The moral bias behind your search results | Andreas Ekström

翻译人员: Jingyao Xue 校对人员: Zhiting Chen

So whenever I visit a school and talk to students,

每当我参观一所学校， 和学生们交谈，

I always ask them the same thing:

我总是问他们同样的问题：

Why do you Google?

为什么你用Google？

Why is Google the search engine of choice for you?

为什么你首选Google 作为搜索引擎呢？

Strangely enough, I always get the same three answers.

奇怪的是，我总是得到 三个相同的回答。

One, "Because it works,"

第一个回答， “因为它是个管用的搜索工具，”

which is a great answer; that's why I Google, too.

这是最好的答案，也是 我使用Google的原因。

Two, somebody will say,

第二个回答是，

"I really don't know of any alternatives."

”我真的不知道其他 任何可代替的方法。“

It's not an equally great answer and my reply to that is usually,

这不是一个很好的回答， 而我的回应总是，

"Try to Google the word 'search engine,'

”尝试在Google上 键入搜索引擎，

you may find a couple of interesting alternatives."

你可以发现一些有趣的， 可代替的搜索方法。”

And last but not least, thirdly,

最后一个回答 并非不重要，

inevitably, one student will raise her or his hand and say,

如我所预料的， 一个学生举手说，

"With Google, I'm certain to always get the best, unbiased search result."

”使用Google， 我确信总得到 最好和客观公正的搜索结果。“

Certain to always get the best, unbiased search result.

一定得到最好和客观 公正的搜索结果。

Now, as a man of the humanities,

作为一个学习人文科学的人，

albeit a digital humanities man,

尽管我是研究科技 和人文科学间关系的，

that just makes my skin curl,

这个回答听起来 使我心里发毛，

even if I, too, realize that that trust, that idea of the unbiased search result

尽管我也知道， 这种对搜索结果的信任，

is a cornerstone in our collective love for and appreciation of Google.

是以我们共同喜爱和对 Google的认识为基础的。

I will show you why that, philosophically, is almost an impossibility.

我会证明给你看，这种想法 在哲学思维分析上几乎不存在。

But let me first elaborate, just a little bit, on a basic principle

但首先，我对网上查询的 基本原则说明一下，

behind each search query that we sometimes seem to forget.

我们有时似乎忘记，每个 搜索查询背后的原则。

So whenever you set out to Google something,

因此，每当你用Google 查询所需的东西，

start by asking yourself this: "Am I looking for an isolated fact?"

首先要问自己： “我要寻找个别的事实吗？”

What is the capital of France?

例如，法国的首都是哪里？

What are the building blocks of a water molecule?

水分子的构成要素是什么？

Great -- Google away.

使用Google搜索 都能找到这些答案。

There's not a group of scientists who are this close to proving

没有一个严密的 科学家小组去证明

that it's actually London and H30.

这些答案，实际上 会是伦敦和H30。

You don't see a big conspiracy among those things.

在这些答案中， 你不会找到什么阴谋。

We agree, on a global scale,

我们同意，在全球范围内，

what the answers are to these isolated facts.

这些个别事实通过 网上查询能找到答案。

But if you complicate your question just a little bit and ask something like,

但是，如果你搜索 一些稍复杂的问题，

"Why is there an Israeli-Palestine conflict?"

例如，“为什么出现以色列 和巴勒斯坦之间的冲突？“

You're not exactly looking for a singular fact anymore,

你不会只寻找单独 形式的个别事实，

you're looking for knowledge,

你要找的是知识，

which is something way more complicated and delicate.

这些知识会表现出很 复杂和微妙的形式。

And to get to knowledge,

为了获得这些知识，

you have to bring 10 or 20 or 100 facts to the table

你必须要搜索10，20或 100个相关的事实

and acknowledge them and say, "Yes, these are all true."

并认可这些事实，说： ”这些都是真实的。“

But because of who I am,

但是，取决于我是谁，

young or old, black or white, gay or straight,

年龄、肤色、性别取向，

I will value them differently.

人们以不同方式， 去评估这些事实。

And I will say, "Yes, this is true,

我会说，“这是真实的，

but this is more important to me than that."

但对我来说，这些事实 比那些更重要。”

And this is where it becomes interesting,

事情在这里变得微妙，

because this is where we become human.

你可以在这里发现人性。

This is when we start to argue, to form society.

当我们开始对这些事实 进行辩论，这就形成社会。

And to really get somewhere, we need to filter all our facts here,

要获得真实的信息，我们需要 对所有的事实进行过滤，

through friends and neighbors and parents and children

通过我们的朋友， 邻居，父母和小孩

and coworkers and newspapers and magazines,

还有通过同事， 报纸和杂志，

to finally be grounded in real knowledge,

最后会得到真实的知识，

which is something that a search engine is a poor help to achieve.

这些真实的知识， 很难依赖网上搜索引擎的帮助。

So, I promised you an example just to show you why it's so hard

让我举例证明

to get to the point of true, clean, objective knowledge --

要得到真实， 公正和客观的知识有多困难 －

as food for thought.

提供另一个观点供你们思考。

I will conduct a couple of simple queries, search queries.

在这里，我会进行一些 简单的搜索查询。

We'll start with "Michelle Obama,"

我们查询“Michelle Obama，”

the First Lady of the United States.

美国第一夫人。

And we'll click for pictures.

我们点击她的图片。

It works really well, as you can see.

你能看到，这个查询 方法很管用。

It's a perfect search result, more or less.

或多或少这是一个 完美的搜索结果。

It's just her in the picture, not even the President.

只有她在相片中，她的 总统丈夫不在相片中。

How does this work?

搜索引擎是如何工作呢？

Quite simple.

很简单。

Google uses a lot of smartness to achieve this, but quite simply,

Google使用大量智能技术 呈现搜索结果，但很简单，

they look at two things more than anything.

它们更多看两样东西比 任何其他东西。

First, what does it say in the caption under the picture on each website?

首先，每个网站上 的图片标题是什么？

Does it say "Michelle Obama" under the picture?

在图片下有显示 “Michelle Obama” 吗？

Pretty good indication it's actually her on there.

有的话会是很好的指标。

Second, Google looks at the picture file,

第二，Google显示图像文件，

the name of the file as such uploaded to the website.

该文件的名字会传到网站。

Again, is it called "MichelleObama.jpeg"?

被称为“MichellObama.jpeg”吗？

Pretty good indication it's not Clint Eastwood in the picture.

非常好的指标，它不是 Clint Eastwood的图片。

So, you've got those two and you get a search result like this -- almost.

你已经获得两个搜索结果，而且 你会得到其他像这样类似的结果。

Now, in 2009, Michelle Obama was the victim of a racist campaign,

在2009年，Michelle Obama 是一个种族主义运动的受害者，

where people set out to insult her through her search results.

有些人通过网上搜索结果， 有意对她进行侮辱。

There was a picture distributed widely over the Internet

有一张她的相片在 互联网上广泛散布，

where her face was distorted to look like a monkey.

她的脸被扭曲， 看起来像一只猴子。

And that picture was published all over.

这张相片全世界 被公布了。

And people published it very, very purposefully,

这些人很故意地 发布这张相片，

to get it up there in the search results.

登载在互联网上，在她 相关图片的搜索结果中。

They made sure to write "Michelle Obama" in the caption

这些人确信，要用 “MichelleObama”作标题，

and they made sure to upload the picture as "MichelleObama.jpeg," or the like.

而且，他们确信上传的图片 为“MichelleObama.jpeg"。

You get why -- to manipulate the search result.

你明白这个原因--- 操纵搜索结果。

And it worked, too.

而且，它也管用。

So when you picture-Googled for "Michelle Obama" in 2009,

当你Google搜索图片， ”MichelleObama“ 2009，

that distorted monkey picture showed up among the first results.

被扭曲了的猴子图片会 出现在第一个搜索结果中。

Now, the results are self-cleansing,

现在，这图片在 搜索结果中自我消失了，

and that's sort of the beauty of it,

这是Google的优点，

because Google measures relevance every hour, every day.

因为Google每小时， 每天都测量信息的相关性。

However, Google didn't settle for that this time,

但是， Google在这个事件上没有妥协，

they just thought, "That's racist and it's a bad search result

Google只是想，”这是种族主义 的表现，是糟糕的搜索结果，

and we're going to go back and clean that up manually.

我们用手工清除它。

We are going to write some code and fix it,"

我们会编写一些 代码并修复它。”

which they did.

修复这个恶作剧。

And I don't think anyone in this room thinks that was a bad idea.

我不认为，这里会有人觉得 这个主意不好，

Me neither.

我也不觉得。

But then, a couple of years go by,

但是，几年过去后，

and the world's most-Googled Anders,

世界上最多用Google 搜索的是Anders，

Anders Behring Breivik,

Anders Behring Breivik，

did what he did.

他为所欲为做他想做的事。

This is July 22 in 2011,

在2011年7月22日，

and a terrible day in Norwegian history.

在挪威史上是很可怕的一天。

This man, a terrorist, blew up a couple of government buildings

Anders是恐怖分子， 他炸毁了几栋政府大楼，

walking distance from where we are right now in Oslo, Norway

在挪威首都奥斯陆，离我们 现在所在地仅几步之遥，

and then he traveled to the island of Utøya

然后，他前往Utoya岛

and shot and killed a group of kids.

并开枪打死一群孩子。

Almost 80 people died that day.

近80人在那天丧生。

And a lot of people would describe this act of terror as two steps,

很多人描述他的 恐怖行为有两个步骤，

that he did two things: he blew up the buildings and he shot those kids.

他做了两件事：炸毁建筑物 和枪杀那些孩子。

It's not true.

这不是真的。

It was three steps.

该包括三个步骤。

He blew up those buildings, he shot those kids,

他炸毁那些建筑物， 他枪杀孩子，

and he sat down and waited for the world to Google him.

还有，他坐下来并等待 全世界的人用Google搜索他。

And he prepared all three steps equally well.

他同样出色地 准备了这三个步骤。

And if there was somebody who immediately understood this,

如果有人立刻明白他的目的，

it was a Swedish web developer,

这个人就是瑞典互联网的开发员，

a search engine optimization expert in Stockholm, named Nikke Lindqvist.

名叫Nikke Lindqvist，是斯德哥尔摩 的搜索引擎优化专家。

He's also a very political guy

他也是一个非常 政治化的家伙，

and he was right out there in social media, on his blog and Facebook.

他出现在社会媒体，博客 和Facebook上。

And he told everybody,

他告诉每一个人，

"If there's something that this guy wants right now,

“如果他现在想做一件事情，

it's to control the image of himself.

就是去控制他自己的图像。

Let's see if we can distort that.

让我们看看是否 能扭曲图像。

Let's see if we, in the civilized world, can protest against what he did

在文明世界里，让我们看看 能否反对他所做的事

through insulting him in his search results."

在他的搜索结果中侮辱他。“

And how?

该怎样做呢？

He told all of his readers the following,

他告诉所有的读者，

"Go out there on the Internet,

“在互联网上，

find pictures of dog poop on sidewalks --

找人行道上狗屎的照片 --

find pictures of dog poop on sidewalks --

找人行道上狗屎的照片 --

publish them in your feeds, on your websites, on your blogs.

把这些照片公布在食物上， 你的网站和博客上。

Make sure to write the terrorist's name in the caption,

务必写恐怖分子 的名字作为题目，

make sure to name the picture file "Breivik.jpeg."

务必给图像文件取名 为“Breivik.jpeg".

Let's teach Google that that's the face of the terrorist."

让我们教导Google， 恐怖分子的脸是怎样的。”

And it worked.

这个方法管用。

Two years after that campaign against Michelle Obama,

在那场用来反对Michelle Obama 的战役两年后，

this manipulation campaign against Anders Behring Breivik worked.

反对Anders Behring Breivik 的操纵战役有效。

If you picture-Googled for him weeks after the July 22 events from Sweden,

在瑞典7月22日的事件发生几周后， 如果你Google搜索Anders的图片，

you'd see that picture of dog poop high up in the search results,

你会看到有狗屎的相片 排在他的搜索结果前面，

as a little protest.

这作为对他的小抗议。

Strangely enough, Google didn't intervene this time.

奇怪的是，这次Google 没有干预这件事。

They did not step in and manually clean those search results up.

Google没有介入，并 手工清除那些搜索结果。

So the million-dollar question,

所以，重要的问题是，

is there anything different between these two happenings here?

这两个发生的事件在 Google上有什么不同？

Is there anything different between what happened to Michelle Obama

这两者之间有不同吗？ 发生在Michelle Obama的事，

and what happened to Anders Behring Breivik?

和发生在Anders Behring Breivik的事？

Of course not.

当然这是折然不同的事。

It's the exact same thing,

其实，这是同样的事情，

yet Google intervened in one case and not in the other.

但Google介入一个案件， 另一个不介入。

Why?

为什么？

Because Michelle Obama is an honorable person, that's why,

因为Michelle Obama是一个 体面和正直的人，这就是原因，

and Anders Behring Breivik is a despicable person.

而Anders Behring Breivik 是一个卑鄙的人。

See what happens there?

你都看到这里发生的事吗？

An evaluation of a person takes place

对一个人的评价。

and there's only one power-player in the world

在世界上，只有唯一 一个权力重要参与者，

with the authority to say who's who.

有权威决定谁是谁，

"We like you, we dislike you.

“我们喜欢你，我们不喜欢你。

We believe in you, we don't believe in you.

我们相信你，我们不相信你。

You're right, you're wrong. You're true, you're false.

你是对的，你是错的； 你是真的，你是假的。

You're Obama, and you're Breivik."

你是Obama，而你 是Breivik。”

That's power if I ever saw it.

我在任何时候都看到， 这就是权力。

So I'm asking you to remember that behind every algorithm

所以，我要求你记住， 每个计算机操作的背后

is always a person,

总有一个人，

a person with a set of personal beliefs

这个人有一套个人的信念

that no code can ever completely eradicate.

从来没有破译密码能 彻底根除这种信念。

And my message goes out not only to Google,

我发布这些信息表明， 不仅Google是这样，

but to all believers in the faith of code around the world.

而且世界上所有相信 计算机操作的人都这样。

You need to identify your own personal bias.

你需要确定自己 个人的偏见。

You need to understand that you are human

你需要明白，你是人

and take responsibility accordingly.

并承担相应的责任。

And I say this because I believe we've reached a point in time

我说这些，因为我相信 我们已经及时达成共识，

when it's absolutely imperative

这是绝对必要的，

that we tie those bonds together again, tighter:

我们又把这个纽带 更加拉紧在一起：

the humanities and the technology.

人文科学和科技，

Tighter than ever.

比以往任何时候更紧密。

And, if nothing else, to remind us that that wonderfully seductive idea

而且，如果没有其他东西， 提醒我们这个极其吸引人的

of the unbiased, clean search result

公正，干净的搜索结果的想法

is, and is likely to remain, a myth.

是一个神话。

Thank you for your time.

谢谢。

(Applause)

（鼓掌）