TEDGlobal 2010
David McCandless: The beauty of data visualization
David McCandless：資料視覺化的美麗
Filmed:
Readability: 4.3
3,250,094 views
David McCandless將全世界的軍隊開銷、媒體活動、Facebook的狀態更新等複雜的資料，轉化成簡單又美麗的圖表。他提倡人們應該使用設計工具，來整理當今過多的資訊，找出獨特的模式與關連性，也許就能改變我們對這個世界的看法。
David McCandless - Data journalist
David McCandless draws beautiful conclusions from complex datasets -- thus revealing unexpected insights into our world. Full bio
David McCandless draws beautiful conclusions from complex datasets -- thus revealing unexpected insights into our world. Full bio
Double-click the English transcript below to play the video.
00:15
It feels like we're all suffering
0
0
2000
目前看起來，
00:17
from information overload or data glut.
1
2000
3000
我們正在承受資訊的過度氾濫。
00:20
And the good news is there might be an easy solution to that,
2
5000
2000
好消息是，現在有一個簡單的方式，
00:22
and that's using our eyes more.
3
7000
2000
讓我們能有效的理解這些資訊。
00:24
So, visualizing information, so that we can see
4
9000
2000
在視覺化資訊中，
00:26
the patterns and connections that matter
5
11000
3000
我們能瞭解資料的模式與關連性，
00:29
and then designing that information so it makes more sense,
6
14000
3000
資訊經過設計，會讓資訊更有意義，
00:32
or it tells a story,
7
17000
2000
甚至陳述某個故事，
00:34
or allows us to focus only on the information that's important.
8
19000
3000
或是引領我們專注在資料的重點上。
00:38
Failing that, visualized information can just look really cool.
9
23000
3000
視覺化資訊可不僅僅是看起來很酷炫的東西。
00:41
So, let's see.
10
26000
2000
讓我們一起來看看。
00:45
This is the $Billion Dollar o-Gram,
11
30000
2000
這是"十億美元圖表(Billion Dollar Gram)"，
00:47
and this image arose
12
32000
2000
這張圖是在與媒體界接觸了
00:49
out of frustration I had
13
34000
2000
各種以十億美元為單位的事件後，
00:51
with the reporting of billion-dollar amounts in the press.
14
36000
2000
讓我感到沮喪的情形下，所繪製出來的。
00:53
That is, they're meaningless without context:
15
38000
3000
然而，沒有內容，數字就沒有意義。
00:56
500 billion for this pipeline,
16
41000
2000
輸油管花了5千億美金。
00:58
20 billion for this war.
17
43000
2000
戰爭花了2百億美金。
01:00
It doesn't make any sense, so the only way to understand it
18
45000
2000
為了體會這些金額大小的意義，我們只能透過
01:02
is visually and relatively.
19
47000
2000
視覺化還有相對性。
01:04
So I scraped a load of reported figures
20
49000
2000
所以，我從各種不同的來源
01:06
from various news outlets
21
51000
2000
蒐集了許多相關報導的數字，
01:08
and then scaled the boxes according to those amounts.
22
53000
3000
然後根據數字設定格子大小。
01:11
And the colors here represent the motivation behind the money.
23
56000
3000
而顏色代表的是使用這些金錢的動機。
01:14
So purple is "fighting,"
24
59000
3000
像紫色是戰爭，
01:17
and red is "giving money away," and green is "profiteering."
25
62000
3000
紅色是贈送，綠色是收益。
01:20
And what you can see straight away
26
65000
2000
這樣就讓各位能直接了解
01:22
is you start to have a different relationship to the numbers.
27
67000
2000
這些數字之間的差異。
01:24
You can literally see them.
28
69000
2000
讓各位能更快的了解。
01:26
But more importantly, you start to see
29
71000
2000
但更重要的是，
01:28
patterns and connections between numbers
30
73000
2000
這能讓各位發現，在過去各種報導中，
01:30
that would otherwise be scattered across multiple news reports.
31
75000
3000
這些數字間未被提起的模式與關連性。
01:33
Let me point out some that I really like.
32
78000
3000
讓我來告訴各位其中隱藏的事實。
01:36
This is OPEC's revenue, this green box here --
33
81000
2000
這塊綠色是OPEC(石油輸出國組織)的總收益，
01:38
780 billion a year.
34
83000
2000
一年7800億美金。
01:40
And this little pixel in the corner -- three billion --
35
85000
3000
右下角這一小塊，30億美金，
01:43
that's their climate change fund.
36
88000
3000
是該組織投資的氣候變遷基金。
01:46
Americans, incredibly generous people --
37
91000
2000
而極度慷慨的美國人，
01:48
over 300 billion a year, donated to charity every year,
38
93000
3000
每年做慈善的金額都超過3000億美金，
01:51
compared with the amount of foreign aid
39
96000
2000
相較於其他前17大工業國，
01:53
given by the top 17 industrialized nations
40
98000
2000
它們每年所捐增的總額
01:55
at 120 billion.
41
100000
2000
也不過1200億美金。
01:57
Then of course,
42
102000
2000
當然，
01:59
the Iraq War, predicted to cost just 60 billion
43
104000
2000
伊拉克戰爭，在2003年時，
02:01
back in 2003.
44
106000
3000
預計只需要花600億美金。
02:04
And it mushroomed slightly. Afghanistan and Iraq mushroomed now
45
109000
3000
這數字後來爆增。阿富汗/伊拉克戰爭目前開銷
02:07
to 3,000 billion.
46
112000
3000
已經來到3兆美金。
02:10
So now it's great
47
115000
2000
非常龐大的數字，
02:12
because now we have this texture, and we can add numbers to it as well.
48
117000
2000
因為我們有前車之鑑，這個數字也會再往上調整。
02:14
So we could say, well, a new figure comes out ... let's see African debt.
49
119000
3000
讓我們看另一個新的數字... 非洲的負債。
02:17
How much of this diagram do you think might be taken up
50
122000
2000
各位猜猜非洲各國的負債中
02:19
by the debt that Africa owes to the West?
51
124000
2000
西方國家持有多少?
02:21
Let's take a look.
52
126000
2000
讓我們看看。
02:23
So there it is:
53
128000
2000
來了，
02:25
227 billion is what Africa owes.
54
130000
2000
目前非洲各國的負債金額是2270億美金。
02:27
And the recent financial crisis,
55
132000
2000
而最近的金融風暴....
02:29
how much of this diagram might that figure take up?
56
134000
2000
各位認為這個數字的色塊是怎樣?
02:31
What has that cost the world? Let's take a look at that.
57
136000
3000
這個事件造成的成本是多少?讓我們來瞧瞧。
02:34
Dooosh -- Which I think is the appropriate sound effect
58
139000
3000
夭壽。這大概是最適合的形容詞了。
02:37
for that much money:
59
142000
2000
這麼大一筆的數字。
02:39
11,900 billion.
60
144000
4000
11兆9000億美金。
02:45
So, by visualizing this information,
61
150000
2000
所以，資訊透過視覺化，
02:47
we turned it into a landscape
62
152000
2000
資訊會呈現為圖像，
02:49
that you can explore with your eyes,
63
154000
2000
你就能用眼睛尋找蛛絲馬跡，
02:51
a kind of map really, a sort of information map.
64
156000
2000
就像是某種資訊地圖。
02:53
And when you're lost in information,
65
158000
2000
當你被大量資訊迷惑時，
02:55
an information map is kind of useful.
66
160000
3000
這種資訊地圖就能派上用場。
02:58
So I want to show you another landscape now.
67
163000
2000
現在，我要給各位看看另一種圖像。
03:00
We need to imagine what a landscape
68
165000
2000
各位想像一下，
03:02
of the world's fears might look like.
69
167000
3000
把世界上令人驚恐的事件，用圖形表示會是怎樣。
03:05
Let's take a look.
70
170000
2000
讓我們來看看。
03:07
This is Mountains Out of Molehills,
71
172000
2000
這圖叫"小題大作(mountains out of mole hills)"，
03:09
a timeline of global media panic.
72
174000
2000
這張圖是全球媒體造成恐慌的時間軸。
03:11
(Laughter)
73
176000
2000
(笑)
03:13
So, I'll label this for you in a second.
74
178000
2000
讓我很快的講解一下圖表。
03:15
But the height here, I want to point out,
75
180000
2000
色塊的高低起伏，
03:17
is the intensity of certain fears
76
182000
2000
是指某特定事件
03:19
as reported in the media.
77
184000
2000
在某時段被媒體報導的強度。
03:21
Let me point them out.
78
186000
2000
讓我為各位說明顏色的意義。
03:23
So this, swine flu -- pink.
79
188000
4000
粉紅色的是豬流感。
03:27
Bird flu.
80
192000
2000
這是禽流感。
03:29
SARS -- brownish here. Remember that one?
81
194000
3000
SARS，是這個淡褐色的，
03:32
The millennium bug,
82
197000
3000
這是千禧蟲事件，
03:35
terrible disaster.
83
200000
2000
可怕的電腦病毒。
03:37
These little green peaks
84
202000
2000
這些小小的綠色部分，
03:39
are asteroid collisions.
85
204000
2000
是小行星撞擊地球的消息。
03:41
(Laughter)
86
206000
2000
(笑)
03:43
And in summer, here, killer wasps.
87
208000
2000
這是今年夏天的殺人峰事件。
03:45
(Laughter)
88
210000
8000
(笑)
03:53
So these are what our fears look like
89
218000
2000
這些是從媒體開始報導到結束之間
03:55
over time in our media.
90
220000
2000
對這些事件報導的程度。
03:57
But what I love -- and I'm a journalist --
91
222000
2000
因為我是個記者，
03:59
and what I love is finding hidden patterns; I love being a data detective.
92
224000
3000
所以我喜歡尋找那隱藏的模式，我喜歡當個資料偵探。
04:02
And there's a very interesting and odd pattern hidden in this data
93
227000
3000
資料中總是隱藏著有趣且古怪的模式，
04:05
that you can only see when you visualize it.
94
230000
2000
除了把資料視覺化，不然根本沒辦法發覺。
04:07
Let me highlight it for you.
95
232000
2000
讓我為各位點出這些地方。
04:09
See this line, this is a landscape for violent video games.
96
234000
3000
紅色部分，是針對暴力電玩的報導強度。
04:12
As you can see, there's a kind of odd, regular pattern in the data,
97
237000
3000
各位可以看到，有點奇怪，資料出現了一些規律，
04:15
twin peaks every year.
98
240000
2000
每年的報導都出現2次尖峰。
04:17
If we look closer, we see those peaks occur
99
242000
2000
若我們仔細看，我們能發現這2次尖峰，
04:19
at the same month every year.
100
244000
3000
都是發生在特定的月份。
04:22
Why?
101
247000
2000
為什麼?
04:24
Well, November, Christmas video games come out,
102
249000
2000
因為耶誕節前，電玩遊戲會紛紛在11月推出，
04:26
and there may well be an upsurge in the concern about their content.
103
251000
3000
所以很多媒體就會針對這些遊戲內容做出評論。
04:29
But April isn't a particularly massive month
104
254000
3000
但四月對電玩業者而言，
04:32
for video games.
105
257000
2000
又不是什麼重要的月份。
04:34
Why April?
106
259000
2000
為什麼四月也會這樣?
04:36
Well, in April 1999 was the Columbine shooting,
107
261000
3000
因為1999年4月發生了科倫拜校園槍擊事件，
04:39
and since then, that fear
108
264000
2000
從那時候開始，
04:41
has been remembered by the media
109
266000
2000
這件事情就被媒體銘記，
04:43
and echoes through the group mind gradually through the year.
110
268000
2000
並在每年的這個時候重新報導。
04:45
You have retrospectives, anniversaries,
111
270000
3000
像是回顧展、周年紀念日、
04:48
court cases, even copy-cat shootings,
112
273000
3000
法庭案件、甚至有模仿的槍擊事件，
04:51
all pushing that fear into the agenda.
113
276000
3000
這些助力讓此事件一再的被報導。
04:54
And there's another pattern here as well. Can you spot it?
114
279000
2000
這裡面其實還有一個模式，各位注意到了嗎?
04:56
See that gap there? There's a gap,
115
281000
2000
看到這個缺口了嗎?
04:58
and it affects all the other stories.
116
283000
2000
這個大缺口是其他事件所導致的。
05:00
Why is there a gap there?
117
285000
2000
是什麼事件?
05:02
You see where it starts? September 2001,
118
287000
3000
這缺口從什麼時候開始的? 2001年9月。
05:05
when we had something very real
119
290000
2000
因為這個時間點
05:07
to be scared about.
120
292000
2000
我們有個最令人驚悚的(911)事件。
05:09
So, I've been working as a data journalist for about a year,
121
294000
3000
我作為資料記者大約已經一年的時間。
05:12
and I keep hearing a phrase
122
297000
2000
這期間我常聽到一句話
05:14
all the time, which is this:
123
299000
3000
這句話就是：
05:17
"Data is the new oil."
124
302000
2000
"資料是種新石油"
05:19
Data is the kind of ubiquitous resource
125
304000
3000
資料就像是某種普遍的資源，
05:22
that we can shape to provide new innovations and new insights,
126
307000
3000
我們可以使之塑形，以提供我們新思想跟新洞察，
05:25
and it's all around us, and it can be mined very easily.
127
310000
3000
這種資源就在我們身邊，非常容易取得。
05:28
It's not a particularly great metaphor in these times,
128
313000
3000
現在這時間點，是不太適合把資料比喻成石油，
05:31
especially if you live around the Gulf of Mexico,
129
316000
3000
尤其是如果你住在墨西哥灣附近的話，
05:34
but I would, perhaps, adapt this metaphor slightly,
130
319000
2000
但其實我只對這種比方贊同一點點，
05:36
and I would say that data is the new soil.
131
321000
3000
我認為，"資料是一種新土壤"。
05:40
Because for me, it feels like a fertile, creative medium.
132
325000
3000
對我來說，資料就像是一種肥沃、有創造力的媒介。
05:43
Over the years, online,
133
328000
2000
過去幾年，
05:45
we've laid down
134
330000
3000
我們已經在網路上
05:48
a huge amount of information and data,
135
333000
2000
放置了非常大量的資訊和資料，
05:50
and we irrigate it with networks and connectivity,
136
335000
2000
我們利用網路和連結灌溉它們，
05:52
and it's been worked and tilled by unpaid workers and governments.
137
337000
3000
透過政府和網路志工在這上面不斷耕種。
05:55
And, all right, I'm kind of milking the metaphor a little bit.
138
340000
3000
喔，對啦，我也可以用擠奶作為譬喻。
05:58
But it's a really fertile medium,
139
343000
3000
這是非常豐沃的媒介，
06:01
and it feels like visualizations, infographics, data visualizations,
140
346000
3000
形象、圖表資料、視覺化資料，
06:04
they feel like flowers blooming from this medium.
141
349000
3000
都像是從這媒介中，生長出來的茂盛花海。
06:07
But if you look at it directly,
142
352000
2000
不過，若直接觀看資料圖表，
06:09
it's just a lot of numbers and disconnected facts.
143
354000
2000
看起來就像一堆數字，和一堆不相干的事件。
06:11
But if you start working with it and playing with it in a certain way,
144
356000
3000
如果將這些資訊用特別的方式整理一下，
06:14
interesting things can appear and different patterns can be revealed.
145
359000
3000
有趣的事情就會浮現，各式各樣的模式就會顯露出來了。
06:17
Let me show you this.
146
362000
2000
讓我示範給各位看。
06:19
Can you guess what this data set is?
147
364000
3000
各位能猜到這些資料在陳述什麼嗎?
06:22
What rises twice a year,
148
367000
2000
1年內會出現2次尖峰，
06:24
once in Easter
149
369000
2000
一次是在復活節(3、4月)，
06:26
and then two weeks before Christmas,
150
371000
2000
另一次是聖誕節的前2周，
06:28
has a mini peak every Monday,
151
373000
2000
而每星期的星期一都會有一次小高峰，
06:30
and then flattens out over the summer?
152
375000
2000
然後在夏天的時候特別平穩。
06:32
I'll take answers.
153
377000
2000
有人要猜猜看嗎?
06:34
(Audience: Chocolate.) David McCandless: Chocolate.
154
379000
2000
觀眾：巧克力。
06:36
You might want to get some chocolate in.
155
381000
3000
這跟巧克力有一點點相關。
06:39
Any other guesses?
156
384000
2000
有其他答案嗎?
06:41
(Audience: Shopping.) DM: Shopping.
157
386000
2000
觀眾：購物。
06:43
Yeah, retail therapy might help.
158
388000
3000
購物也許能紓緩這件事。
06:46
(Audience: Sick leave.)
159
391000
2000
觀眾：請病假。
06:48
DM: Sick leave. Yeah, you'll definitely want to take some time off.
160
393000
2000
請病假。對，這時候會想要休息一下。
06:50
Shall we see?
161
395000
2000
答案揭曉摟。
06:53
(Laughter)
162
398000
8000
(分手潮---資料來源：Facebook的狀態更新)
07:01
(Applause)
163
406000
3000
(掌聲)
07:04
So, the information guru Lee Byron and myself,
164
409000
3000
Lee Byron和我一起做這項統計，
07:07
we scraped 10,000 status Facebook updates
165
412000
3000
我們抓了1萬筆Facebook上的個人狀態，
07:10
for the phrase "break-up" and "broken-up"
166
415000
2000
關鍵字是"分手"，
07:12
and this is the pattern we found --
167
417000
2000
然後我們發現了這個模式，
07:14
people clearing out for Spring Break,
168
419000
2000
春假前先分手(才能玩樂)
07:16
(Laughter)
169
421000
5000
(笑)
07:21
coming out of very bad weekends on a Monday,
170
426000
2000
在度過了幾個糟糕週末後的星期一，
07:23
being single over the summer,
171
428000
3000
然後單身渡過整個夏天。
07:26
and then the lowest day of the year, of course: Christmas Day.
172
431000
3000
接下來是一整年分手數的最低點，聖誕節。
07:29
Who would do that?
173
434000
3000
誰會在這時候分手阿?
07:32
So there's a titanic amount of data out there now,
174
437000
2000
現在我們有許多的數據，
07:34
unprecedented.
175
439000
2000
是前所未有的。
07:37
But if you ask the right kind of question,
176
442000
2000
不過，若你的質疑正確，
07:39
or you work it in the right kind of way,
177
444000
2000
或是你蒐集正確的資料，
07:41
interesting things can emerge.
178
446000
3000
有趣的事情就會跑出來。
07:44
So information is beautiful. Data is beautiful.
179
449000
3000
資訊是美麗的，資料是美麗的。
07:47
I wonder if I could make my life beautiful.
180
452000
3000
我希望將我的人生弄的很美麗。
07:50
And here's my visual C.V.
181
455000
2000
這是我的"視覺履歷"(visual C.V)。
07:52
I'm not quite sure I've succeeded.
182
457000
2000
我不太確定是否成功了。
07:54
Pretty blocky, the colors aren't that great.
183
459000
2000
上色的小方塊。顏色不是這麼好看。
07:56
But I wanted to convey something to you.
184
461000
3000
但我想傳達一些訊息給各位。
07:59
I started as a programmer,
185
464000
2000
我第一份工作是程式設計師，
08:01
and then I worked as a writer for many years, about 20 years,
186
466000
2000
然後我當作家大約20年，
08:03
in print, online and then in advertising,
187
468000
2000
作品出版後，我又進了廣告業，
08:05
and only recently have I started designing.
188
470000
3000
直到最近我才開始從事設計。
08:08
And I've never been to design school.
189
473000
2000
我從未接受設計方面的教育。
08:10
I've never studied art or anything.
190
475000
3000
也從來沒學過任何美術學科。
08:13
I just kind of learned through doing.
191
478000
2000
我透過實作來學習。
08:15
And when I started designing,
192
480000
2000
當我開始從事設計的時候，
08:17
I discovered an odd thing about myself.
193
482000
2000
我就發現自己有些怪怪的。
08:19
I already knew how to design,
194
484000
2000
雖然我已經知道該如何設計，
08:21
but it wasn't like I was amazingly brilliant at it,
195
486000
3000
但是結果並不如我想像中的令人驚奇，
08:24
but more like I was sensitive
196
489000
2000
而我對某些元素相當敏感，
08:26
to the ideas of grids and space
197
491000
2000
像是方格、空間、
08:28
and alignment and typography.
198
493000
2000
直線對齊、排版設計。
08:30
It's almost like being exposed
199
495000
2000
可能是因為
08:32
to all this media over the years
200
497000
2000
我過去這幾年都在媒體界工作
08:34
had instilled a kind of dormant design literacy in me.
201
499000
3000
而那類型的設計素養就被深植在腦中。
08:37
And I don't feel like I'm unique.
202
502000
2000
我不覺得我是獨一無二的。
08:39
I feel that everyday, all of us now
203
504000
2000
我只是感覺到，人們每天
08:41
are being blasted by information design.
204
506000
3000
都被資訊設計(information design)所轟炸。
08:44
It's being poured into our eyes through the Web,
205
509000
2000
透過網頁不斷注入到我們的眼睛裡，
08:46
and we're all visualizers now;
206
511000
2000
我們都善於在腦海中想像，
08:48
we're all demanding a visual aspect
207
513000
2000
因此我們需要將資訊
08:50
to our information.
208
515000
3000
呈現為視覺化的面貌。
08:53
There's something almost quite magical about visual information.
209
518000
3000
視覺資訊有一些地方是很不可思議的。
08:56
It's effortless, it literally pours in.
210
521000
3000
它很簡單，確實地融入我們。
08:59
And if you're navigating a dense information jungle,
211
524000
3000
若你想引領人們在濃密的資訊叢林中航行，
09:02
coming across a beautiful graphic
212
527000
2000
透過華麗的圖像
09:04
or a lovely data visualization,
213
529000
2000
或是美麗的視覺化資訊，
09:06
it's a relief, it's like coming across a clearing in the jungle.
214
531000
3000
那就輕鬆多了，這就變成像是通過叢林中的空地。
09:09
I was curious about this, so it led me
215
534000
2000
我對這方面的知識相當好奇，
09:11
to the work of a Danish physicist
216
536000
2000
因此我與一位丹麥物理學家合作
09:13
called Tor Norretranders,
217
538000
2000
他叫Tor Norretranders，
09:15
and he converted the bandwidth of the senses into computer terms.
218
540000
3000
他將人的感官寬帶轉換成電腦用語。
09:19
So here we go. This is your senses,
219
544000
2000
開始了，這是你的感官，
09:21
pouring into your senses every second.
220
546000
2000
你的感官時時刻刻都在接受資訊。
09:23
Your sense of sight is the fastest.
221
548000
3000
視覺對資訊的反應是最快的。
09:26
It has the same bandwidth as a computer network.
222
551000
3000
就像是電腦寬頻網路的速度。
09:29
Then you have touch, which is about the speed of a USB key.
223
554000
3000
然後是觸覺，大概像USB傳輸的速度。
09:32
And then you have hearing and smell,
224
557000
2000
接下來才是你的聽覺和嗅覺，
09:34
which has the throughput of a hard disk.
225
559000
2000
這就像硬碟的存取速度。
09:36
And then you have poor old taste,
226
561000
2000
然後是緩慢的味覺，
09:38
which is like barely the throughput of a pocket calculator.
227
563000
3000
它的反應速度大概就像計算機的存取速度。
09:41
And that little square in the corner, a naught .7 percent,
228
566000
3000
在角落有個小方塊，大概佔感官的0.7%，
09:44
that's the amount we're actually aware of.
229
569000
3000
這是我們真正的意識。
09:47
So a lot of your vision --
230
572000
2000
你透過眼睛接受的大量資訊，
09:49
the bulk of it is visual, and it's pouring in.
231
574000
2000
像圖中的大方塊般龐大，不停的灌入。
09:51
It's unconscious.
232
576000
2000
這過程是無意識的。
09:53
The eye is exquisitely sensitive
233
578000
3000
而且眼睛對於圖形的顏色、形狀、組合模式，
09:56
to patterns in variations in color, shape and pattern.
234
581000
3000
是非常敏感的。
09:59
It loves them, and it calls them beautiful.
235
584000
2000
眼睛喜愛這些圖形，認為它們很美麗。
10:01
It's the language of the eye.
236
586000
2000
這是眼睛的語言。
10:03
If you combine the language of the eye with the language of the mind,
237
588000
2000
若你把眼睛和心靈的語言組合起來，
10:05
which is about words and numbers and concepts,
238
590000
3000
這二者都針對文字、數字、概念的陳述，
10:08
you start speaking two languages simultaneously,
239
593000
3000
你便能同時述說二種語言，
10:11
each enhancing the other.
240
596000
3000
就能增加兩者的感受。
10:14
So, you have the eye, and then you drop in the concepts.
241
599000
3000
所以，用眼睛看某事物時，心裡就會出現概念。
10:17
And that whole thing -- it's two languages
242
602000
2000
而這整個過程，是因為
10:19
both working at the same time.
243
604000
2000
這二種語言同時運作的關係。
10:21
So we can use this new kind of language, if you like,
244
606000
2000
若你願意，我們能用另一種新語言，
10:23
to alter our perspective or change our views.
245
608000
3000
這種語言可以改變我們的認知和觀點。
10:26
Let me ask you a simple question
246
611000
2000
讓我問各位一個簡單的問題，
10:28
with a really simple answer:
247
613000
2000
只需要簡短的答案。
10:30
Who has the biggest military budget?
248
615000
2000
哪個國家的軍事預算最高?
10:32
It's got to be America, right?
249
617000
2000
就是美國了，對吧?
10:34
Massive. 609 billion in 2008 --
250
619000
2000
非常龐大的數字。2008年的6090億..
10:36
607, rather.
251
621000
2000
更正，是6070億。
10:38
So massive, in fact, that it can contain
252
623000
2000
實際上，這麼大的數字，
10:40
all the other military budgets in the world inside itself.
253
625000
3000
可以吃下全世界其他國家軍事預算的總和。
10:43
Gobble, gobble, gobble, gobble, gobble.
254
628000
2000
幾乎吃個精光。
10:45
Now, you can see Africa's total debt there
255
630000
2000
我們用非洲國家的總負債(中間方塊)
10:47
and the U.K. budget deficit for reference.
256
632000
2000
和英國預算赤字(右側方塊)做個比較。
10:49
So that might well chime
257
634000
2000
這應該也符合
10:51
with your view that America
258
636000
2000
各位對美國的看法，
10:53
is a sort of warmongering military machine,
259
638000
3000
像是好戰者、戰爭機器、
10:56
out to overpower the world
260
641000
2000
企圖征服世界、
10:58
with its huge industrial-military complex.
261
643000
3000
大號的軍事工業複合體(industrial-military complex)。
11:01
But is it true that America has the biggest military budget?
262
646000
3000
但是美國真的是擁有最高國防預算的國家嗎?
11:04
Because America is an incredibly rich country.
263
649000
2000
因為美國是個非常富有的國家。
11:06
In fact, it's so massively rich
264
651000
2000
事實上，美國的富有程度，
11:08
that it can contain the four other
265
653000
2000
能包含世界上
11:10
top industrialized nations' economies
266
655000
2000
前4大工業國家的經濟產值
11:12
inside itself, it's so vastly rich.
267
657000
3000
實在是非常富有。
11:15
So its military budget is bound to be enormous.
268
660000
3000
所以這樣的國家，軍事預算自然就龐大。
11:18
So, to be fair and to alter our perspective,
269
663000
2000
要從公平的角度，來調整我們的認知，
11:20
we have to bring in another data set,
270
665000
2000
我們得要有另一套資料來分析，
11:22
and that data set is GDP, or the country's earnings.
271
667000
2000
這套資料就是GDP，或說是一國的收入。
11:24
Who has the biggest budget as a proportion of GDP?
272
669000
2000
哪一國的軍事預算佔GDP最重?
11:26
Let's have a look.
273
671000
2000
讓我們來瞧瞧。
11:28
That changes the picture considerably.
274
673000
3000
圖形跟剛剛有很大的不同喔。
11:31
Other countries pop into view that you, perhaps, weren't considering,
275
676000
3000
很多國家都跑出來了，搞不好這些國家你想都沒想過，
11:34
and American drops into eighth.
276
679000
2000
美國的名次掉到第8了。
11:36
Now you can also do this with soldiers.
277
681000
2000
現在我們來看看有關軍人的部份。
11:38
Who has the most soldiers? It's got to be China.
278
683000
2000
誰擁有的軍人最多?應該是中國。
11:40
Of course, 2.1 million.
279
685000
2000
沒錯，210萬人。
11:42
Again, chiming with your view
280
687000
2000
又來了，再度符合各位的印象，
11:44
that China has a militarized regime
281
689000
2000
因為中國是個軍事主義政權，
11:46
ready to, you know, mobilize its enormous forces.
282
691000
2000
像是隨時要展現他們強大的武力。
11:48
But of course, China has an enormous population.
283
693000
3000
但事實上，中國擁有龐大的人口。
11:51
So if we do the same,
284
696000
2000
若我們做跟剛剛一樣的事，
11:53
we see a radically different picture.
285
698000
2000
我們就會看到非常不一樣的圖形。
11:55
China drops to 124th.
286
700000
2000
中國掉到了第124名。
11:57
It actually has a tiny army
287
702000
2000
從另一種資料面來看，
11:59
when you take other data into consideration.
288
704000
3000
中國的軍隊規模實在有夠小。
12:02
So, absolute figures, like the military budget,
289
707000
2000
像軍事預算這種絕對數字，
12:04
in a connected world,
290
709000
2000
在這互相連結的世界裡，
12:06
don't give you the whole picture.
291
711000
2000
反而沒辦法讓你看到完整的事實。
12:08
They're not as true as they could be.
292
713000
2000
事物的真相不如它表面所示。
12:10
We need relative figures that are connected to other data
293
715000
3000
我們需要用其他資料的相關數字作比較，
12:13
so that we can see a fuller picture,
294
718000
2000
這樣才能了解最完整的事實，
12:15
and then that can lead to us changing our perspective.
295
720000
2000
進一步改變我們的認知。
12:17
As Hans Rosling, the master,
296
722000
2000
Hans Rosling教授，
12:19
my master, said,
297
724000
3000
同時也是我的老師，他說：
12:22
"Let the dataset change your mindset."
298
727000
3000
"用數據改變思維"
12:26
And if it can do that, maybe it can also change your behavior.
299
731000
3000
若真能改變思維，那同樣的也能改變你的行為。
12:29
Take a look at this one.
300
734000
2000
各位看這裡。
12:31
I'm a bit of a health nut.
301
736000
2000
我是個注重健康的人。
12:33
I love taking supplements and being fit,
302
738000
3000
雖然我喜歡攝取一些營養品、健身，
12:36
but I can never understand what's going on in terms of evidence.
303
741000
3000
但是我永遠搞不清楚這些東西會帶來的好處。
12:39
There's always conflicting evidence.
304
744000
2000
有很多牴觸的說法。
12:41
Should I take vitamin C? Should I be taking wheatgrass?
305
746000
2000
我應該攝取維他命C嗎?還是應該吃小麥草?
12:43
This is a visualization of all the evidence
306
748000
2000
畫面上所呈現的圖像就是
12:45
for nutritional supplements.
307
750000
2000
營養品所提供好處的圖像化。
12:47
This kind of diagram is called a balloon race.
308
752000
3000
這種圖形稱為熱汽球(balloon race)。
12:50
So the higher up the image,
309
755000
2000
較高的圓圈，
12:52
the more evidence there is for each supplement.
310
757000
3000
是有強烈證據證明其功效的營養品。
12:55
And the bubbles correspond to popularity as regards to Google hits.
311
760000
3000
而圓圈大小是對應在Google上查詢的次數。
12:58
So you can immediately apprehend
312
763000
3000
這樣你就能馬上看出營養品之間，
13:01
the relationship between efficacy and popularity,
313
766000
3000
效果和受歡迎程度的比較。
13:04
but you can also, if you grade the evidence,
314
769000
3000
若再根據這些證據做排序，
13:07
do a "worth it" line.
315
772000
2000
可以畫上一條"值得攝取(worth it)"線。
13:09
So supplements above this line are worth investigating,
316
774000
3000
在這條線上的營養素是最值得探討的，
13:12
but only for the conditions listed below,
317
777000
3000
它們適用於圓圈內下方小字的情況。
13:15
and then the supplements below the line
318
780000
3000
而在這條線之下的營養素，
13:18
are perhaps not worth investigating.
319
783000
2000
其實，並不這麼值得探討。
13:20
Now this image constitutes a huge amount of work.
320
785000
3000
要弄出這張圖片可是件大工程。
13:23
We scraped like 1,000 studies from PubMed,
321
788000
3000
我們從PubMed搜尋引擎中抓出一千多份的研究報告，
13:26
the biomedical database,
322
791000
2000
PubMed會連到生物醫學資料庫，
13:28
and we compiled them and graded them all.
323
793000
3000
然後我們將這些報告做匯整排序。
13:31
And it was incredibly frustrating for me
324
796000
2000
這讓我感到非常沮喪的是，
13:33
because I had a book of 250 visualizations to do for my book,
325
798000
3000
我的書需要有250張的視覺化圖像，
13:36
and I spent a month doing this,
326
801000
2000
然後我花了一個月整理出這堆圈圈，
13:38
and I only filled two pages.
327
803000
2000
卻只能塞滿2頁。
13:40
But what it points to
328
805000
2000
不過，這張圖點出了
13:42
is that visualizing information like this
329
807000
2000
視覺化的資訊
13:44
is a form of knowledge compression.
330
809000
2000
就是一種知識濃縮的型態。
13:46
It's a way of squeezing an enormous amount
331
811000
2000
這種方式能夠壓縮非常大量的
13:48
of information and understanding
332
813000
2000
資訊和知識
13:50
into a small space.
333
815000
2000
濃縮在一起。
13:52
And once you've curated that data, and once you've cleaned that data,
334
817000
2000
透過資料的搜集整頓，
13:54
and once it's there,
335
819000
2000
一但完成，
13:56
you can do cool stuff like this.
336
821000
2000
你也能做出這麼酷炫的東西。
13:58
So I converted this into an interactive app,
337
823000
2000
我將這張圖轉換成一個互動的應用程式，
14:00
so I can now generate this application online --
338
825000
2000
便產生了這個線上應用程式，
14:02
this is the visualization online --
339
827000
2000
這種線上的視覺化程式，
14:04
and I can say, "Yeah, brilliant."
340
829000
2000
我想說："哇嗚，超棒的"。
14:06
So it spawns itself.
341
831000
2000
這張圖能自己更新資料。
14:08
And then I can say, "Well, just show me the stuff
342
833000
2000
只要我說："我只想看
14:10
that affects heart health."
343
835000
2000
會影響心臟健康的東西就好"。
14:12
So let's filter that out.
344
837000
2000
點選項，程式會開始過濾。
14:14
So heart is filtered out, so I can see if I'm curious about that.
345
839000
3000
有關心臟保養的營養品就出現了。
14:17
I think, "No, no. I don't want to take any synthetics,
346
842000
2000
如果我想說："喔不，我是不吃人工合成物的。
14:19
I just want to see plants and --
347
844000
3000
只要告訴我
14:22
just show me herbs and plants. I've got all the natural ingredients."
348
847000
3000
藥草類和植物類食品就行了，我只要天然的成分"。
14:25
Now this app is spawning itself
349
850000
2000
然後這程式又會開始從資料中
14:27
from the data.
350
852000
2000
過濾出想要的資訊。
14:29
The data is all stored in a Google Doc,
351
854000
2000
這些資料會被儲存成Google文件，
14:31
and it's literally generating itself from that data.
352
856000
3000
嚴格來說，資料會自動轉成文件檔案。
14:34
So the data is now alive; this is a living image,
353
859000
2000
因此，這些資料是活的，是個活生生的圖像，
14:36
and I can update it in a second.
354
861000
2000
我只需要幾秒鐘就能更新資料。
14:38
New evidence comes out. I just change a row on a spreadsheet.
355
863000
2000
我只需要改變選單裡的選項，新的資料就跑出來了。
14:40
Doosh! Again, the image recreates itself.
356
865000
4000
再一次，夭壽！這些圖像能自我創造。
14:44
So it's cool.
357
869000
2000
它非常棒。
14:46
It's kind of living.
358
871000
3000
它是活生生的。
14:49
But it can go beyond data,
359
874000
2000
它超越了資料本身，
14:51
and it can go beyond numbers.
360
876000
2000
它能突顯數字的涵意。
14:53
I like to apply information visualization
361
878000
2000
我想要應用資料視覺化的技術
14:55
to ideas and concepts.
362
880000
3000
在各種不同領域上。
14:58
This is a visualization
363
883000
2000
畫面上的是
15:00
of the political spectrum,
364
885000
2000
政治光譜(度量政治傾向的工具)的視覺化。
15:02
an attempt for me to try
365
887000
2000
我試圖了解
15:04
and understand how it works
366
889000
2000
這張圖該怎麼運作，
15:06
and how the ideas percolate down
367
891000
2000
還有政府的立場
15:08
from government into society and culture,
368
893000
2000
該如何滲透到社會、文化、
15:10
into families, into individuals, into their beliefs
369
895000
3000
家庭、個人、甚至到個人信念，
15:13
and back around again in a cycle.
370
898000
3000
然後這個影響關係會形成一個迴圈。
15:16
What I love about this image
371
901000
2000
我喜歡這張圖的地方是
15:18
is it's made up of concepts,
372
903000
2000
它是由許多概念所構成，
15:20
it explores our worldviews
373
905000
2000
它探討了我們當今的世界觀，
15:22
and it helps us -- it helps me anyway --
374
907000
2000
它能幫助我們
15:24
to see what others think,
375
909000
2000
觀察到他人心裡所想，
15:26
to see where they're coming from.
376
911000
2000
並觀察到這些想法從何而來。
15:28
And it feels just incredibly cool to do that.
377
913000
3000
這張圖實在是太棒了。
15:31
What was most exciting for me
378
916000
3000
在設計這張圖時，
15:34
designing this
379
919000
2000
最令我感到興奮的是，
15:36
was that, when I was designing this image,
380
921000
2000
當我正在進行設計的時候，
15:38
I desperately wanted this side, the left side,
381
923000
3000
我拼老命想把左派陣營
15:41
to be better than the right side --
382
926000
2000
弄得比右派陣營好(註：左派提倡自由，右派提倡集權)
15:43
being a journalist, a Left-leaning person --
383
928000
3000
因為我是一位記者、一位左派人士。
15:46
but I couldn't, because I would have created
384
931000
2000
但我不能這樣做，否則會出現
15:48
a lopsided, biased diagram.
385
933000
3000
一個傾斜、具有偏見的圖像。
15:51
So, in order to really create a full image,
386
936000
3000
所以，為了忠實呈現完整的圖像，
15:54
I had to honor the perspectives on the right-hand side
387
939000
3000
我必須要有身為右派人士的榮耀感，
15:57
and at the same time, uncomfortably recognize
388
942000
3000
然而，這過程中，我發現到自己身上
16:00
how many of those qualities were actually in me,
389
945000
3000
也有許多對立者的特質
16:03
which was very, very annoying and uncomfortable.
390
948000
2000
這真的非常討厭和不舒服。
16:05
(Laughter)
391
950000
4000
(笑)
16:09
But not too uncomfortable,
392
954000
2000
其實也沒這麼誇張啦，
16:11
because there's something unthreatening
393
956000
3000
因為這張圖，對於觀察他人政治觀感
16:14
about seeing a political perspective,
394
959000
2000
並不會構成太大的威脅，
16:16
versus being told or forced to listen to one.
395
961000
3000
相對的，還能促使人去聆聽另一方的聲音。
16:19
You're capable of holding conflicting viewpoints
396
964000
3000
這是真的，當你能清楚看見對方的立場，
16:22
joyously when you can see them.
397
967000
2000
就會愉悅地看待衝突的觀念。
16:24
It's even fun to engage with them
398
969000
2000
甚至樂於與對方接觸，
16:26
because it's visual.
399
971000
2000
因為一切視覺化了。
16:28
So that's what's exciting to me,
400
973000
2000
這就是讓我感到興奮的地方，
16:30
seeing how data can change my perspective
401
975000
2000
看到這些美麗、迷人的資料
16:32
and change my mind midstream --
402
977000
2000
是如何改變我們的認知，
16:34
beautiful, lovely data.
403
979000
3000
改變我們根深蒂固的觀念。
16:38
So, just to wrap up,
404
983000
2000
來做個總結，
16:40
I wanted to say
405
985000
2000
我想說的是
16:42
that it feels to me that design is about solving problems
406
987000
2000
我的感覺是，設計能解決問題
16:44
and providing elegant solutions,
407
989000
3000
同時也能提供優雅的解決方案。
16:47
and information design is about
408
992000
2000
資訊設計
16:49
solving information problems.
409
994000
2000
能解決資訊的問題。
16:51
It feels like we have a lot of information problems
410
996000
2000
此時此刻，
16:53
in our society at the moment,
411
998000
2000
我們的社會上擁有許多資訊問題，
16:55
from the overload and the saturation
412
1000000
2000
資訊的過載與飽和，
16:57
to the breakdown of trust and reliability
413
1002000
2000
使資料的信任度和可靠性受到打擊，
16:59
and runaway skepticism and lack of transparency,
414
1004000
2000
引發排山倒海的質疑，透明度也大大降低，
17:01
or even just interestingness.
415
1006000
2000
或說失去資料的趣味性。
17:03
I mean, I find information just too interesting.
416
1008000
2000
其實資訊是非常有趣的。
17:05
It has a magnetic quality that draws me in.
417
1010000
3000
它就像具有磁性般的吸引了我。
17:09
So, visualizing information
418
1014000
2000
視覺化的資訊
17:11
can give us a very quick solution to those kinds of problems.
419
1016000
3000
能快速的針對各種問題提供解答。
17:14
Even when the information is terrible,
420
1019000
2000
即使資訊夾帶的是負面、糟糕的消息，
17:16
the visual can be quite beautiful.
421
1021000
3000
視覺化能讓它變的非常美麗。
17:19
Often we can get clarity
422
1024000
3000
同時能得到清晰的思維
17:22
or the answer to a simple question very quickly,
423
1027000
2000
同時快速的回答簡單的問題，
17:24
like this one,
424
1029000
2000
就像這個，
17:26
the recent Icelandic volcano.
425
1031000
3000
最近的冰島火山。
17:29
Which was emitting the most CO2?
426
1034000
2000
請問是何者排放了最多的二氧化碳?
17:31
Was it the planes or the volcano,
427
1036000
2000
是飛機，還是火山，
17:33
the grounded planes or the volcano?
428
1038000
2000
是那些停飛的飛機還是火山?(2010/4，冰島火山爆發，造成歐洲航線大亂)
17:35
So we can have a look.
429
1040000
2000
我們能看見真相為何。
17:37
We look at the data and we see:
430
1042000
2000
各位看看這資料顯示的情形，
17:39
Yep, the volcano emitted 150,000 tons;
431
1044000
2000
沒錯，火山釋放了15萬噸的二氧化碳；
17:41
the grounded planes would have emitted
432
1046000
2000
而這些飛機如果沒有停飛的話，
17:43
345,000 if they were in the sky.
433
1048000
3000
會釋放34萬5千噸的二氧化碳。
17:46
So essentially, we had our first carbon-neutral volcano.
434
1051000
3000
所以實際上，這是我們第一座碳中和火山。
17:49
(Laughter)
435
1054000
2000
(笑)
17:51
(Applause)
436
1056000
9000
(掌聲)
18:00
And that is beautiful. Thank you.
437
1065000
3000
這就是資料的美麗之處，感謝你們。
18:03
(Applause)
438
1068000
8000
(掌聲)
ABOUT THE SPEAKER
David McCandless - Data journalistDavid McCandless draws beautiful conclusions from complex datasets -- thus revealing unexpected insights into our world.
Why you should listen
David McCandless makes infographics -- simple, elegant ways to see information that might be too complex or too big, small, abstract or scattered to otherwise be grasped. In his new book, Information Is Beautiful (in the US, it's being called The Visual Miscellaneum), McCandless and his cadre of info designers take a spin through the world of visualized data, from hard stats on politics and climate to daffy but no less important trends in pop music.
McCandless' genius is not so much in finding jazzy new ways to show data -- the actual graphics aren't the real innovation here -- as in finding fresh ways to combine datasets to let them ping and prod each other. Reporting the number of drug deaths in the UK every year is interesting; but mapping that data onto the number of drug deaths reported by the UK press, broken down by drug, is utterly fascinating (more deaths by marijuana were reported than in fact occurred, by a factor of 484%). McCandless contributes a monthly big-think graphic to the Guardian's Data Blog, and makes viral graphics for his blog Information Is Beautiful.
David McCandless | Speaker | TED.com