How do you organize a group of individuals

一群人究竟如何搞定事情？

so that the output of the group

你如何組織一群人，

is something coherent and of lasting value,

讓這個團體的產出，

instead of just being chaos?

帶來某種一致性與延續性的價值。

And the economic framing of that problem

而並非只是一團混亂？

is called coordination costs.

把這個難題用經濟學的架構/術語來解釋

And a coordination cost is essentially all of the financial

我們稱之為「協調成本」。

or institutional difficulties in arranging group output.

協調成本基本上是安排群體產出成果時，

And we've had a classic answer for coordination costs,

所面臨到的財務與機構/組織問題。

which is, if you want to coordinate the work of a group of people,

對於協調成本有一個古典的答案來回應，

you start an institution, right? You raise some resources.

那就是：如果你想要協調一群人順利產出成果

You found something. It can be private or public.

那就發起一個組織吧。沒錯吧？你募集一些資源

It can be for profit or not profit. It can be large or small.

找到某些東西。這個組織可以是私人的或公共的

But you get these resources together.

它可以是營利或非營利組織；大機構或小型組織。

You found an institution, and you use the institution

但是當你把這些資源湊在一起

to coordinate the activities of the group.

你創立了一個機構，你運用這個機構來

More recently, because the cost of letting groups

協調一群人的活動。

communicate with each other has fallen through the floor --

而隨著近年來，形成團體彼此相互溝通

and communication costs are one of the big

的成本下跌到不像樣的程度

inputs to coordination -- there has been a second answer,

溝通成本比重上佔大部分

which is to put the cooperation into the infrastructure,

於是協調成本的第二個答案浮現了出來

to design systems that coordinate the output

那就是將彼此的合作建置在基礎架構中，

of the group as a by-product of the operating of the system,

設計出除了基本運作之外，同時能夠協調一群人

without regard to institutional models.

與產出成果的系統，讓一群人能夠順利地

So, that's what I want to talk about today.

產出成果，而不用訴諸於機構的模式。

I'm going to illustrate it with some fairly concrete examples,

所以這就是我今天想要談的內容。

but always pointing to the broader themes.

我將會舉一些相當具體的範例來作闡述，

So, I'm going to start by trying to answer a question

但是總會指向較廣的主題

that I know each of you will have asked yourself at some point or other,

我將會從試著回答一個問題來作為開始

and which the Internet is purpose-built to answer,

我知道各位在某個時刻也曾問過自己

which is, where can I get a picture of a roller-skating mermaid?

並且網際網路就是被用來回答這個問題的。

So, in New York City, on the first Saturday of every summer,

這個問題是：我要從哪找到一張美人魚溜直排輪的照片？

Coney Island, our local, charmingly run-down amusement park,

在紐約市，每年夏天的第一個星期六，

hosts the Mermaid Parade. It's an amateur parade;

康尼島，我們在地的、迷人的遊樂園區

people come from all over the city; people get all dressed up.

會舉辦美人魚遊行。這是一個業餘的遊行活動

Some people get less dressed up.

人們從紐約的四面八方湧來，盛裝打扮。

Young and old, dancing in the streets.

有些人盛裝打扮的比較清涼。

Colorful characters, and a good time is had by all.

年輕人跟熟男熟女，在街道上跳舞。

And what I want to call your attention to is not the Mermaid Parade itself,

所有人物都是色彩繽紛，大家都很享受這個時刻。

charming though it is, but rather to these photos.

我想要讓各位注意的不是美人魚遊行本身，

I didn't take them. How did I get them?

雖然它很迷人，我想要專注在這些照片上。

And the answer is: I got them from Flickr.

這些照片不是我拍的。我怎麼找到這些照片的？

Flickr is a photo-sharing service

答案是：我從 Flickr 上面找到了這些照片。

that allows people to take photos, upload them,

Flickr 是一個照片分享的服務

share them over the Web and so forth.

讓人們拍照、上傳照片，

Recently, Flickr has added an additional function called tagging.

在網路上彼此分享這些照片。

Tagging was pioneered by Delicious and Joshua Schachter.

最近 Flickr 增加了一個新功能：標籤（tagging）。

Delicious is a social bookmarking service.

標籤首先由Del.icio.us/Joshua Schachter所帶動

Tagging is a cooperative infrastructure answer to classification.

Del.icio.us 是一個社會書籤服務。

Right? If I had given this talk last year,

標籤是一種回答分類問題的答案：合作基礎架構。

I couldn't do what I just did,

如果我去年就做這場演講的話，

because I couldn't have found those photos.

我將無法展示那些剛剛秀的照片，

But instead of saying,

因為我找不到這些照片。

we need to hire a professional class of librarians

如果真的要作的話，

to organize these photos once they're uploaded,

我們需要雇一組專業的圖書館館員

Flickr simply turned over to the users

來組織這些上傳的許多照片，

the ability to characterize the photos.

Flickr簡單地讓使用者自己來管理

So, I was able to go in and draw down photos that had been tagged

它把標示照片的功能提供給了使用者。

"Mermaid Parade." There were 3,100 photos taken by 118 photographers,

所以我能在其中找到夠多上面標有我要的標籤的照片

all aggregated and then put under this nice, neat name,

「美人魚遊行」。共有 118 位攝影者，拍攝了 3100 張照片，

shown in reverse chronological order.

所有這些照片都被整理起來，放在簡潔有力的名稱底下，

And I was then able to go and retrieve them

以相反的時間順序來顯示。

to give you that little slideshow.

於是我可以搜尋、找到這些照片

Now, what hard problem is being solved here?

來作一場小小的照片展示。

And it's -- in the most schematic possible view,

我們現在正在解決的，是什麼樣的問題？

it's a coordination problem, right?

從最概略的可能觀點來檢視，

There are a large number of people on the Internet,

這是一個協調的問題。

a very small fraction of them have photos of the Mermaid Parade.

在網際網路上有非常多的人，

How do we get those people together to contribute that work?

其中一小群的人擁有美人魚遊行的照片。

The classic answer is to form an institution, right?

我們要如何讓那些人一起貢獻這個作品？

To draw those people into some prearranged structure

傳統的答案會是，成立一個組織吧？

that has explicit goals.

為了要吸引人們加入某些預先設計好的結構，

And I want to call your attention to

這些結構擁有明確的目標。

some of the side effects of going the institutional route.

請注意：

First of all, when you form an institution,

機構這種作法有一些副作用。

you take on a management problem, right?

首先，當你組成一個機構/組織時，

No good just hiring employees,

你馬上就有管理上的問題。

you also have to hire other employees to manage those employees

不只是聘僱員工而已。

and to enforce the goals of the institution and so forth.

你還需要聘僱另外一些員工來管理這些員工

Secondly, you have to bring structure into place.

並且強迫執行機構的目標...等等。

Right? You have to have economic structure.

再來，你必需要把結構放到實際的空間當中。

You have to have legal structure.

你還必需要有經濟結構。

You have to have physical structure.

你必需要有法律結構。

And that creates additional costs.

你必需要有實體結構。

Third, forming an institution is inherently exclusionary.

這些都造成了額外的成本。

You notice we haven't got everybody who has a photo.

第三，形成機構/組織天生就會排除異己。

You can't hire everyone in a company, right?

你注意到不是所有有照片的人都被納進組織中。

You can't recruit everyone into a governmental organization.

你沒有辦法在一個公司中僱用每一個人！對吧?

You have to exclude some people.

你也沒有辦法把所有人都僱用到政府裡。

And fourth, as a result of that exclusion,

你總是得排除某些人。

you end up with a professional class. Look at the change here.

第四，作為排除的結果，

We've gone from people with photos to photographers.

你將製造出一個專業階級。看看這個改變。

Right? We've created a professional class of photographers

我們從有照片的人們，變成了攝影師。

whose goal is to go out and photograph the Mermaid Parade,

我們創造了一個攝影師的專業階級

or whatever else they're sent out to photograph.

其目的是為了要去拍攝美人魚遊行

When you build cooperation into the infrastructure,

或其他任何指定要拍的東西。

which is the Flickr answer,

當你把合作建置在基礎架構中的時候，

you can leave the people where they are

這也是 Flickr 的答案，

and you take the problem to the individuals, rather than

你可以讓人們留在原地

moving the individuals to the problem.

或將這個問題帶到他們面前，讓每個人自己來解決，

You arrange the coordination in the group, and by doing that

而不是叫每個人移動來遷就這個問題。

you get the same outcome, without the institutional difficulties.

藉由這種安排，你在團體中設計協調的進行

You lose the institutional imperative.

讓你省去承擔機構的困擾、得到相同的產出。

You lose the right to shape people's work when it's volunteer effort,

你失去了機構的命令力量。

but you also shed the institutional cost,

當大家都是志工，你失去了型塑人們產出的權利，

which gives you greater flexibility.

但是你同時也減少了機構的成本，

What Flickr does is it replaces planning with coordination.

讓你擁有了更大的彈性。

And this is a general aspect of these cooperative systems.

Flickr 所作的是，它以協調取代了規劃。

Right. You'll have experienced this in your life

這是在這些合作系統中的一種普遍面向。

whenever you bought your first mobile phone,

在生活中你一定經歷過類似的片刻：

and you stopped making plans.

當你買了第一隻手機，

You just said, "I'll call you when I get there."

你便不再作規劃或計畫。

"Call me when you get off work." Right?

你只是說，「我到了再撥電話給你」。

That is a point-to-point replacement of coordination with planning.

「當你下班的時候 call 我」對吧?

Right. We're now able to do that kind of thing with groups.

那就是一種取代了規劃的、點對點的協調行動。

To say instead of, we must make an advance plan,

我們現在能夠跟一群人進行那樣子的協調。

we must have a five-year projection

不用再說，我們一定要作一個多先進的計畫、

of where the Wikipedia is going to be, or whatever,

我們必須要往後規劃五年的未來，

you can just say, let's coordinate the group effort,

或維基百科將會被帶往何處等等。

and let's deal with it as we go,

你可以只是說，我們一起來協調看看吧，

because we're now well-enough coordinated

我們邊做邊看好了，

that we don't have to take on the problems of deciding in advance what to do.

因為我們現在可以充分地彼此協調

So here's another example. This one's somewhat more somber.

不用再頭痛預先設想要做什麼。

These are photos on Flickr tagged "Iraq."

這裡是另外一個例子：這個例子更為陰暗。

And everything that was hard about the coordination cost

這些是 Flickr 網站上標註 Iraq 的照片。

with the Mermaid Parade is even harder here.

以協調成本來說，一切都非常困難

There are more pictures. There are more photographers.

比美人魚遊行還要困難的多。

It's taken over a wider geographic area.

有更多的照片，更多的攝影者。

The photos are spread out over a longer period of time.

照片涵蓋範圍包括更多地理區域。

And worst of all, that figure at the bottom,

拍攝時間跨越更長的一段時間。

approximately ten photos per photographer, is a lie.

而且更糟糕的是，看看底下的數字，

It's mathematically true,

「每個攝影者平均貢獻10張照片」這是假的。

but it doesn't really talk about anything important --

數學上來說是真的，

because in these systems, the average isn't really what matters.

但是沒有任何重要的意義

What matters is this.

因為這些系統中，平均數並不重要。

This is a graph of photographs tagged Iraq

真正重要的是：

as taken by the 529 photographers who contributed the 5,445 photos.

這是所有有標註 Iraq 的照片的貢獻數據圖

And it's ranked in order of number of photos taken per photographer.

是由529名攝影者，貢獻了5,445張照片。

You can see here, over at the end,

依照攝影者貢獻照片數目來加以排序。

our most prolific photographer has taken around 350 photos,

你可以看到在一端，

and you can see there's a few people who have taken hundreds of photos.

貢獻最多的攝影者拍攝了350張照片，

Then there's dozens of people who've taken dozens of photos.

一些人拍了將近數百張照片。

And by the time we get around here,

數十位攝影者拍攝上傳了數十張照片。

we get ten or fewer photos, and then there's this long, flat tail.

我們現在來看這裡，

And by the time you get to the middle,

我們看到十張或更少的照片貢獻者很多，有很長、平坦的尾部分佈。

you've got hundreds of people

接著我們走到圖表中間，

who have contributed only one photo each.

看到有數百人

This is called a power-law distribution.

每個人只有貢獻一張照片。

It appears often in unconstrained social systems

這就是所謂的冪次分佈。

where people are allowed to contribute as much or as little as they like --

常常在沒有設限的社會系統中出現

this is often what you get. Right?

當人們被允許貢獻多少都沒有關係時，

The math behind the power-law distribution is that whatever's in the nth position

這常常是我們所得到的結果。

is doing about one-nth of whatever's being measured,

冪次定律後面的數學原理就是：無論什麼在第 n 個位置

relative to the person in the first position.

其測量的結果是 1/n，

So, we'd expect the tenth most prolific photographer

相對於第 1 個位置的測量結果。

to have contributed about a tenth of the photos,

所以我們期待第十位貢獻最多的攝影者

and the hundredth most prolific photographer

他所貢獻的照片數量是第一名的 1/10，

to have contributed only about a hundred as many photos

而第 100 名的貢獻者

as the most prolific photographer did.

貢獻結果是 1/100

So, the head of the curve can be sharper or flatter.

相較於貢獻最多的攝影者。

But that basic math accounts both for the steep slope

所以這個曲線的頭部可以變得更為尖銳或平坦。

and for the long, flat tail.

但是基本數學說明了斜率

And curiously, in these systems, as they grow larger,

以及長長的、平坦的尾部。

the systems don't converge; they diverge more.

令人覺得有趣的是，在這些系統中，當他們規模成長，

In bigger systems, the head gets bigger

系統並不會收斂，反而更為發散。

and the tail gets longer, so the imbalance increases.

在較大的系統中，頭部變得更大

You can see the curve is obviously heavily left-weighted. Here's how heavily:

尾部則變得更長。不平衡的狀況更為增加。

if you take the top 10 percent of photographers contributing to this system,

你可以看到曲線很明顯地嚴重左傾；我們來看程度有多嚴重。

they account for three quarters of the photos taken --

如果你取前 10% 的攝影者的貢獻作品，

just the top 10 percent most prolific photographers.

它們佔了約 ¾ 的照片總數

If you go down to five percent,

僅僅只有前 10% 的攝影者的貢獻而已。

you're still accounting for 60 percent of the photos.

如果你取前 5% 的貢獻成果，

If you go down to one percent, exclude 99 percent of the group effort,

你就涵蓋了 60% 的照片。

you're still accounting for almost a quarter of the photos.

如果你取 1% 的成果，排除眾人 99% 的努力成果，

And because of this left weighting,

你仍然涵蓋了幾乎 ¼ 的照片總數。

the average is actually here, way to the left.

而且因為這樣的左傾，

And that sounds strange to our ears,

平均數實際就落在左側。

but what ends up happening is that 80 percent of the contributors

即便聽起來很怪，

have contributed a below-average amount.

最終實際的狀況是，80%的貢獻者

That sounds strange because we expect average and middle

只有低於平均數的貢獻。

to be about the same, but they're not at all.

這聽起來很怪，因為我們期待平均數與中數

This is the math underlying the 80/20 rule. Right?

應該是相同的；但是並不如此。

Whenever you hear anybody talking about the 80/20 rule,

這就是 80/20 法則後面的數學邏輯。

this is what's going on. Right?

每當你聽到有人談到 80/20 法則，

20 percent of the merchandise accounts for 80 percent of the revenue,

這就是實際的情形。

20 percent of the users use 80 percent of the resources --

20%的商品帶來 80% 的利潤，

this is the shape people are talking about when that happens.

20% 的使用者使用著 80% 的系統資源，

Institutions only have two tools: carrots and sticks.

這就是人們在討論時實際發生的資料形狀。

And the 80 percent zone is a no-carrot and no-stick zone.

機構只有兩種工具：胡蘿蔔跟棍子。

The costs of running the institution mean that you cannot

80% 的區域都是沒有胡蘿蔔跟棍子的地方。

take on the work of those people easily in an institutional frame.

運作機構的成本，意味著你沒有辦法

The institutional model always pushes leftwards,

將那些人們的成果簡易地用機構的框架來取得。

treating these people as employees.

機構模型總是會往左邊推擠，

The institutional response is,

希望將這些頭部的人們當作員工。

I can get 75 percent of the value for 10 percent of the hires -- great,

機構的反應是

that's what I'll do.

我可以從所僱用的人們當中的10%取得75%的價值的話...太棒了。

The cooperative infrastructure model says,

我就會這樣作。

why do you want to give up a quarter of the value?

合作架構的模型則是問：

If your system is designed

為什麼你希望放棄 ¼ 的價值？

so that you have to give up a quarter of the value,

如果你的系統被設計

re-engineer the system.

成你必需要放棄 ¼ 的價值，

Don't take on the cost that prevents you

那麼趕快去改造它吧。

from getting to the contributions of these people.

別讓成本阻礙了你

Build the system so that anybody can contribute at any amount.

不讓你從人們的貢獻中獲得成果；

So the coordination response asks not,

打造這個系統，讓任何人都能夠隨意貢獻

how are these people as employees, but rather,

所以協調並非這樣地回應提問，

what is their contribution like? Right?

這些人如何可以被雇為己用，而是換個方式提問：

We have over here Psycho Milt, a Flickr user,

他們的貢獻長什麼樣子？

who has contributed one, and only one, photo titled "Iraq."

以這張 Flickr 使用者 Psycho Milt 的照片為例，

And here's the photo. Right. Labeled, "Bad Day at Work."

他只有貢獻一張照片，只有一張照片標註著 Iraq。

Right? So the question is,

就是這張照片。名稱寫著：工作不順的一天。

do you want that photo? Yes or no.

所以問題是：

The question is not, is Psycho Milt a good employee?

你想要這張照片嗎？或許想，或許不想。

And the tension here is between institution as enabler

這個問題不是，Psycho Milt 是不是一個好的員工？

and institution as obstacle.

緊張關係就存在於，機構到底是一個促成者，

When you're dealing with the left-hand edge

還是一個阻礙者。

of one of these distributions,

當你在處理左側邊緣的這些資料

when you're dealing with the people who spend a lot of time

這些分佈當中的其中一筆資料時，

producing a lot of the material you want,

當你在跟花了很多時間的這些人們

that's an institution-as-enabler world.

他們生產一大堆你所需要的素材，

You can hire those people as employees, you can coordinate their work

那就是機構作為促成者的世界。

and you can get some output.

你可以把那些人都聘作員工，你可以協調他們的工作

But when you're down here, where the Psycho Milts of the world

而且你可以得到某些產出。

are adding one photo at a time,