Frederic Kaplan: How I built an information time machine

譯者: NAN-KUN WU 審譯者: Willy Feng

This is an image of the planet Earth.

這是地球的照片

It looks very much like the Apollo pictures

看起來和廣為人知的阿波羅計劃中 拍到的照片很像

that are very well known.

看起來和廣為人知的阿波羅計劃中 拍到的照片很像

There is something different;

但其中有些差異

you can click on it,

你可以點擊它

and if you click on it,

如果你點擊它

you can zoom in on almost any place on the Earth.

你就可以放大地球上的幾乎每個地方

For instance, this is a bird's-eye view

舉例來說，這是洛桑聯邦理工學院的鳥瞰圖

of the EPFL campus.

舉例來說，這是洛桑聯邦理工學院的鳥瞰圖

In many cases, you can also see

很多時候你也可以從旁邊的街上

how a building looks from a nearby street.

看到某棟建築物的的長相

This is pretty amazing.

這真的很神奇

But there's something missing in this wonderful tour:

但這趟美妙的旅程還少了一樣東西：

It's time.

時間

i'm not really sure when this picture was taken.

我不確定這張照片是什麼時候拍的

I'm not even sure it was taken

我甚至無法確定這是在鳥瞰圖的當下拍的

at the same moment as the bird's-eye view.

我甚至無法確定這是在鳥瞰圖的當下拍的

In my lab, we develop tools

在我的實驗室的研發工具

to travel not only in space

不止用在空間的旅行

but also through time.

也用在時間的旅行

The kind of question we're asking is

我們想問的問題是

Is it possible to build something

有辦法做出過往時間的Google Maps嗎？

like Google Maps of the past?

Can I add a slider on top of Google Maps

有辦法在Google Maps上增加一個滑桿

and just change the year,

只要改變年份

seeing how it was 100 years before,

就可以看到100年前

1,000 years before?

1000年前的樣子嗎？

Is that possible?

這可能嗎？

Can I reconstruct social networks of the past?

我能夠重建過去的社群網路嗎？

Can I make a Facebook of the Middle Ages?

我可以建立中世紀時的Facebook嗎？

So, can I build time machines?

我能夠建造時光機嗎？

Maybe we can just say, "No, it's not possible."

我們可以說：「不，這不可能。」

Or, maybe, we can think of it from an information point of view.

或者，我們可以從資訊的角度來看

This is what I call the information mushroom.

我把這個稱之為資訊蘑菇

Vertically, you have the time.

縱軸是時間

and horizontally, the amount of digital information available.

橫軸是我們擁有的數位資訊

Obviously, in the last 10 years, we have much information.

很明顯的，在過去的10年中 我們擁有很大量的資訊

And obviously the more we go in the past, the less information we have.

也看得出來，越早以前資訊就越少

If we want to build something like Google Maps of the past,

如果我們想建立過去的Google Maps

or Facebook of the past,

或是過去的Facebook

we need to enlarge this space,

我們就必須擴大這個空間

we need to make that like a rectangle.

像是一個長方形

How do we do that?

這要如何做到呢？

One way is digitization.

一個辦法是數位化

There's a lot of material available --

有很多的材料可以用--

newspaper, printed books, thousands of printed books.

報紙、印刷書本

好幾千本印刷書本

我可以把它們全部數位化

I can digitize all these.

I can extract information from these.

我可以從當中截取資訊

Of course, the more you go in the past, the less information you will have.

當然，當時間越往前 你擁有的資訊就越少

So, it might not be enough.

所以，這可能不夠

So, I can do what historians do.

因此，我可以做史學家們做的事

I can extrapolate.

我可以推論

This is what we call, in computer science, simulation.

這就是我們在電腦科學中所稱的模擬

If I take a log book,

如果我有一本航海日誌

I can consider, it's not just a log book

我可以不只把它看成是一位威尼斯船長 特定的航程

of a Venetian captain going to a particular journey.

I can consider it is actually a log book

我可以把它看成是當時許多航程的代表

which is representative of many journeys of that period.

I'm extrapolating.

我就是在推論

If I have a painting of a facade,

如果我有一幅建築外觀的繪畫

I can consider it's not just that particular building,

我可以不只把它看作特定的一棟建築

but probably it also shares the same grammar

這棟建築可能和其它遺失了資訊的 建築有著相同的風格

of buildings where we lost any information.

So if we want to construct a time machine,

如果我們想要打造一台時光機

we need two things.

我們需要兩樣東西

We need very large archives,

大量的檔案

and we need excellent specialists.

也需要很厲害的專家

The Venice Time Machine,

威尼斯時光機 (The Venice Time Machine)

the project I'm going to talk to you about,

也就是我接下來要跟你們談論的計劃

is a joint project between the EPFL

這個計劃是由洛桑聯邦理工學院和 威尼斯大學所合作

and the University of Venice Ca'Foscari.

There's something very peculiar about Venice,

威尼斯有個特點

that its administration has been

它的管理非常非常的官僚

very, very bureaucratic.

They've been keeping track of everything,

他們持續紀錄了每一件事情

almost like Google today.

就像Google一樣

At the Archivio di Stato,

在政府典藏中心 (Archivio di Stato)

you have 80 kilometers of archives

有長達80公里的檔案

documenting every aspect

記錄了超過1000年來 威尼斯生活的每個面相

of the life of Venice over more than 1,000 years.

You have every boat that goes out,

每一艘出海的船

every boat that comes in.

每一艘返航的船

You have every change that was made in the city.

在這城市裡面的每一個變化

This is all there.

都記錄得一清二楚

We are setting up a 10-year digitization program

我們正建立一個10年的數位化計劃

which has the objective of transforming

要把這個龐大的檔案 轉換成一個巨大的數位系統

this immense archive

into a giant information system.

The type of objective we want to reach

我們想達成的目標是

is 450 books a day that can be digitized.

每天可以數位化450本書

Of course, when you digitize, that's not enough,

當然，只有數位化是不夠的

because these documents,

因為這些文件大部份都是用拉丁、 托斯卡納、威尼斯方言寫的

most of them are in Latin, in Tuscan,

in Venetian dialect,

so you need to transcribe them,

所以有時必須轉錄、翻譯它們 並編成索引

to translate them in some cases,

to index them,

and this is obviously not easy.

而這理所當然並不容易

In particular, traditional optical character recognition method

特別是當傳統的光學字符識別法 運用在手寫稿上

that can be used for printed manuscripts,

they do not work well on the handwritten document.

效果不若列印稿的時候

So the solution is actually to take inspiration

解決辦法是借助另外一個方法：語音識別

from another domain: speech recognition.

This is a domain of something that seems impossible,

這個方法看似不可能 但卻是可行的

which can actually be done,

simply by putting additional constraints.

只需要加入額外的條件

If you have a very good model

如果你有一個很好的語言模型 曾經被使用過

of a language which is used,

if you have a very good model of a document,

如果你有一個很好的文件模型 那它們就已經很完整了

how well they are structured.

And these are administrative documents.

而這些是官方文件 很多都非常完整

They are well structured in many cases.

If you divide this huge archive into smaller subsets

當你把這巨大的檔案分成許多小子集

where a smaller subset actually shares similar features,

而每個子集都有共同的特徵的話

then there's a chance of success.

成功的機會就很大

If we reach that stage, then there's something else:

如果我們達到這個階段 就能夠做其他事情：

we can extract from this document events.

汲取這個文件中的資訊

Actually probably 10 billion events

事實上這個檔案中大約可以汲取出100億個事件

can be extracted from this archive.

And this giant information system

這個巨型資訊系統可以用多種方法搜尋

can be searched in many ways.

You can ask questions like,

例如，你可以問這樣的問題

"Who lived in this palazzo in 1323?"

「誰在1323年住在這座皇宮裡面？」

"How much cost a sea bream at the Realto market

「1434年時Realto市場裡一條鯛魚多少錢？」

in 1434?"

"What was the salary

「在十年中Murano玻璃工匠的薪水多高？」

of a glass maker in Murano

maybe over a decade?"

You can ask even bigger questions

你也可以問更大的問題

because it will be semantically coded.

因為它是用語意編碼的

And then what you can do is put that in space,

然後你就可以把它放在空間中

because much of this information is spatial.

因為大部份的資訊都是空間性的

And from that, you can do things like

接下來就可以重建這座城市非凡的旅程

reconstructing this extraordinary journey

of that city that managed to have a sustainable development

這座城市設法永續發展超過1000年

over a thousand years,

managing to have all the time

設法讓時間和環境取得平衡

a form of equilibrium with its environment.

You can reconstruct that journey,

你可以重建這趟旅程

visualize it in many different ways.

用不同方法看到它

But of course, you cannot understand Venice if you just look at the city.

但是當然，如果只是用看的 你無法了解威尼斯

You have to put it in a larger European context.

你要把它放到一個更廣大、 歐洲的概念裡面

So the idea is also to document all the things

所以這個想法同時也要把檔案 放在歐洲的層級中

that worked at the European level.

We can reconstruct also the journey

我們也可以重建威尼斯這個海上帝國的旅程

of the Venetian maritime empire,

how it progressively controlled the Adriatic Sea,

了解它是如何積極地控制亞德里亞海的

how it became the most powerful medieval empire

如何成為中世紀控制了大部分 東方以及南方航線的最強大的帝國

of its time,

controlling most of the sea routes

from the east to the south.

But you can even do other things,

你甚至可以做到其它事

because in these maritime routes,

因為這些航線有相似之處

there are regular patterns.

You can go one step beyond

可以更進一步建立模擬系統

and actually create a simulation system,

建立地中海模擬器

create a Mediterranean simulator

which is capable actually of reconstructing

它可以重建我們遺失的資訊

even the information we are missing,

which would enable us to have questions you could ask

讓你在用路線規劃時的 問題可以得到解答

like if you were using a route planner.

"If I am in Corfu in June 1323

「如果我在1323年的科孚，想去康士坦丁堡 我要去哪裡搭船？」

and want to go to Constantinople,

where can I take a boat?"

Probably we can answer this question

我們或許可以在一天，兩天或三天的 準確度之內回答這個問題

with one or two or three days' precision.

"How much will it cost?"

「要花多少錢？」

"What are the chance of encountering pirates?"

「有多大的機率遇到海盜？」

Of course, you understand,

當然，你懂的，這種計劃最大的挑戰 在於修飾、量化

the central scientific challenge of a project like this one

is qualifying, quantifying and representing

並在每一步驟重現不確定以及不一致的資料

uncertainty and inconsistency at each step of this process.

There are errors everywhere,

這份文件裡到處都有錯誤

errors in the document, it's the wrong name of the captain,

文件裡的錯誤包括把 船長的名字搞錯了，而有些船根本沒下過水

some of the boats never actually took to sea.

There are errors in translation, interpretative biases,

也有翻譯的錯誤，翻譯時的誤解

and on top of that, if you add algorithmic processes,

此外，在用演算法的時候會有辨認上的錯誤

you're going to have errors in recognition,

汲取上的錯誤

errors in extraction,

so you have very, very uncertain data.

所以你會得到相當相當不確定的數據

So how can we detect and correct these inconsistencies?

那我們要如何找到並修正這些不一致之處呢？

How can we represent that form of uncertainty?

我們如何重現不確定性？

It's difficult. One thing you can do

很困難。你可以做的一件事是把每一個步驟建檔

is document each step of the process,

not only coding the historical information

不只是把歷史訊息編碼

but what we call the meta-historical information,

也把所謂的元歷史(meta-history)訊息

how is historical knowledge constructed,

也就是建立歷史知識的根基的每一步都建檔

documenting each step.

That will not guarantee that we actually converge

這沒辦法保證我們確實涵蓋到 威尼斯的每一個故事

toward a single story of Venice,

but probably we can actually reconstruct

但我們確實根據完整文件重建了威尼斯 每一個可能發生過的故事

a fully documented potential story of Venice.

Maybe there's not a single map.

也許地圖不止一幅

Maybe there are several maps.

可能有好幾幅地圖

The system should allow for that,

系統應該要可以到這樣子

because we have to deal with a new form of uncertainty,

因為我們得面對一個新型態的不確定性

which is really new for this type of giant databases.

包含這樣龐大且新穎的資料庫

And how should we communicate

那我們要如何把這個新研究告訴大眾呢？

this new research to a large audience?

Again, Venice is extraordinary for that.

威尼斯是個絕佳的地方

With the millions of visitors that come every year,

由於每年有好幾百萬的遊客

it's actually one of the best places

這裡是試著創造未來博物館最好的地點之一

to try to invent the museum of the future.

Imagine, horizontally you see the reconstructed map

想想，平躺的是建立出的特定年份的地圖

of a given year,

and vertically, you see the document

立著的是用來重建這幅地圖的文件 例如，圖畫

that served the reconstruction,

paintings, for instance.

Imagine an immersive system that permits

想像一個讓你彷彿置身於特定年份的 威尼斯的沈浸式系統

to go and dive and reconstruct the Venice of a given year,

some experience you could share within a group.

一種讓你能和別人分享的經驗

On the contrary, imagine actually that you start

相反的，想像你能夠從一份 威尼斯手稿開始

from a document, a Venetian manuscript,

and you show, actually, what you can construct out of it,

看看你到底能夠從當中建立什麼

how it is decoded,

它是如何被解碼、當中所蘊含的概念是如何重建

how the context of that document can be recreated.

This is an image from an exhibit

這張圖片是現在在日內瓦的展覽當中展出

which is currently conducted in Geneva

with that type of system.

它就是用這類系統所建立的

So to conclude, we can say that

總而言之

research in the humanities is about to undergo

人文研究正走向一場

an evolution which is maybe similar

像是生命科學30年前所經歷的革命

to what happened to life sciences 30 years ago.

It's really a question of scale.

事實上，問題在於它的規模

We see projects which are

我們看到

這些計劃遠超出任何一個 小型研究團隊所能夠做的

much beyond any single research team can do,

and this is really new for the humanities,

這在人文領域也很新穎

which very often take the habit of working

人文領域習慣小團隊或是只和數個研究員合作

in small groups or only with a couple of researchers.

When you visit the Archivio di Stato,

當你造訪政府典藏中心時

you feel this is beyond what any single team can do,

你會感覺到這超過任何一個團隊做得到的

and that should be a joint and common effort.

這必須依賴團隊合作

So what we must do for this paradigm shift

在這個典範轉移中

is actually foster a new generation

我們必須培養一些新一代

of "digital humanists"

準備好迎接這個進步的 「數位」人文學家

that are going to be ready for this shift.

I thank you very much.

謝謝

(Applause)

（掌聲）