue only to people in the US.

A second controversy surrounds the intended Book Rights Registry. Similar agencies representing the collective interests of artists are familiar in other parts of the media industry, for instance, the music world. But these typically are the creation of a legislative process or operate under close antitrust scrutiny. The settlement tries to combine two conflicting objectives - to maximise the revenues to authors and publishers while ensuring the widest possible access to the out-of-print works. Whether the complex system of incentives it creates can have the desired effect is a source of considerable unease.

"The library subscription could be excessively expensive," says Mr Courant in Michigan, reflecting a widely held concern. Gary Reback, a Silicon Valley antitrust lawyer, adds that the registry may have an incentive to license its book rights only to Google in order to keep prices up, rather than encourage competing distributors.

Countering this, Mr Clancy contends that Google's business model is based on obtaining the widest possible distribution: "Google's interest is to make it cheap." Even if libraries do not buy a subscription, he adds, the terminal they will receive for visitors to access Google's digital files will leave them better off than now.

W ith scrutiny intensifying on both sides of the Atlantic, a moment of truth is at hand for Google and its new allies (including Pearson Education and Penguin, sister companies of the Financial Times). They can push ahead with their settlement and risk provoking a backlash. Or they can try to adjust the terms to defuse some of the criticisms.

Those changes could be relatively easy to make, say opponents. Representatives of wider interests, such as libraries and readers, could be included on the book registry to prevent it limiting distribution only to Google or seeking excessive prices, says Peter Brantley, director of the Internet Archive, a non-profit organisation that is working on a digital archive of its own.

The court that is due to approve the class action settlement could also find ways to extend the "orphan works" protections to distributors other than Google, says Randal Picker, a law professor at the University of Chicago - though legal opinions are divided on whether this is possible. Google itself says it supports the idea of legislation to resolve the problem.

With the Department of Justice set to issue its verdict in less than a month, its behind-the-scenes discussions with many of the interested parties have been intensifying, according to people involved. There is so far no public indication that any voluntary changes to the complex book settlement will be forthcoming. But it seems increasingly likely that adjustments will be needed if the millions of tracts, treatises, thrillers and tragedies already embedded in Google's vast memory bank are once more to see the full light of day.

The market in prospect

Now for a new - or very old - type of browser war . . .

Behind Google's efforts to win friends and mollify critics in the book world lies a simple message: its vision for digital books is at least more appealing than that advanced by Amazon.com.

Though digital editions of books are still only a tiny part of the overall market, Amazon has created a model for how this business might work. Its Kindle reader and linked digital store, with books sold only in its proprietary format, echo the iPod/iTunes model with which Apple conquered digital music. Amazon's pricing power and tight control are starting to stir up the same concerns among publishers and booksellers that Apple aroused in the music industry.

Dan Clancy, the former rocket scientist who is in charge of Google's books project, cannot be drawn into mentioning the rival's name but says: "If there's a single player you should be concerned about in the digital books market, it's not Google." In Google's vision, books dematerialise and move into the "cloud" - they sit as digital files in its data centres and can be accessed on any internet-connected device, not tied to a single gadget like the Kindle.

The company's bet, says Mr Clancy, is that just as consumers have shown they want to download music and manage it on their home computers, they would rather browse a giant bookshelf in the sky when they are looking for something to read - and would rather have a choice of device on which to read it.

The works it is copying from universities create a foundation for this virtual bookshelf. They cannot be downloaded but there will be limited powers to copy and paste sections of works. Through deals with publishers, Google hopes to extend this approach to encompass new books, which have far greater commercial potential.

In another attempt to set itself apart from Amazon, Google is also seeking to position itself as an ally of traditional booksellers - though given the fear and uncertainty stirred up by the move towards digital books, this is proving a tough sell. In the future, while the "cloud books" reside on Google's servers, other retailers will be able to sell access to them. Eventually, says Mr Clancy, those digital rights could be sold through brick-and-mortar booksellers.

"Our core business is not selling books," he says. "Our core business is search and display. We will work well with people whose core business is selling books."

chijiaoshunv

在人们今年池边消遣读物的清单上，1960年对芬兰女学生的社会学研究或1894年出版的如何打板球的手册一定不会排在首位。

这些已绝版很久的著作更有可能躺在阁楼上积灰，在人们书架的角落渐渐被遗忘，或者像这两本书那样，分别在哈佛和威斯康辛大学图书馆里朽烂。估计美国各家图书馆馆藏不同书目4000万册，其中一半以上再也不可能重新被出版商相中。

这让谷歌的一项努力看起来既雄心万丈又不切实际。谷歌对美国顶尖研究图书馆进行深度挖掘，为它能得到的所有著作制作数字副本。该项目差不多始于五年前，现已开始从其他国家的图书馆扫描版权过期著作。一个所有目前存世书籍——哪怕是现如今几乎没什么人会感兴趣的书籍——的数字档案库正将这家互联网公司“整合全世界的信息”的使命推向极端。

然而，大量正逐渐被人遗忘的文学残品现处于将要濒临危急关头的图书界激烈争论的中心。

在美国因数字化项目面临版权诉讼后，谷歌去年达成一项似乎对所有人都有一定意义的和解协议，即出版商和作者，公共和大学图书馆，因为前者因此能有机会用长期被人遗忘的著作来赚钱，后者因此有办法跳过死的木头书架而进入数字时代；还有读者们，他们因此可以读到原本触不可及的数以百万计的著作。

但与美国图书出版业达成的这一有待法院批准的协议，已激起了人们的激情，这种激情总是依附在图书这种最具文化性的产品上。尤其是，该交易引起人们担心，认为随着出版业转向数字，颠覆数十年来建立起来的制衡机制，一个更集中化的行业将会出现。

“图书出版界确实真正从分散化中得益——只要有思想，或能使用印刷机，任何人都能参与，”纽约法学院(New York Law School)副教授詹姆斯·格里梅尔曼(James Grimmelmann)表示，他也是该协议的主要批评者。他补充说，给谷歌太多老旧绝版著作的权利，可能为其在范围更广的数字图书市场占据支配地位创造条件。“对过去的控制将转化成对图书未来的控制。”

美国司法部对此给予了足够的关切，发起了一场针对该协议竞争性后果的调查：下月中旬，司法部将向主审该案的法庭递交自己的看法。在此之前，欧盟委员会已就这一问题召开了听证会，以考虑此举对欧洲图书行业和作者权利的影响。

协议主要焦点在于仍在版权期内的绝版图书。这些著作占美国所有图书的60%左右，使它们成为庞大的——如果使用严重不足——知识资源。虽然谷歌最初将这些著作数字化的单干方法引起了担心和诉讼，但该协议使之成了美国图书界的盟友。除非版权所有者选择不参与该计划，否则由出版商和作者的代表管理的图书版权登记处(Book Rights Registry)将有权把美国境内绝版图书的电子版权授给谷歌。

接着谷歌将通过其搜索服务，让这些著作的一些部分开放，将整个数据库签售给美国境内大学和其他院校——为美国境内每座图书馆提供一个利用这些宝藏的免费终端——并出售在其电脑上的完整版单件著作的访问权。谷歌将保留这些销售所得的37%，其余部分给登记处支付给版权所有者。

chijiaoshunv

谷歌图书举措负责人丹·克兰西(Dan Clancy)表示，这项事业将耗费“数亿美元”。然而旧图书没有什么生意可言：估计二手书籍在价值250亿美元的美国图书市场中占不到10亿美元。这种抱负的规模决定了它是那种只有谷歌会去考虑——或能够负担得起的事。

前司法部律师戴维·巴尔托(David Balto)坚持认为，与该协议将带来的好处相比，任何反垄断担忧都不值一提。“从竞争政策角度来看，谷歌正在做一件不可思议的事情，对于冒数百万美元风险做一件以前从未做过的事的人们，你是不会加以惩罚的，”他表示。

即便抨击该协议的批评人士也承认，那将带来可观的直接利益，在美国首次让成百万图书可以在很广的范围内被人们得到。然而，它对数字图书市场形成的潜在长期影响决定了该协议将招致监管机构的密切审查，不论它有什么立竿见影的吸引力。

批评人士尤其担心在两方面可能给了谷歌太多权利，同时也在数字著作定价方面留给了出版商和作者群不成比例的影响力，会有损读者利益。

第一点担心的是谷歌将拥有独家销售无法找到版权所有者的数字著作的权利。这些所谓的“孤品”绝版可能占了所有绝版大部头著作中的很大一部分：密歇根大学图书馆馆长保罗·库朗(Paul Courant)估计，在他的800万册馆藏书中，这些书共计有100万至250万册。

国会曾试图解放这些著作，以便销售它们的时候能不冒日后遭版权所有人索赔的风险，但这些努力失败了。但私人诉讼最终可不经由国会干预是美国集体诉讼法的一个特色：由于谷歌和新的图书登记处能自由销售那些著作权所有者不积极选择脱离法庭批准协议的著作，他们将拥有任何不归属于任何一方名义下的著作版权。

但即便让谷歌成为这些孤品的独家销售商，益处大过反垄断担忧么？“谷歌肯定将在绝版图书领域发挥重要作用——但绝版图书并非正好就是热门商业财产，”库朗先生表示。与之平衡的是给读者们的利益：“能利用这些孤品著作比什么也没有要好得太多了。”

反对者们表示，这种说法低估了谷歌长期的潜在价值。拥有世界上最综合的藏书可能让它成为图书购买者们默认的首选，令亚马逊(Amazon.com)全球最大书店的提法相形见绌。“你更可能先去找谷歌，因为他们将有更多书目，”法学院的格里梅尔曼表示。

国际上对于孤品著作的争论也开始响应起来，尤其是在法国，当地出版商提起对谷歌的诉讼原定于下月听审。

根据历史悠久的国际版权协议《伯尔尼公约》，版权所有人不必在每个国家登记来保护他们的权利。美国协议的选择不参与条款先否定掉不主动响应者的权利，好像公然对抗该公约。

谷歌的克兰西先生表示，围绕该案的舆论和创造单一登记处来管理权利，应鼓励更多美国以外的权利所有者前来响应。但由于一些欧洲出版商已怀疑他们的权利被美国诉讼划定了界限，一种发自内心的反对已在积累——尤其是由于只有美国境内人士才能得到该协议的好处。

第二点争议围绕着计划成立的图书版权登记处。在媒介行业的其他领域，代表艺术家集体利益的类似机构是很常见的，如在音乐界。但这些通常是法律程序创建的，或在密切的反垄断监督下运作的。该协议试图将两个相互矛盾的目的结合起来——让作者和出版商的收益最大化，同时保证可能访问绝版著作的范围最大。它所创造的这一复杂激励系统是否能有期望中的效果，人们还不能完全放心。

chijiaoshunv

“图书馆签约订购可能非常昂贵，”密歇根的库朗先生表示。这点令人普遍担心。硅谷反垄断律师加里·瑞贝克(Gary Reback)补充说，该登记处可能有动机，为抬高价格，只将图书版权授予谷歌，而不是鼓励分销商参与竞争。

对此克兰西先生进行了反驳。他声称，谷歌的商业模式是基于获得最广销售可能性。“谷歌的兴趣在于让价格便宜，”他补充说，即便图书馆不订购，他们会接受让访问者访问谷歌电子文件的终端，并因此而更加受益。

由于大西洋两岸都在加强审查，谷歌及其新盟友（包括《金融时报》(Financial Times)姐妹公司培生教育(Pearson Education)和企鹅(Penguin)）即将迎来关键时刻。他们能在和解与风险激起反弹的情况下向前推进。或者他们可能设法调整条款以减少一些批评。

反对者们表示，那些改变可能相对容易做到。互联网档案馆(Internet Archive)主任彼得·布兰特利(Peter Brantley)表示，让更多不同的利益方，如图书馆和读者，能在图书登记处问题上参与其中，以防止限制只向谷歌进行销售。互联网档案馆是一家非营利性机构，正致力于它自己的数字档案馆。

芝加哥大学法律教授兰德尔·皮克(Randal Picker)表示，应批准该集体诉讼协议的法庭也能找到办法扩大“孤本著作”保护到除谷歌以外的销售商，尽管这是否有可能，法律上还存在意见分歧。谷歌自己表示，支持以立法解决问题。

随着司法部将在一个月不到的时间内发出裁决，据与此有密切关系的人士透露，正在加强与众多利益团体的幕后商讨。迄今为止，没有公开迹象表明任何一方将主动做出改变来接受这一复杂的图书协议。但如果数百万计的已内嵌在谷歌庞大存储体上的章节片段、专著、戏剧和悲剧作品再一次重见天日的话，似乎越来越有可能需要这一格局有所调整。

可期待的市场

现今新型的——也可以说很老式的——浏览器之争……

在谷歌努力争取支持和安抚图书界批评人士的背后透露出一条简单的信息：谷歌数字图书的愿景至少比亚马逊提出的更吸引人。

尽管数字版图书仍只是总体市场的一小部分，但亚马逊创造出了一个该业务可能运作的模式。其Kindle阅读器和相关联的数码商店，图书只以其专有格式出售，重复了苹果(Apple)征服数码音乐领域的iPod/iTunes模式。亚马逊的定价能力，以及牢固控制力，开始在出版商和书商中激起与苹果在音乐界所引起的一模一样的担心。

我们虽然无法让前火箭科学家、现负责谷歌图书项目的丹·克兰西提及竞争对手的名字，但他表示：“如果在数码图书市场你应该担心一个玩家的话，那它不是谷歌。”在谷歌看来，图书无纸化和转变为“云”——它们以数码文件形式存在于数据中心里，任何上网设备都能访问，不与像Kindle这样的单独某个设备关联。

克兰西先生表示，公司确信，就像消费者们表现出来的那样，他们想要下载音乐，并在他们的家庭电脑上进行管理，他们找点阅读内容时宁愿浏览一个庞大的“云中书架”——并且宁愿可以有选择在什么设备上进行阅读。

chijiaoshunv

谷歌从各所大学复制来的著作创造了一个虚拟书架。他们是不能下载的，但会有复制著作章节片段的有限权利。通过与出版商的协议，谷歌希望扩展这种方法，将商业潜力大很多的新书也包括到其中。

在另一项将自己与亚马逊区分开的尝试中，谷歌也在争取将自身定位为传统书商的盟友——尽管向数字图书的转变激起了担心和不确定，这说明人们很难接受。未来，虽然“云中图书”位于谷歌服务器上，但其他零售商将能出售这些图书的访问权。克兰西先生表示，最终，那些数码版权可能通过实体书商来进行销售。

“我们的核心业务不是卖书，”他表示。“我们的核心业务是搜索和陈列。我们将与核心业务是卖书的人良好合作。”


译者/红岭