WEBVTT
00:00:03.614 --> 00:00:06.280
the world we live in is a wash with data
00:00:06.388 --> 00:00:09.857
that comes pouring in from everywhere around us
00:00:10.349 --> 00:00:14.330
on is own this data is just noise and confusion
00:00:14.515 --> 00:00:16.393
to make sense of data
00:00:16.393 --> 00:00:18.250
to find the meaning in this
00:00:18.250 --> 00:00:22.650
we need a powerful brunch of science STATISTICS
00:00:22.650 --> 00:00:27.542
believe me thats nothing boring about
statistics , especially not today when we
00:00:27.542 --> 00:00:28.442
can make the data
00:00:28.442 --> 00:00:30.423
seen
00:00:30.423 --> 00:00:32.575
with statistics we can really make
sense of the world
00:00:32.575 --> 00:00:35.091
and there is more
00:00:35.091 --> 00:00:41.170
with statistics that data deluge as it
been called , is leading us to an ever
00:00:41.170 --> 00:00:42.730
greater understanding
00:00:42.730 --> 00:00:46.286
of life on Earth and the Universe beyond
00:00:46.348 --> 00:00:51.540
and thanks to the incredible power
of today's computers it may
00:00:51.540 --> 00:00:57.760
fundamentally transformed the process of
scientific discovery
00:00:57.760 --> 00:00:59.070
I kid you not
00:00:59.070 --> 00:01:06.070
statistics is now the sexiest subject
around
00:01:12.891 --> 00:01:15.483
Sztokholm, Szwecja
00:01:22.698 --> 00:01:25.860
did you know that there is a one million boats in
Sweden
00:01:25.860 --> 00:01:28.269
thats one boat per nine people
00:01:28.269 --> 00:01:31.614
is the highest number of boats per person in Europe
00:01:31.614 --> 00:01:34.053
Sweden also has .. the most McDonalds per person in Europe
00:01:34.274 --> 00:01:35.615
..... the most computers per person in Europe
00:01:36.558 --> 00:01:38.597
.. the most nuclear reactor per person in Europe
00:01:41.770 --> 00:01:45.870
being a statistician you dont like telling
your profession at dinner parties
00:01:45.870 --> 00:01:46.759
but really
00:01:46.759 --> 00:01:51.098
statistician shouldn't be shy becouse
everyone wants to understand whats going on
00:01:51.467 --> 00:01:53.030
and statystics
00:01:53.030 --> 00:01:58.846
gives us a perspective on the world we
live in that we cant get in any other way
00:01:59.092 --> 00:02:03.639
the average person spends 24 years of their life asleep
00:02:04.008 --> 00:02:06.691
statistics tells us whether the things we think
00:02:06.691 --> 00:02:09.168
and believe are actually true
00:02:09.264 --> 00:02:13.949
most men think theyre better drivers than women
00:02:14.179 --> 00:02:19.495
but they have almost double the fatal accidens per mile driven
00:02:19.618 --> 00:02:22.719
a statystics are far more useful
00:02:22.719 --> 00:02:25.529
than we usually like to admit
00:02:25.529 --> 00:02:29.499
in the last recession I was listen this
famous calling to a talk-radiostation
00:02:29.945 --> 00:02:31.099
the man complain
00:02:31.099 --> 00:02:35.836
in times like this when unemployment raised up to a thirteen percent and income
00:02:35.836 --> 00:02:39.003
is fallen by five percent
and suicide raised up climbing
00:02:39.234 --> 00:02:43.533
I get so angry that the government
is wasting money on things like
00:02:43.533 --> 00:02:45.296
connection of statystics
00:02:48.273 --> 00:02:50.680
Im not officially a statystician
00:02:50.680 --> 00:02:56.172
strictly speaking my field is global
health
00:02:58.061 --> 00:03:03.498
but i got really obsessed with stats
when I realized how much people in
00:03:03.498 --> 00:03:07.439
Sweden just don't know about the
rest of the world
00:03:07.701 --> 00:03:12.069
I started in our medical university Karolinska Instytut
an undergraduate
00:03:12.069 --> 00:03:13.979
course called global health
00:03:13.979 --> 00:03:17.072
these students coming to our section at
the highest grade you can get in the
00:03:17.072 --> 00:03:20.333
swedish college system so i thought
maybe they know everything i'm going to
00:03:20.333 --> 00:03:21.639
teach them about
00:03:21.639 --> 00:03:23.419
so I did the pretest
00:03:23.419 --> 00:03:25.677
when they came and one of the question
00:03:25.677 --> 00:03:29.419
from which I learned a lot was this one: which country has the highest child
00:03:29.419 --> 00:03:32.749
mortality of this five pairs
00:03:32.749 --> 00:03:36.759
I wont put you at the test here
but is Turkey which is highest, there Poland
00:03:36.759 --> 00:03:39.420
Russia,Pakistan and South Africa
00:03:40.235 --> 00:03:41.980
and these were the results of the
swedish students
00:03:41.980 --> 00:03:44.879
one point eight right answer
of the five possible
00:03:44.879 --> 00:03:48.250
that means that there was a place for a
professor of international health
00:03:48.250 --> 00:03:50.349
and for my course
00:03:50.349 --> 00:03:55.339
but one late night when I was compiling
the report I really realised
00:03:55.339 --> 00:04:01.199
my discovery , I have shown that swedish top
students know statisticaly
00:04:01.199 --> 00:04:06.469
significantly less about the world than
the chimpanzees
00:04:06.469 --> 00:04:09.069
because the chimpanzee would score
00:04:09.069 --> 00:04:12.480
half right, if I gave them two bananas
which Sri Lanka or Turkey
00:04:12.480 --> 00:04:14.290
would be right half of the cases
00:04:14.290 --> 00:04:18.100
but the students are not that. I did also
an unethical study of the
00:04:18.100 --> 00:04:20.630
professors of the Karolinska Instytut
00:04:20.630 --> 00:04:24.270
that hands of Nobel Prize in
medicine and they are on par with the
00:04:24.270 --> 00:04:27.029
chimpanzee there
00:04:28.614 --> 00:04:32.890
today there's more information
accessible than ever before
00:04:32.890 --> 00:04:37.800
and I work with my team at the Gapminder Fundation
using new tools
00:04:37.800 --> 00:04:41.949
that help everyone make sense of the
changing world
00:04:41.949 --> 00:04:46.529
we draw all the masses of data thats now
freely available from international
00:04:46.529 --> 00:04:50.470
institutions like the UN and the World Bank
00:04:50.470 --> 00:04:54.174
and its become my mission to share
the insights from this data
00:04:54.297 --> 00:04:55.874
with anyone who listen
00:04:55.874 --> 00:04:57.054
and to reveal
00:04:57.089 --> 00:05:00.922
how statistics is nothing to be frighten
00:05:02.753 --> 00:05:05.159
Im going to provide you a view
00:05:05.159 --> 00:05:07.449
of the global health situation
00:05:07.449 --> 00:05:09.409
across mankind
00:05:09.409 --> 00:05:11.349
and Im going to do that
00:05:11.349 --> 00:05:14.527
in a hopefully enjoyable way
so relax
00:05:14.527 --> 00:05:17.210
so we did this software which displaced it like this
00:05:17.210 --> 00:05:19.168
every bubble here is a country
00:05:19.168 --> 00:05:20.633
this is China
00:05:20.633 --> 00:05:21.795
this is India
00:05:21.795 --> 00:05:23.819
the size of the bubble is the population and
00:05:23.819 --> 00:05:26.150
I done the stage of race here
00:05:26.150 --> 00:05:28.530
between the sort of yellowish Ford here
00:05:28.530 --> 00:05:30.712
and the red Toyota down there
00:05:30.712 --> 00:05:31.990
and the brownish Volvo
00:05:33.098 --> 00:05:38.110
the Toyota has a very bad start down here, and US Ford is going offroad
00:05:38.110 --> 00:05:39.319
back, and the Volvo is doing quite fine
00:05:39.319 --> 00:05:42.430
this is the war, the Toyota got off tracking out to our best country
00:05:42.430 --> 00:05:43.981
coming on to a healthier side of Sweden
00:05:43.981 --> 00:05:47.900
thats it why I sold the Volvo and both the Toyota
00:05:49.870 --> 00:05:51.820
may have been drinking forward when
china phil donahue
00:05:51.820 --> 00:05:53.119
planning by multiple
00:05:53.119 --> 00:05:56.949
diners covered in this endeavor more
student central planning and they were
00:05:56.949 --> 00:06:00.059
dennis one morning it would be looking
at
00:06:00.059 --> 00:06:02.289
united states
00:06:02.289 --> 00:06:04.340
or they broke my frame
00:06:04.340 --> 00:06:08.990
washington blvd phil rachel identified
on the back of the fda_ relating and
00:06:08.990 --> 00:06:15.990
it makes life interesting
00:06:20.789 --> 00:06:23.329
when we come to the usa_
00:06:23.329 --> 00:06:24.639
leaders in
00:06:24.639 --> 00:06:28.889
cars
00:06:28.889 --> 00:06:32.110
there are many here who shared my vision
00:06:32.110 --> 00:06:34.319
public dave accessible in use
00:06:34.319 --> 00:06:36.449
part everyone
00:06:36.449 --> 00:06:38.550
the city of san francisco
00:06:38.550 --> 00:06:39.650
is in the league
00:06:39.650 --> 00:06:43.269
opening upbeat state government every
00:06:43.269 --> 00:06:48.639
even the police department is releasing
olympics crime reports
00:06:48.639 --> 00:06:50.600
is sufficient crime
00:06:50.600 --> 00:06:56.330
turned into a wonderfully attractive man
by two of the city's computer whizzes
00:06:56.330 --> 00:07:03.330
it's communities that these pics
connection
00:07:07.850 --> 00:07:09.570
crime's going as a matter of
00:07:09.570 --> 00:07:12.349
crime reports from the san francisco
police department
00:07:12.349 --> 00:07:16.810
uh... showing young dot sohn maps uh...
for citizens to be able to see patterns
00:07:16.810 --> 00:07:19.690
of crime around their neighborhoods in
san francisco
00:07:19.690 --> 00:07:23.759
the map is not just about individual
crimes but about broader patterns the
00:07:23.759 --> 00:07:26.209
show you where crime is clustered
00:07:26.209 --> 00:07:29.569
around the city which areas of high
crimes which a resume relatively low
00:07:29.569 --> 00:07:31.409
crime
00:07:31.409 --> 00:07:32.719
me
00:07:32.719 --> 00:07:34.060
dates
00:07:34.060 --> 00:07:37.909
don't
00:07:37.909 --> 00:07:39.139
here at the top of
00:07:39.139 --> 00:07:40.440
jones street
00:07:40.440 --> 00:07:43.380
nob hill
00:07:43.380 --> 00:07:45.729
quite nice neighborhood
00:07:45.729 --> 00:07:49.440
put the crime maps jealous is the
relationship between
00:07:49.440 --> 00:07:51.680
topography attempt to crime
00:07:51.680 --> 00:07:58.680
this was a higher up the helpful s
progress
00:07:59.099 --> 00:08:03.339
to the flats
00:08:03.339 --> 00:08:05.639
especially as soon as you get into that
00:08:05.639 --> 00:08:12.639
hello rely areas to demonstrate the
prime just to skyrocket
00:08:19.889 --> 00:08:21.569
we're hearing
00:08:21.569 --> 00:08:25.690
uh... uptown tenderloin district
00:08:25.690 --> 00:08:30.559
is one of the oldest condenses
neighborhoods in san francisco
00:08:30.559 --> 00:08:32.820
this is where you go to buy drugs
00:08:32.820 --> 00:08:37.289
right around here
00:08:37.289 --> 00:08:41.930
dissociate lots of uh... aggravated
assaults lots a lot of thefts
00:08:41.930 --> 00:08:45.300
basically they're huge part of the car
that happens in the
00:08:45.300 --> 00:08:50.280
city happens just by two just viruses
moderates
00:08:50.280 --> 00:08:53.010
k
00:08:53.010 --> 00:08:55.550
can you
00:08:55.550 --> 00:09:00.400
diffused beam him please sirens in your
neighborhood you can use them and to
00:09:00.400 --> 00:09:02.820
find out why
00:09:02.820 --> 00:09:07.700
if you are denied in an unfamiliar part
of town deconcini and that was streets
00:09:07.700 --> 00:09:09.230
to avoid
00:09:09.230 --> 00:09:13.520
if the neighbor gets bergen and you can
see isn't the one off why has there been
00:09:13.520 --> 00:09:17.550
a spike in logan crime
00:09:17.550 --> 00:09:20.360
you can use their neighborhood and a
word about its safety
00:09:20.360 --> 00:09:24.259
the fact that we have the ability to
turn off all the nighttime in the middle
00:09:24.259 --> 00:09:27.329
of the day crimes and show you just the
things that are happening during the
00:09:27.329 --> 00:09:28.100
commute
00:09:28.100 --> 00:09:30.300
is a statistical operation
00:09:30.300 --> 00:09:33.410
but i think to people that are
interacting with the thing it feels very
00:09:33.410 --> 00:09:37.160
much more like they're just sort of
browsing the website or shopping on
00:09:37.160 --> 00:09:41.370
amazon there their looking at data and
they don't realize they're doing
00:09:41.370 --> 00:09:44.260
statistics
00:09:44.260 --> 00:09:46.150
what's most exciting for me
00:09:46.150 --> 00:09:49.300
is that public statistics is making
citizen
00:09:49.300 --> 00:09:50.340
powerful
00:09:50.340 --> 00:09:57.340
and the authorities moral compass
00:10:01.180 --> 00:10:04.480
we have community meetings typically use
a tent
00:10:04.480 --> 00:10:10.180
and what citizens are now doing are
bringing printouts of the maps which
00:10:10.180 --> 00:10:14.950
show where crimes are taking place and
they're demanding services from the
00:10:14.950 --> 00:10:19.070
police department and the police
department is now having to change how
00:10:19.070 --> 00:10:20.970
they police
00:10:20.970 --> 00:10:23.540
how they provide policing services
00:10:23.540 --> 00:10:29.010
highs and the date is showing what is
working and what isn't
00:10:29.010 --> 00:10:34.109
people in san francisco roadster using
public data to max social inequalities
00:10:34.109 --> 00:10:37.110
and see how to improve society
00:10:37.110 --> 00:10:40.080
possibilities parts and
00:10:40.080 --> 00:10:41.450
i think ur dream
00:10:41.450 --> 00:10:46.430
government data analysis project would
really be focused on life information on
00:10:46.430 --> 00:10:50.339
stuff that was being reported and pushed
out of the world over the internet as it
00:10:50.339 --> 00:10:55.570
was happening you know trash pickups
traffic accidents pluses
00:10:55.570 --> 00:10:58.920
through the kind of staffs gathering
power of the internet it's possible to
00:10:58.920 --> 00:11:03.900
really begin to see that the workings of
the city's displayed as a unified in
00:11:03.900 --> 00:11:08.050
advance
00:11:08.050 --> 00:11:10.080
that's why we are heading
00:11:10.080 --> 00:11:14.240
towards the world on fri data for the
former statistical insights that come
00:11:14.240 --> 00:11:14.880
from it
00:11:14.880 --> 00:11:16.400
accessible to everyone
00:11:16.400 --> 00:11:18.320
empowering us as citizens
00:11:18.320 --> 00:11:22.350
electing household dot rulers to walk
home
00:11:22.350 --> 00:11:27.180
along waiting for him as the districts
began
00:11:27.180 --> 00:11:33.800
specter's x art is central to our store
one enter our governments in our society
00:11:33.800 --> 00:11:38.210
it was all rumors out there was stalked
in the collection of statistics in the
00:11:38.210 --> 00:11:45.210
first place the order that wanted to
asked
00:11:47.410 --> 00:11:49.990
in fact wind statistics
00:11:49.990 --> 00:11:50.850
comes from
00:11:50.850 --> 00:11:52.520
misstate
00:11:52.520 --> 00:11:55.950
more than statistics began two centuries
ago
00:11:55.950 --> 00:11:59.710
one think ought to go in and spread and
never stop
00:11:59.710 --> 00:12:03.590
and guess who was fun
00:12:03.590 --> 00:12:06.600
the chinese have confucius
00:12:06.600 --> 00:12:10.310
the italians have done vientiane the
british have shakespeare
00:12:10.310 --> 00:12:12.829
and we have picked up the elected
00:12:12.829 --> 00:12:16.540
the first ever systematic collection of
statistics
00:12:16.540 --> 00:12:21.240
since the year seventeen forty nine we
have collected
00:12:21.240 --> 00:12:23.510
date dot on every
00:12:23.510 --> 00:12:24.640
that it's a bit
00:12:24.640 --> 00:12:29.640
and we are proud of
00:12:29.640 --> 00:12:34.150
the cabin baggage he courted information
from every parish in sweden
00:12:34.150 --> 00:12:34.889
it was you
00:12:34.889 --> 00:12:39.220
quantity of data and it was the first
time antigovernment
00:12:39.220 --> 00:12:46.010
could get an accurate picture on xp
00:12:46.010 --> 00:12:50.490
uh... huh
00:12:50.490 --> 00:12:53.630
the greatest military power in northern
new
00:12:53.630 --> 00:12:58.040
popeye seventeen forty nine always start
with
00:12:58.040 --> 00:13:02.420
and other conference growing stronger
00:13:02.420 --> 00:13:04.110
room armstrong
00:13:04.110 --> 00:13:06.950
for to have twenty million people
00:13:06.950 --> 00:13:10.480
enough to rival the britain and frogs
00:13:10.480 --> 00:13:14.040
de
00:13:14.040 --> 00:13:19.080
party mood swings and that's this
impasse
00:13:19.080 --> 00:13:24.540
analysis of abell back if you need that
sweet land two million inhabitants
00:13:24.540 --> 00:13:28.030
singing and playing what what pollen
decline
00:13:28.030 --> 00:13:30.920
it also had a very small population
00:13:30.920 --> 00:13:37.920
the government was horrified by this
strike what is the enemies
00:13:38.030 --> 00:13:41.350
opted to have been directed towards the
showed that many women died in
00:13:41.350 --> 00:13:42.700
childbirth
00:13:42.700 --> 00:13:44.690
and many children died young
00:13:44.690 --> 00:13:48.900
self-government took action to improve
the health of the people
00:13:48.900 --> 00:13:50.070
this
00:13:50.070 --> 00:13:54.050
is the beginning awards and sweet
00:13:54.050 --> 00:13:59.859
it took more than fifteen years before
the war strings belgians danes dutch
00:13:59.859 --> 00:14:02.540
french germans italians
00:14:02.540 --> 00:14:03.929
playing in the british
00:14:03.929 --> 00:14:10.929
caught up with sweet and in collecting
and using statistics
00:14:24.380 --> 00:14:30.030
it was cool political risks was looking
for reasons useful statistics
00:14:30.030 --> 00:14:34.530
don't skip them much more control
understanding of calls the society house
00:14:34.530 --> 00:14:37.089
working parents developing and
00:14:37.089 --> 00:14:38.499
essentially so they could
00:14:38.499 --> 00:14:43.850
control it
00:14:43.850 --> 00:14:47.970
guest governments who woke up to the
following statistics
00:14:47.970 --> 00:14:49.740
right across your room
00:14:49.740 --> 00:14:53.090
nineteenth-century societies went and
that is perfect
00:14:53.090 --> 00:14:55.160
it
00:14:55.160 --> 00:14:57.690
despite the state's doctor britton
00:14:57.690 --> 00:15:01.100
with its royal statistical sense like a
mom
00:15:01.100 --> 00:15:06.440
with soon statisticians nearby
00:15:06.440 --> 00:15:10.430
i loved looking at old copies of the
rules to cisco such a x's
00:15:10.430 --> 00:15:12.140
full of sorts polled stuff
00:15:12.140 --> 00:15:15.820
deserve wonderful papers in eighteen
forty it's which shows a matter of
00:15:15.820 --> 00:15:16.400
england
00:15:16.400 --> 00:15:21.530
and the rates of boston to uni scanty
and so you can identifier it quickly the
00:15:21.530 --> 00:15:24.240
areas with high rates of blasted
00:15:24.240 --> 00:15:28.620
bikini sangeeta toys next recital off
that norfolk seems to talk to the past
00:15:28.620 --> 00:15:33.440
decade waiting for tips
00:15:33.440 --> 00:15:37.230
one of the founders of the royal
statistical society was the great
00:15:37.230 --> 00:15:39.850
victoria mathematician an inventor
00:15:39.850 --> 00:15:42.510
charles bandage
00:15:42.510 --> 00:15:47.850
in eighteen forty two he read the latest
poll unbiased equally great victorian
00:15:47.850 --> 00:15:50.550
and for intensive
00:15:50.550 --> 00:15:53.560
vision of the scene contain the lines
00:15:53.560 --> 00:15:54.560
feel a cop
00:15:54.560 --> 00:15:58.390
and feel like i don't have a rollins
before the morning
00:15:58.390 --> 00:16:03.320
every moment dot is a man every moment
one is bold
00:16:03.320 --> 00:16:05.629
sloppiness that decision was about the
ch
00:16:05.629 --> 00:16:09.569
that he could not containing the cell's
he dashed off a letter to tennis on
00:16:09.569 --> 00:16:14.260
explaining that the course of population
growth the lines should reach
00:16:14.260 --> 00:16:18.910
every moment dot is a man and one under
sixteen is born
00:16:18.910 --> 00:16:24.020
and he added that the except finger is
one point zero six seven but something
00:16:24.020 --> 00:16:31.020
must be conceded to the north of meat
00:16:32.340 --> 00:16:36.000
in the nineteenth century scholars or we
work to do
00:16:36.000 --> 00:16:39.300
mason work in mission missile sites
00:16:39.300 --> 00:16:43.000
they were holding update almost
everything
00:16:43.000 --> 00:16:46.290
but numbers of the dylan ten minutes
00:16:46.290 --> 00:16:48.240
you have to animal house
00:16:48.240 --> 00:16:54.860
and that's what makes the tasks
00:16:54.860 --> 00:17:00.120
you're when the first statisticians
began to get to greet swimming analyzing
00:17:00.120 --> 00:17:01.050
the data
00:17:01.050 --> 00:17:03.290
pcs upon the average
00:17:03.290 --> 00:17:07.720
according to the average of everyone
with
00:17:07.720 --> 00:17:14.720
that what's so great the bottom have
reached is that you can take a mass
00:17:15.150 --> 00:17:22.150
something and use it to the single
numbers and googling
00:17:22.390 --> 00:17:28.130
and we trust is unique collected lines
reviews have reduced their content price
00:17:28.130 --> 00:17:34.560
for populations
00:17:34.560 --> 00:17:38.930
mhm
00:17:38.930 --> 00:17:43.860
with unhooked it might have called
newspaper
00:17:43.860 --> 00:17:46.400
and short tensioner
00:17:46.400 --> 00:17:49.490
exhorting people to fulfill accelerated
00:17:49.490 --> 00:17:52.700
crashed her friend against all
00:17:52.700 --> 00:17:54.850
devastating hideous
00:17:54.850 --> 00:17:56.750
powerful sink
00:17:56.750 --> 00:17:59.860
none of the second one
00:17:59.860 --> 00:18:04.140
the young man he didn't have a driving
license was driving a car under the
00:18:04.140 --> 00:18:08.330
influence of drugs and alcohol
00:18:08.330 --> 00:18:10.470
questions that the district
00:18:10.470 --> 00:18:12.169
was remarkable
00:18:12.169 --> 00:18:14.120
absolutely remarkable
00:18:14.120 --> 00:18:15.640
field at the number
00:18:15.640 --> 00:18:17.210
the people who die
00:18:17.210 --> 00:18:18.760
each year
00:18:18.760 --> 00:18:20.180
in traffic crashes
00:18:20.180 --> 00:18:24.750
incendiary a constant
00:18:24.750 --> 00:18:26.740
all these individual dents
00:18:26.740 --> 00:18:28.580
somehow when you some of them all up
00:18:28.580 --> 00:18:30.320
stays the same number
00:18:30.320 --> 00:18:31.960
every year
00:18:31.960 --> 00:18:33.230
and every year
00:18:33.230 --> 00:18:36.130
two-and-a-half times as many men
00:18:36.130 --> 00:18:39.120
traffic crashes since women and it's a
constant
00:18:39.120 --> 00:18:40.409
and every year
00:18:40.409 --> 00:18:42.970
the rich indulgent estoppel
00:18:42.970 --> 00:18:44.730
to reaching into it
00:18:44.730 --> 00:18:47.570
that these remarkable regularities
00:18:47.570 --> 00:18:49.130
so that
00:18:49.130 --> 00:18:51.560
individual particular things
00:18:51.560 --> 00:18:57.030
sum up sum up into a social phenomena
00:18:57.030 --> 00:18:59.950
let's see what's we then have done the
houston post about
00:18:59.950 --> 00:19:03.850
that social programs that's where we
were reminded shift attention a little
00:19:03.850 --> 00:19:05.410
bit changing running
00:19:05.410 --> 00:19:10.559
and averages from entire country when is
the average income tried mechanically
00:19:10.559 --> 00:19:13.390
family size and called the marshall
00:19:13.390 --> 00:19:15.010
but they are getting up
00:19:15.010 --> 00:19:16.310
billy wright was a bit
00:19:16.310 --> 00:19:18.000
singapore advice toddler
00:19:18.000 --> 00:19:20.659
defeat is the most probably found in the
world
00:19:20.659 --> 00:19:24.429
marshall and only played should be a
real if looking at the time for them to
00:19:24.429 --> 00:19:28.210
get independent but then they started to
grow the economy and they made a social
00:19:28.210 --> 00:19:29.920
investment they caught the wave in the
area
00:19:29.920 --> 00:19:32.960
throughout the magnificent help system
that he could help you with this
00:19:32.960 --> 00:19:39.960
remember never forget what i had nothing
on the
00:19:41.530 --> 00:19:42.350
useful
00:19:42.350 --> 00:19:49.330
as averages are safe don't tell you the
whole story
00:19:49.330 --> 00:19:50.429
on average
00:19:50.429 --> 00:19:53.210
swedish people i was like a less than
two x
00:19:53.210 --> 00:19:57.360
this is because a few people only have
one leg or no legs
00:19:57.360 --> 00:19:58.390
and no one
00:19:58.390 --> 00:20:00.260
affiliate
00:20:00.260 --> 00:20:02.710
almost everybody in sweden
00:20:02.710 --> 00:20:07.020
has more blended average number alexis
00:20:07.020 --> 00:20:14.020
variation that guy's yes those important
that the other
00:20:16.960 --> 00:20:19.460
but how do you get a handle on millions
00:20:19.460 --> 00:20:23.530
for this to transform numbers into
shapes
00:20:23.530 --> 00:20:28.760
next look again at the number of adult
women in sweden for heath heights
00:20:28.760 --> 00:20:30.659
looking digitizes shade
00:20:30.659 --> 00:20:36.810
shows how much they are heights vary
from the average in hawaii blackberries
00:20:36.810 --> 00:20:39.220
the shape to set the date that makes its
way
00:20:39.220 --> 00:20:41.620
if distribution
00:20:41.620 --> 00:20:46.200
this if the income distribution all
shylock nineteen seventy
00:20:46.200 --> 00:20:50.179
this is being done here is the jewish
and or the united states nineteen
00:20:50.179 --> 00:20:54.190
seventy-eight almost no overlap and what
has happened
00:20:54.190 --> 00:20:58.590
china's growing it's not so equal any
longer and he's appearing here
00:20:58.590 --> 00:21:01.030
overlooking the united states
00:21:01.030 --> 00:21:08.030
almost like a ghost isn't it time
previous get a
00:21:17.430 --> 00:21:23.420
physician-assisted slow distribution
systems once she says to himself says
00:21:23.420 --> 00:21:26.270
the states
00:21:26.270 --> 00:21:29.769
dance card president on the south
estimated
00:21:29.769 --> 00:21:32.580
he did the machine that could be a
producer
00:21:32.580 --> 00:21:37.390
and he found thinking so many different
sets of nationalist betty mehendi than
00:21:37.390 --> 00:21:39.630
normal distribution
00:21:39.630 --> 00:21:42.720
sleepless pieces arms
00:21:42.720 --> 00:21:44.169
non-cancerous
00:21:44.169 --> 00:21:46.720
or stephen beard families
00:21:46.720 --> 00:21:52.370
the normal distribution shaped return
time and time again
00:21:52.370 --> 00:21:56.560
under statisticians soon from any on the
regulations
00:21:56.560 --> 00:22:02.360
each produced by a particular kind of
natural or social
00:22:02.360 --> 00:22:03.940
everest at the station
00:22:03.940 --> 00:22:05.780
their favorite
00:22:05.780 --> 00:22:10.460
fossil distribution procedure is raised
by favorite distribution texas absolute
00:22:10.460 --> 00:22:15.970
crap done
00:22:15.970 --> 00:22:20.460
inquest solution a describes how likely
it is that out of the ordinary things
00:22:20.460 --> 00:22:25.610
will happen emotional on the muscle what
we know that when everybody gets revises
00:22:25.610 --> 00:22:29.640
enough we won't lose again three buses
and gruesome
00:22:29.640 --> 00:22:34.510
amazingly the partial shade please show
us the probability that in any given our
00:22:34.510 --> 00:22:39.860
word when we get four five or six months
is one of us is the time
00:22:39.860 --> 00:22:43.900
the exact shape changes pediatrics
00:22:43.900 --> 00:22:47.919
but whether it's how many people will
win the lottery check
00:22:47.919 --> 00:22:52.740
or how many people in the former cold
symptoms mimic plus one shape we give
00:22:52.740 --> 00:22:57.690
the probabilities
00:22:57.690 --> 00:23:01.249
the wonderful example but this was
applied to the late nineteenth century
00:23:01.249 --> 00:23:05.470
was to can teach she is the number of
prashant offices calorie offices and
00:23:05.470 --> 00:23:07.460
kicked to death by the horses
00:23:07.460 --> 00:23:11.380
as some use of the nuns and here's the
ones from his too
00:23:11.380 --> 00:23:14.120
up to seven i think one ticking died
here
00:23:14.120 --> 00:23:19.970
and buttons distribution permanent
bizarre taf crushing cavalry officer
00:23:19.970 --> 00:23:21.720
skip taft cases
00:23:21.720 --> 00:23:25.380
usually obey the constitution
00:23:25.380 --> 00:23:32.380
just
00:23:34.520 --> 00:23:36.530
tests
00:23:36.530 --> 00:23:41.780
cert
00:23:41.780 --> 00:23:43.460
going
00:23:43.460 --> 00:23:47.400
so statisticians use she c you then
00:23:47.400 --> 00:23:53.320
candidate ability using images of warmth
kinds to communicate statistics to a
00:23:53.320 --> 00:23:54.789
wider public
00:23:54.789 --> 00:23:56.090
because
00:23:56.090 --> 00:23:57.980
it's a story in the numbers
00:23:57.980 --> 00:24:03.780
he's told by a beautiful and clever
image thank everyone understand
00:24:03.780 --> 00:24:06.529
of the pioneers of statistically
graphics
00:24:06.529 --> 00:24:13.529
my favorite is florence unlike
00:24:22.710 --> 00:24:26.380
tire they're not many people don't
realize it actually she was there was a
00:24:26.380 --> 00:24:28.190
passion statistician
00:24:28.190 --> 00:24:31.190
and not just of a vehicle out
00:24:31.190 --> 00:24:35.270
she said that understand god's thoughts
we must study statistics but these are
00:24:35.270 --> 00:24:37.450
the measure of his purpose
00:24:37.450 --> 00:24:42.190
it's just it's less river villages
detailer alternative
00:24:42.190 --> 00:24:46.380
flashes nine-years-old she started
collecting data heard it was different
00:24:46.380 --> 00:24:50.820
fruits and vegetables she felt prelim
into different tables trying to organize
00:24:50.820 --> 00:24:55.010
them in some standard for so we have one
of the playing fields first to school
00:24:55.010 --> 00:25:02.010
tables at the age of mine
00:25:04.860 --> 00:25:08.039
immediately fifties florence nightingale
00:25:08.039 --> 00:25:11.850
went to the crime he had to cancel
british cash infusion
00:25:11.850 --> 00:25:14.980
she was forty five i want to discuss
00:25:14.980 --> 00:25:18.160
these soldiers being blown to bits on
the battlefield
00:25:18.160 --> 00:25:22.290
there were many many more soldiers dying
from the sees his state court in the
00:25:22.290 --> 00:25:25.590
armies filthy hostels
00:25:25.590 --> 00:25:29.409
so florence nightingale began compton
content
00:25:29.409 --> 00:25:35.730
for two years she recorded mortality
data unique too touristy
00:25:35.730 --> 00:25:40.980
it was over she persuaded the government
to set up a royal commission of inquiry
00:25:40.980 --> 00:25:45.100
and gathered her dating devastating
report
00:25:45.100 --> 00:25:48.269
this amended took place in the
statistically history books
00:25:48.269 --> 00:25:50.340
on the graphics shields
00:25:50.340 --> 00:25:51.830
and one in particular
00:25:51.830 --> 00:25:54.299
the polar area graph
00:25:54.299 --> 00:25:56.650
for each month of the war
00:25:56.650 --> 00:26:00.360
usually low wage represented the
soldiers who are dying from preventable
00:26:00.360 --> 00:26:02.390
disease
00:26:02.390 --> 00:26:05.370
much more of a bridges more deaths from
00:26:05.370 --> 00:26:07.500
owns and the plaque wages
00:26:07.500 --> 00:26:10.990
deaths from accidents and public works
00:26:10.990 --> 00:26:13.840
nineteen days graphics mercele cleaner
00:26:13.840 --> 00:26:14.619
me working
00:26:14.619 --> 00:26:17.330
possible treatment
00:26:17.330 --> 00:26:20.049
the usual cigarette from start to get
storms just adjust
00:26:20.049 --> 00:26:24.530
tables in tables and figures are you've
got to really tedious stocks that unless
00:26:24.530 --> 00:26:28.220
you're actually dedicated statisticians
will be quite difficult spot the patents
00:26:28.220 --> 00:26:29.250
quite naturally
00:26:29.250 --> 00:26:30.840
but an visualizations
00:26:30.840 --> 00:26:34.710
they tell a story to tell a story
immediately and use of color
00:26:34.710 --> 00:26:36.400
and usage schenk
00:26:36.400 --> 00:26:39.850
you do to really tell a powerful story
and now it is of course we can make
00:26:39.850 --> 00:26:41.370
things move as well
00:26:41.370 --> 00:26:44.790
for such here would love to have played
with should have been
00:26:44.790 --> 00:26:49.590
jews wonderful animation semantics
00:26:49.590 --> 00:26:51.380
ru
00:26:51.380 --> 00:26:54.149
today a hundred and fifty years old
00:26:54.149 --> 00:26:57.959
nightingale is graphics are rightly
regarded as a class in
00:26:57.959 --> 00:27:02.109
they lead to a revolution in nursing
health care in hygiene hospitals
00:27:02.109 --> 00:27:06.900
worldwide wheat cv innumerable times
that
00:27:06.900 --> 00:27:11.640
the statistical graphics has become an
art form where it's a very old learned
00:27:11.640 --> 00:27:17.180
by deciding to restore passionate about
it is relaxing there
00:27:17.180 --> 00:27:19.970
will lure
00:27:19.970 --> 00:27:23.169
remember more
00:27:23.169 --> 00:27:27.420
in rice's two billion pounds are ground
00:27:27.420 --> 00:27:31.100
this image arose frustration with the
reporting of
00:27:31.100 --> 00:27:34.480
billion panna medicine a media fire in
japan in pounds for this war
00:27:34.480 --> 00:27:37.870
fifty billion pounds this or still
doesn't common sense is known as the two
00:27:37.870 --> 00:27:41.270
enormous to get your mind round source
craigslist data from various news
00:27:41.270 --> 00:27:43.460
sources in trade the star graham
00:27:43.460 --> 00:27:49.050
squares hero scaled according to the
billion pan-am outs
00:27:49.050 --> 00:27:51.880
we see numbers visualize like this
00:27:51.880 --> 00:27:54.240
the start of a different kind of
relationship with them
00:27:54.240 --> 00:27:57.029
uh... because of see patterns you see
the scale of them
00:27:57.029 --> 00:27:59.710
here at the corner this little square
thirty seven billion
00:27:59.710 --> 00:28:03.309
i was the cost projected cost iraq war
in two thousand three
00:28:03.309 --> 00:28:05.460
as you can see it's grown exponentially
00:28:05.460 --> 00:28:06.589
of the last few years
00:28:06.589 --> 00:28:08.309
so the total cost house around about
00:28:08.309 --> 00:28:11.210
two thousand five hundred billion
00:28:11.210 --> 00:28:14.540
it's funny 'cause when you visualize two
sticks on this you understand them when
00:28:14.540 --> 00:28:15.970
you understand them
00:28:15.970 --> 00:28:22.970
you can really start puts things into
perspective
00:28:24.530 --> 00:28:28.900
we should learn stations right at the
heart of my own walked to
00:28:28.900 --> 00:28:31.140
i teach global health
00:28:31.140 --> 00:28:33.940
and i know having the data is not enough
00:28:33.940 --> 00:28:40.000
i have to show it in ways people both
enjoy and i understand
00:28:40.000 --> 00:28:44.580
i'm going to try something i've never
done before animating the dating really
00:28:44.580 --> 00:28:45.980
stinks
00:28:45.980 --> 00:28:50.870
with a bit of technical assistance from
the crew
00:28:50.870 --> 00:28:52.500
so here we go
00:28:52.500 --> 00:28:54.360
first taxes for help
00:28:54.360 --> 00:28:55.940
life expectancy
00:28:55.940 --> 00:29:01.620
from twenty-five years seventy-five
years and out here access for work
00:29:01.620 --> 00:29:03.159
income per person
00:29:03.159 --> 00:29:04.240
four hundred
00:29:04.240 --> 00:29:07.040
four thousand and forty thousand dollars
00:29:07.040 --> 00:29:09.600
so down here
00:29:09.600 --> 00:29:10.649
and sick
00:29:10.649 --> 00:29:15.580
and cop here is rich and health
00:29:15.580 --> 00:29:18.840
i want to show the world two hundred
years ago
00:29:18.840 --> 00:29:21.410
in eighteen ten
00:29:21.410 --> 00:29:23.000
here come all the countries
00:29:23.000 --> 00:29:26.289
you know ground station away hideously
00:29:26.289 --> 00:29:29.850
acrobats tells us army or and the
american deal
00:29:29.850 --> 00:29:34.179
and the soyuz over the country bubble
soulless eyes of the population
00:29:34.179 --> 00:29:38.760
and eighteen ten it is pretty crowded
down there wasn't it congress were
00:29:38.760 --> 00:29:42.539
sickened pork life expectancy will be
lost
00:29:42.539 --> 00:29:47.460
in oregon please everyting you kate and
then the lance was like it better off
00:29:47.460 --> 00:29:48.799
not much
00:29:48.799 --> 00:29:49.720
and now
00:29:49.720 --> 00:29:52.920
wise dot the world
00:29:52.920 --> 00:29:57.410
industrial revolution makes companies in
europe and elsewhere
00:29:57.410 --> 00:29:59.210
move away from the rest
00:29:59.210 --> 00:30:04.049
apologize countries in asia and africa
best backed down there
00:30:04.049 --> 00:30:08.540
and eventually the western countries get
healthier and healthier
00:30:08.540 --> 00:30:14.430
and now we slow down to show the impact
on the first world war and the spanish
00:30:14.430 --> 00:30:15.850
flu epidemic
00:30:15.850 --> 00:30:18.549
what does it
00:30:18.549 --> 00:30:22.990
and now i speed up through the nineteen
twenties and the nineteen thirties and
00:30:22.990 --> 00:30:26.910
in spite of the great depression was to
come visit for john doe was greater
00:30:26.910 --> 00:30:32.510
wealth and helped and some others try to
follow but most interesting down here
00:30:32.510 --> 00:30:37.340
up often the tragedies of the second
world war we stop a bit to look at the
00:30:37.340 --> 00:30:39.670
world in nineteen forty eight
00:30:39.670 --> 00:30:42.200
nineteen forty eight was a great year
00:30:42.200 --> 00:30:43.610
the war was over
00:30:43.610 --> 00:30:48.080
sweden topped the metal table at the
winter olympics and i was born
00:30:48.080 --> 00:30:49.480
buffet differences
00:30:49.480 --> 00:30:53.019
between the congress of the world was
wider than ever
00:30:53.019 --> 00:30:57.140
united states was in the frog japan was
catching up
00:30:57.140 --> 00:31:02.030
brazil is way behind iran was getting a
little richard from lloyd but still had
00:31:02.030 --> 00:31:03.080
short life
00:31:03.080 --> 00:31:05.509
and dictation giants
00:31:05.509 --> 00:31:09.759
china india pakistan bangladesh in
indonesia and they were still poor
00:31:09.759 --> 00:31:11.190
instinct on here
00:31:11.190 --> 00:31:12.999
but he'll play is about to happen
00:31:12.999 --> 00:31:14.320
gila delegate
00:31:14.320 --> 00:31:18.840
in my lifetime former colonies gained
independence and then finally they
00:31:18.840 --> 00:31:23.620
starting to get healthier and healthy it
and healthier and in the nineteen
00:31:23.620 --> 00:31:24.510
seventies
00:31:24.510 --> 00:31:28.060
then countries indication that the
american student of catch up with the
00:31:28.060 --> 00:31:30.080
western countries the meeting with me
00:31:30.080 --> 00:31:34.509
emerging economies sunny method of
follows some african rostock in civil
00:31:34.509 --> 00:31:39.830
war and i was hit by a charity and now
we can see the wall today in the most
00:31:39.830 --> 00:31:43.340
up-to-date statistics
00:31:43.340 --> 00:31:45.980
most people today in the mid
00:31:45.980 --> 00:31:48.919
but then you should get friends at the
same time between them
00:31:48.919 --> 00:31:51.630
best of congress and the worst of cuts
00:31:51.630 --> 00:31:52.390
and i will say
00:31:52.390 --> 00:31:54.800
lives in the qualities with the income
00:31:54.800 --> 00:31:58.610
these problems shel country average s
but i just think that
00:31:58.610 --> 00:32:02.420
they china i can split it into provinces
00:32:02.420 --> 00:32:05.500
bearing goes shanghai
00:32:05.500 --> 00:32:11.370
it is the same boat as it related and a
interpol reliant on those white show it
00:32:11.370 --> 00:32:12.759
is like pakistan
00:32:12.759 --> 00:32:15.200
any text picket fence
00:32:15.200 --> 00:32:19.860
the rural parts are like calling africa
00:32:19.860 --> 00:32:21.070
and yet
00:32:21.070 --> 00:32:25.360
despite the enormous disparities to date
we have seen two hundred years of
00:32:25.360 --> 00:32:27.420
remarkable progress
00:32:27.420 --> 00:32:31.890
that's a huge historical gap between the
west and the rest is not closed eight
00:32:31.890 --> 00:32:35.860
we have become an entirely new
converging world
00:32:35.860 --> 00:32:39.580
and i see it clearly friend into the
future we can you need to treat a green
00:32:39.580 --> 00:32:44.410
technology and bs its full impossible
that everyone can make it to the healthy
00:32:44.410 --> 00:32:48.550
wealthy corner
00:32:48.550 --> 00:32:50.120
wenjie dot c
00:32:50.120 --> 00:32:53.929
for the last few minutes is a story of
two hundred congress
00:32:53.929 --> 00:32:56.610
showing over two hundred years and
beyond
00:32:56.610 --> 00:32:58.450
it involve plotting on
00:32:58.450 --> 00:33:01.340
one hundred and twenty thousand numbers
00:33:01.340 --> 00:33:08.139
pretty neat
00:33:08.139 --> 00:33:09.149
so
00:33:09.149 --> 00:33:13.450
with statistics we can begin to see
things as they really are
00:33:13.450 --> 00:33:17.290
from tables update dental and reduce
these tributions and issuing this
00:33:17.290 --> 00:33:23.330
nation's statistics gives us a clear
description of the world
00:33:23.330 --> 00:33:28.630
athletes distinctly cannot only
discovered much is happening
00:33:28.630 --> 00:33:35.630
bottles explore why by using the
powerful analytical method correlation
00:33:35.679 --> 00:33:38.429
just looking at one thing at the time is
intended for
00:33:38.429 --> 00:33:42.150
you've got to look at the relationships
between finance has a chance to have a
00:33:42.150 --> 00:33:43.380
very together
00:33:43.380 --> 00:33:46.590
in that sport correlation is a pack and
that's why i stopped trying to
00:33:46.590 --> 00:33:52.940
understand to crysis is that they're
really going on in the world in society
00:33:52.940 --> 00:33:57.240
most of us today would recognize that
crying correlates to poverty
00:33:57.240 --> 00:34:00.340
that infection correlates to poor
sanitation
00:34:00.340 --> 00:34:06.810
and that knowledge of statistics
correlates to being great at balancing
00:34:06.810 --> 00:34:10.589
relations can be very pretty
00:34:10.589 --> 00:34:13.149
i got a joke about silly correlations
00:34:13.149 --> 00:34:15.820
that was that the american i was afraid
of a heart attack
00:34:15.820 --> 00:34:18.969
and he found out that the japanese
00:34:18.969 --> 00:34:23.019
the marriage of fat
00:34:23.019 --> 00:34:25.509
action this heart attacks than the
american
00:34:25.509 --> 00:34:26.670
but on the other hand
00:34:26.670 --> 00:34:28.789
also follow up with a friend
00:34:28.789 --> 00:34:32.829
eat as much practically americans and
they want to know why
00:34:32.829 --> 00:34:39.829
also and less artifacts
00:34:41.889 --> 00:34:45.339
this next round of the lead
groundbreaking coronation
00:34:45.339 --> 00:34:48.970
is the link that was established in the
nineteen fifties between smoking and
00:34:48.970 --> 00:34:51.299
lung cancer
00:34:51.299 --> 00:34:53.689
notlong after the second world war
00:34:53.689 --> 00:34:55.969
a british doctors reach abroad
00:34:55.969 --> 00:34:59.969
investigated lung cancer patient in
twenty long on hospitals
00:34:59.969 --> 00:35:04.349
and he became system that the only thing
they have in common with smoking
00:35:04.349 --> 00:35:05.500
so certain
00:35:05.500 --> 00:35:07.489
she stopped smoking inside
00:35:07.489 --> 00:35:08.789
on the page
00:35:08.789 --> 00:35:11.449
runs social
00:35:11.449 --> 00:35:14.499
moderates a discussion of the early days
just
00:35:14.499 --> 00:35:18.489
linking smoking to lung cancer
simplicial smoking shortly that thing
00:35:18.489 --> 00:35:21.849
that we've come about life's account be
bad for you
00:35:21.849 --> 00:35:24.009
maybe it's jeans
00:35:24.009 --> 00:35:28.450
maybe people would genetically
predisposed to get lung cancer are also
00:35:28.450 --> 00:35:33.119
genetically predisposed to smoke
00:35:33.119 --> 00:35:36.649
maybe it's not the smoking maybe its air
pollution
00:35:36.649 --> 00:35:41.809
that smokers are somehow more exposed to
air pollution non-smokers
00:35:41.809 --> 00:35:43.149
maybe it's not
00:35:43.149 --> 00:35:45.289
smoking maybe it's public
00:35:45.289 --> 00:35:51.769
so now we've got three alternate
explanations apart from chance
00:35:51.769 --> 00:35:53.589
parties coronation
00:35:53.589 --> 00:35:58.300
digging flight cause-and-effect returned
on created the biggest existed to study
00:35:58.300 --> 00:35:59.969
of smoking thing
00:35:59.969 --> 00:36:03.719
he began cracking the life some forty
thousand british dot
00:36:03.719 --> 00:36:06.910
some of whom smoke and some of whom did
00:36:06.910 --> 00:36:09.869
and get this enough data to acquire
anything in mind
00:36:09.869 --> 00:36:14.189
if the doctor smoke within likelihood of
getting cancer
00:36:14.189 --> 00:36:18.229
eventually he not only showed a
correlation between smoking and lung
00:36:18.229 --> 00:36:23.299
cancer platforms to a correlation
between stopping smoking and reducing
00:36:23.299 --> 00:36:24.479
the risk
00:36:24.479 --> 00:36:28.679
this messiahs at its best
00:36:28.679 --> 00:36:31.449
what correlations do not replace
00:36:31.449 --> 00:36:32.999
he's human sort
00:36:32.999 --> 00:36:35.869
good think about what it means
00:36:35.869 --> 00:36:40.079
knew what a good scientist has if he
comes up with the correlation
00:36:40.079 --> 00:36:40.859
ease
00:36:40.859 --> 00:36:45.399
try as hard as she or he possibly can to
disprove it
00:36:45.399 --> 00:36:47.489
to break it down to get rid of it
00:36:47.489 --> 00:36:49.079
to try and refute it
00:36:49.079 --> 00:36:53.459
and if it was stones all those efforts
00:36:53.459 --> 00:36:56.439
demolishing it and it's still standing
out
00:36:56.439 --> 00:37:00.719
then cautiously say weary might have
something here
00:37:00.719 --> 00:37:03.209
bernard
00:37:03.209 --> 00:37:05.699
bed
00:37:05.699 --> 00:37:08.639
benetton
00:37:08.639 --> 00:37:13.919
character
00:37:13.919 --> 00:37:16.739
ho
00:37:16.739 --> 00:37:22.049
henry lee and the scientist date that is
to be able to jim and finance
00:37:22.049 --> 00:37:26.659
the good news is that the more we had
the more correlations will find the more
00:37:26.659 --> 00:37:28.399
theories will test
00:37:28.399 --> 00:37:35.399
and the more discoveries that we are
likely to make
00:37:35.569 --> 00:37:40.920
and history shows park total sum of the
information bruising usually keeps as we
00:37:40.920 --> 00:37:43.389
develop new technologies
00:37:43.389 --> 00:37:47.469
the mention of the printing press kicked
off the first day an information
00:37:47.469 --> 00:37:49.199
explosion
00:37:49.199 --> 00:37:53.089
he complained about the glove the books
that i mean print by the year is
00:37:53.089 --> 00:37:55.339
seventeen hundred
00:37:55.339 --> 00:38:02.299
didn't make six this tax each as high is
montane forest
00:38:02.299 --> 00:38:05.070
then starting in the nineteenth century
00:38:05.070 --> 00:38:07.720
they came a second information
revolution
00:38:07.720 --> 00:38:11.509
with the telegraph gramophone and can
later
00:38:11.509 --> 00:38:13.509
radio and tv
00:38:13.509 --> 00:38:17.119
to obtain long-term information exploded
00:38:17.119 --> 00:38:20.839
them by the nineteen fifties the
information available to us all
00:38:20.839 --> 00:38:25.509
had multiplied six thousand pints
00:38:25.509 --> 00:38:30.279
thanks to the computer linked to the
independently went digital
00:38:30.279 --> 00:38:37.279
mainland of belize we have now he's
wanting mentioned the
00:38:39.359 --> 00:38:44.409
seemed a little denoble politically than
right
00:38:44.409 --> 00:38:51.249
printed page pete wilson killed by
00:38:51.249 --> 00:38:53.390
five megabytes enough of it
00:38:53.390 --> 00:38:57.420
people aren't insured
00:38:57.420 --> 00:38:58.950
pending dot lights
00:38:58.950 --> 00:39:05.929
that's a dvd movies
00:39:05.929 --> 00:39:10.409
to metabolize is himself millions off of
things
00:39:10.409 --> 00:39:14.739
faceplate everyday
00:39:14.739 --> 00:39:20.119
stoplights is the data recorder every
second by the world's largest particle
00:39:20.119 --> 00:39:25.119
accelerator so much higher tiny
fractions
00:39:25.119 --> 00:39:26.879
six exabytes
00:39:26.879 --> 00:39:33.879
is what you have secrecy
00:39:37.749 --> 00:39:39.529
but we had that's not good
00:39:39.529 --> 00:39:44.499
in two thousand nine the internet added
up to five hundred except and in two
00:39:44.499 --> 00:39:47.559
thousand to in yes the here
00:39:47.559 --> 00:39:54.559
stepping down to more than what is it
that way
00:39:55.299 --> 00:39:56.939
back in the world
00:39:56.939 --> 00:39:59.919
here to tell all this data into print
00:39:59.919 --> 00:40:02.610
it would make ninety stacks of both
00:40:02.610 --> 00:40:08.069
sandy treat you from here and all the
way you do this
00:40:08.069 --> 00:40:10.949
dazed and then you just answering
00:40:10.949 --> 00:40:17.509
with today's computers and statistics
i'm confident we can handle it
00:40:17.509 --> 00:40:21.819
when it comes to on the date on the
internet powerhouse of statistical
00:40:21.819 --> 00:40:28.819
analysis is the silicon valley giant
00:40:33.399 --> 00:40:37.659
the average person over their lifetime
is exposed to about a hundred million
00:40:37.659 --> 00:40:43.209
words conversation so if you multiply
that by the six billion people on the
00:40:43.209 --> 00:40:44.450
planet data
00:40:44.450 --> 00:40:48.300
brown words about the world to the
number of words that will has available
00:40:48.300 --> 00:40:51.899
to that he wants to talk
00:40:51.899 --> 00:40:54.059
whose computers happen
00:40:54.059 --> 00:40:58.749
finally every document webpage in image
they can find
00:40:58.749 --> 00:41:04.639
fagan hunt for patents and correlations
in all this date dot doing statistics on
00:41:04.639 --> 00:41:06.669
them that sixty
00:41:06.669 --> 00:41:07.749
and for me
00:41:07.749 --> 00:41:12.499
will be has one project that the
particular excite statistical language
00:41:12.499 --> 00:41:14.869
translation
00:41:14.869 --> 00:41:19.819
who wanted to provide access to all the
winds information matter what languages
00:41:19.819 --> 00:41:22.540
there's just so much information on the
internet you can help
00:41:22.540 --> 00:41:27.710
to translate it all bike and into every
possible language we figured we'd have
00:41:27.710 --> 00:41:32.989
to be able to do machine translation
00:41:32.989 --> 00:41:37.249
in the past program has tried to teach
their computers to free speech language
00:41:37.249 --> 00:41:39.689
as the system dramatically lose
00:41:39.689 --> 00:41:42.819
much like the women which is much more
difficult
00:41:42.819 --> 00:41:44.219
his feet in front
00:41:44.219 --> 00:41:48.880
because no one set the rules to cash in
online which informix subtlety in
00:41:48.880 --> 00:41:51.169
ambiguity
00:41:51.169 --> 00:41:53.019
having eaten hot lunch
00:41:53.019 --> 00:41:55.169
because depart
00:41:55.169 --> 00:41:57.069
well that's all this weekend correct
00:41:57.069 --> 00:42:01.279
written like that it would imply that
the coach has eaten lunch
00:42:01.279 --> 00:42:04.309
it will be far letters
00:42:04.309 --> 00:42:06.279
having eating my lunch
00:42:06.279 --> 00:42:09.249
we departed in the collection
00:42:09.249 --> 00:42:10.480
those rules are helpful
00:42:10.480 --> 00:42:14.440
and they are are useful most the time
but they don't turn out that the true
00:42:14.440 --> 00:42:18.809
all the time ten the insight using
statistical machine translation
00:42:18.809 --> 00:42:23.059
is saying wealthy gotta have all these
exceptions anyways maybe you can get by
00:42:23.059 --> 00:42:27.309
without having its wheels maybe he can
treat everything as an exception nasa
00:42:27.309 --> 00:42:34.309
says we want to
00:42:37.899 --> 00:42:41.759
but the computer is doing when when he's
learning how to translate is too low
00:42:41.759 --> 00:42:47.149
in correlations between boats and
correlations between phrases so we'd be
00:42:47.149 --> 00:42:50.239
we feed this system very large amounts
of data
00:42:50.239 --> 00:42:53.969
and then the system is seeing that uh...
certain good research the phrase
00:42:53.969 --> 00:42:58.739
correlates very often to the other
language
00:42:58.739 --> 00:43:00.069
websites
00:43:00.069 --> 00:43:04.909
parents office translation between in
your fifty seven different languages
00:43:04.909 --> 00:43:09.509
it does this purely a statistically
having correlated the huge collection of
00:43:09.509 --> 00:43:12.479
multilingual texans
00:43:12.479 --> 00:43:16.139
the people to build the system don't
need to know chinese in order to the
00:43:16.139 --> 00:43:19.719
chinese strings just a little beaten
noah rebekah but the expertise that's
00:43:19.719 --> 00:43:23.519
needed is basically knowledge of
statistics knowledge of computer science
00:43:23.519 --> 00:43:27.700
that knowledge of infrastructure to bolt
those very large computational systems
00:43:27.700 --> 00:43:29.999
that we are building for doing that
00:43:29.999 --> 00:43:33.729
okay so then i'm going to invite yahoo
com political good from my office in
00:43:33.729 --> 00:43:34.699
stockholm
00:43:34.699 --> 00:43:40.869
to try the translator for myself
00:43:40.869 --> 00:43:44.199
okay
00:43:44.199 --> 00:43:46.329
linden
00:43:46.329 --> 00:43:49.839
or
00:43:49.839 --> 00:43:54.319
located so it says remains finance
ministers appointed
00:43:54.319 --> 00:43:56.609
and a gold ring in your your
00:43:56.609 --> 00:44:03.609
so here's the problem is that if you're
a cash
00:44:06.199 --> 00:44:09.029
rockport fat cats
00:44:09.029 --> 00:44:12.910
starts
00:44:12.910 --> 00:44:14.489
from school
00:44:14.489 --> 00:44:15.689
na
00:44:15.689 --> 00:44:18.259
these corp
00:44:18.259 --> 00:44:22.479
and his same sex partnerships as
stockholm snoopy shop and his partners
00:44:22.479 --> 00:44:25.029
three-years-old
00:44:25.029 --> 00:44:27.029
hume
00:44:27.029 --> 00:44:28.700
expire
00:44:28.700 --> 00:44:30.769
except after bachchan
00:44:30.769 --> 00:44:35.930
okay so so that's uh... those kinds of
riches and her one of the challenges in
00:44:35.930 --> 00:44:39.829
in translation to be really does
frighten theater kathleen sharp slide
00:44:39.829 --> 00:44:42.529
you excuse me
00:44:42.529 --> 00:44:45.389
causing more often than not it would
probably be it is
00:44:45.389 --> 00:44:46.259
crack
00:44:46.259 --> 00:44:51.399
or the fact that madison i do you do you
do it would be all that and wallet dot
00:44:51.399 --> 00:44:55.429
oc oxford
00:44:55.429 --> 00:44:59.859
cream when sweeney's taking part in a
pickle it's not to win but to be below
00:44:59.859 --> 00:45:00.839
which
00:45:00.839 --> 00:45:04.889
that yes the
00:45:04.889 --> 00:45:06.849
cartwright
00:45:06.849 --> 00:45:09.819
uh... drug-related
00:45:09.819 --> 00:45:10.509
xd
00:45:10.509 --> 00:45:11.589
atpase
00:45:11.589 --> 00:45:13.679
i don't
00:45:13.679 --> 00:45:14.639
aspect
00:45:14.639 --> 00:45:16.739
apart by fax
00:45:16.739 --> 00:45:19.939
partnerships chuck
00:45:19.939 --> 00:45:21.450
controlled lead real good
00:45:21.450 --> 00:45:25.789
but if they succeed with what's next
that'll be a remarkable
00:45:25.789 --> 00:45:29.839
one other exciting possibilities is
combining machine translation technology
00:45:29.839 --> 00:45:31.739
with the speech recognition technology
00:45:31.739 --> 00:45:35.359
now both of these are statistical in
nature them
00:45:35.359 --> 00:45:38.089
machine translation relies on the
statistics
00:45:38.089 --> 00:45:40.379
of mapping from one language to another
00:45:40.379 --> 00:45:44.689
personally speech recognition relies on
the statistics of matt mapping from a
00:45:44.689 --> 00:45:47.419
sound form to the words
00:45:47.419 --> 00:45:48.849
when we put them together
00:45:48.849 --> 00:45:53.089
now we have the capability of having
instant conversation between two people
00:45:53.089 --> 00:45:57.759
that don't speak a common lecture that
uh... i can talk to you in my language
00:45:57.759 --> 00:46:02.699
you hear me in your language and you can
answer back and in real time you can
00:46:02.699 --> 00:46:04.259
make that translation
00:46:04.259 --> 00:46:11.259
can bring to get people together and
allow them to speak
00:46:20.809 --> 00:46:24.540
the intimate is just one how many
technologies
00:46:24.540 --> 00:46:28.339
created together massive amounts of data
00:46:28.339 --> 00:46:32.619
scientists studying our send our
environment
00:46:32.619 --> 00:46:38.189
now use and implantable ancients
instruments to mention the processes up
00:46:38.189 --> 00:46:42.179
our plans
00:46:42.179 --> 00:46:49.179
all around us are sensors continuously
measuring temperature water flow
00:46:49.629 --> 00:46:55.219
and high in order to our satellites busy
engaging cloud formations forest growth
00:46:55.219 --> 00:46:56.729
and snow cover
00:46:56.729 --> 00:47:01.339
scientists beacons instrument in the
uh...
00:47:01.339 --> 00:47:02.649
not
00:47:02.649 --> 00:47:07.939
and pointing up to the skies above are
powerful new telescopes connecting the
00:47:07.939 --> 00:47:14.939
universe
00:47:19.389 --> 00:47:21.570
what's happening in last commonly
00:47:21.570 --> 00:47:28.570
east difficult how profoundly different
torrent update dot is transforming side
00:47:29.059 --> 00:47:34.319
astronomers on our interesting man in
june mysteries of the cost must
00:47:34.319 --> 00:47:36.999
compliance peacekeeping net
00:47:36.999 --> 00:47:43.999
all these new delhi
00:47:49.259 --> 00:47:52.329
because he's a very big place things got
billions of stars in it
00:47:52.329 --> 00:47:56.559
and so to put together a coherent
picture of the whole galaxy requires
00:47:56.559 --> 00:47:58.569
cutting enormous amount of data
00:47:58.569 --> 00:48:02.669
and before you could you afford sky
survey with sensitive diesel detectors
00:48:02.669 --> 00:48:07.009
that meant that you could matt many many
stars all once is very difficult to
00:48:07.009 --> 00:48:14.009
build up enough data on the for the past
00:48:14.049 --> 00:48:19.170
in the past launch surveys of the night
sky had to be done by exposing thousands
00:48:19.170 --> 00:48:21.509
of large photographic dates
00:48:21.509 --> 00:48:28.509
off these thirties scoop takes twenty
five years or more to compete
00:48:28.799 --> 00:48:33.469
then in the nineteen nineties came
digital a strong
00:48:33.469 --> 00:48:38.979
region increasing both moms and the
accessibility affairs
00:48:38.979 --> 00:48:43.019
the slow in the sky survey is the
world's biggest yes
00:48:43.019 --> 00:48:47.889
using a massive digital sensor mounted
on the back of a custom-built
00:48:47.889 --> 00:48:50.119
telescoping new mexico
00:48:50.119 --> 00:48:54.280
it stand the sky indictments tonight
forty two years
00:48:54.280 --> 00:48:58.979
building up a complicit picture in
unprecedented resolutions
00:48:58.979 --> 00:49:03.459
sort of some of the best steepest survey
data that we have in stock
00:49:03.459 --> 00:49:10.459
those on our own galaxy anand until he
sees further away from us
00:49:13.359 --> 00:49:16.429
school the slow moving test is on the
internet
00:49:16.429 --> 00:49:22.029
and we did astronomers have identifying
millions of users to unknown stars and
00:49:22.029 --> 00:49:23.289
galaxies
00:49:23.289 --> 00:49:25.520
you want to call on the fifteenth of
these
00:49:25.520 --> 00:49:32.089
for statistical actions which we proved
disprove or even suggested new theories
00:49:32.089 --> 00:49:37.049
so we have decided it galaxies grounding
becomes florence county cities like the
00:49:37.049 --> 00:49:38.879
one we live in a teammate
00:49:38.879 --> 00:49:44.329
uh... not all at once we're not smoothly
specter five continuously
00:49:44.329 --> 00:49:46.629
incorporating beasley cannibalizing
00:49:46.629 --> 00:49:50.499
of smaller galaxies be dissolved and
then they become part of under the
00:49:50.499 --> 00:49:55.469
guidance systems
00:49:55.469 --> 00:50:01.689
is a startling idea and you still need
to is the evidence to support
00:50:01.689 --> 00:50:06.469
groups of stars that came from
cannibalized galaxies stunned often
00:50:06.469 --> 00:50:07.799
assume they kept
00:50:07.799 --> 00:50:13.479
artistically different prabhat the stars
because they move acted different
00:50:13.479 --> 00:50:15.219
each day despite
00:50:15.219 --> 00:50:17.719
when one of these distribution graphs
00:50:17.719 --> 00:50:19.490
means professional policy
00:50:19.490 --> 00:50:24.459
has found a group of stars or travelling
in a different way to the list
00:50:24.459 --> 00:50:29.689
error that tell-tale patterns she's
looking for
00:50:29.689 --> 00:50:32.039
the evidence is accumulating
00:50:32.039 --> 00:50:36.449
uh... this really is how galaxies girl
born in portland int l tells his ground
00:50:36.449 --> 00:50:40.349
and so this is an important part of of
understanding how galaxies form not only
00:50:40.349 --> 00:50:45.679
are posted every gals
00:50:45.679 --> 00:50:49.449
more database the more discoveries can
be made
00:50:49.449 --> 00:50:52.390
and the technology is getting better all
the
00:50:52.390 --> 00:50:56.549
the next big survey telescope starts at
seventeen twenty fifty
00:50:56.549 --> 00:50:59.939
do you still need docs
00:50:59.939 --> 00:51:06.939
slowed mistaken eighty years still
covered one quarter of the night sky
00:51:06.989 --> 00:51:08.250
the new tenants
00:51:08.250 --> 00:51:10.899
with gandhi and tires scott
00:51:10.899 --> 00:51:12.909
in even greater to resolution
00:51:12.909 --> 00:51:19.909
every three days
00:51:23.589 --> 00:51:26.559
the foster mom something still we have
today
00:51:26.559 --> 00:51:30.389
allows researchers inform sorts of
fields
00:51:30.389 --> 00:51:33.150
test this theory is on the previously i
mean
00:51:33.150 --> 00:51:35.619
mentioned in this case
00:51:35.619 --> 00:51:39.150
but more than this it's many even change
that
00:51:39.150 --> 00:51:42.639
condiment in ways insist
00:51:42.639 --> 00:51:45.350
with the power of today's computers up
00:51:45.350 --> 00:51:47.659
plied to all these things too
00:51:47.659 --> 00:51:54.659
and she was might even able to try to
convince
00:52:04.130 --> 00:52:05.840
potentially profoundly in
00:52:05.840 --> 00:52:06.799
portant
00:52:06.799 --> 00:52:10.169
potential and the significance
00:52:10.169 --> 00:52:13.279
princeton science person even the most
exciting
00:52:13.279 --> 00:52:14.979
waste
00:52:14.979 --> 00:52:19.329
the potential to transform officer
powell scientist do science but even
00:52:19.329 --> 00:52:23.989
more sciences possible on what will come
with a constellation court
00:52:23.989 --> 00:52:28.049
both house and she's done for the war
science as possible discreetly
00:52:28.049 --> 00:52:31.939
compensation
00:52:31.939 --> 00:52:36.609
many of the dynamics of the natural
world nike interplay between the rain
00:52:36.609 --> 00:52:41.109
forest and the atmosphere are still
complex but we don't answer the agreed
00:52:41.109 --> 00:52:42.769
on the staff
00:52:42.769 --> 00:52:47.299
now computers are generating literally
tens of thousands of different
00:52:47.299 --> 00:52:53.349
simulations the hoa fees biological
systems might work dislike creating
00:52:53.349 --> 00:52:58.579
thousands of a hypothetical part of the
world's each and every one of the
00:52:58.579 --> 00:53:04.249
simulations his analyzed with statistics
to see if any part of good match will
00:53:04.249 --> 00:53:07.209
report is observed in nature
00:53:07.209 --> 00:53:11.059
the computers can now put domestic
degenerate
00:53:11.059 --> 00:53:13.059
and discard hypothesis
00:53:13.059 --> 00:53:17.679
with scarcely humans
00:53:17.679 --> 00:53:21.799
east new application six six
00:53:21.799 --> 00:53:24.409
often divided for future science
00:53:24.409 --> 00:53:26.279
is creating a new
00:53:26.279 --> 00:53:28.709
paradigm if you like
00:53:28.709 --> 00:53:31.889
in science midway between conducive
names
00:53:31.889 --> 00:53:34.599
which is increasingly
00:53:34.599 --> 00:53:37.509
between my country rises
00:53:37.509 --> 00:53:40.219
data centric or data driven
00:53:40.219 --> 00:53:41.190
ryland being
00:53:41.190 --> 00:53:43.979
hypothesis driven or experiment retreat
00:53:43.979 --> 00:53:47.960
so is exciting times in terms of the
science in terms of that
00:53:47.960 --> 00:53:49.829
computations
00:53:49.829 --> 00:53:56.829
and in terms of the statistics
00:53:58.319 --> 00:54:02.069
now we thought that sounds a bit
abstract inferior to continue
00:54:02.069 --> 00:54:04.719
how about one final frontier
00:54:04.719 --> 00:54:10.609
good statistics even make sense of your
feelings
00:54:10.609 --> 00:54:11.859
in california
00:54:11.859 --> 00:54:17.179
wariness one computer scientist is
harvesting the internet to try to do you
00:54:17.179 --> 00:54:24.179
find the patterns alvaro innermost
thoughts and emotions
00:54:33.960 --> 00:54:36.620
well this is the madness with madison
00:54:36.620 --> 00:54:40.259
represents a skyscraper views the world
00:54:40.259 --> 00:54:42.060
each of these brightly colored dots
00:54:42.060 --> 00:54:43.979
is an individual feelings
00:54:43.979 --> 00:54:45.779
expressed by someone out there
00:54:45.779 --> 00:54:47.989
blog artery
00:54:47.989 --> 00:54:50.630
and when you click on a diet explodes
00:54:50.630 --> 00:54:53.779
to have revealed the underlying feelings
doctors
00:54:53.779 --> 00:54:56.910
this is what people say they're feeling
00:54:56.910 --> 00:55:01.319
batter
00:55:01.319 --> 00:55:08.319
well
00:55:14.989 --> 00:55:18.230
so every minute we feel fine calls the
world's blogs
00:55:18.230 --> 00:55:22.399
case all the sentences that start with
the words aren't feeling feel
00:55:22.399 --> 00:55:25.229
interest them in the database
00:55:25.229 --> 00:55:27.009
we collect all the feelings
00:55:27.009 --> 00:55:29.199
and week at the most common
00:55:29.199 --> 00:55:36.199
they're better
00:55:37.819 --> 00:55:39.359
the same
00:55:39.359 --> 00:55:40.989
lecture
00:55:40.989 --> 00:55:47.729
sorry
00:55:47.729 --> 00:55:50.859
and we can take a look at anyone feeling
in analyzing
00:55:50.859 --> 00:55:54.109
right now a lot of people are feeling
happy
00:55:54.109 --> 00:55:56.599
we can take a look at all the people who
are happy
00:55:56.599 --> 00:56:00.369
and breakdown by age gender or location
00:56:00.369 --> 00:56:04.279
says bloggers have public profiles we
have that information and so we can ask
00:56:04.279 --> 00:56:05.609
questions like
00:56:05.609 --> 00:56:08.149
are women happier than men
00:56:08.149 --> 00:56:15.149
isn't happy with that the united states
00:56:19.189 --> 00:56:21.150
we find that as people get older
00:56:21.150 --> 00:56:23.579
make you happier
00:56:23.579 --> 00:56:25.400
and moreover we find that
00:56:25.400 --> 00:56:27.209
for younger people
00:56:27.209 --> 00:56:28.820
this as the happiness
00:56:28.820 --> 00:56:30.899
more with exciting
00:56:30.899 --> 00:56:32.809
and as people get water
00:56:32.809 --> 00:56:39.809
this essay happiness more peaceful
00:56:40.619 --> 00:56:44.679
and we also find that women feel love
tomorrow
00:56:44.679 --> 00:56:46.969
also more guilty
00:56:46.969 --> 00:56:49.929
moments could more often than women
00:56:49.929 --> 00:56:56.049
about someone
00:56:56.049 --> 00:56:59.269
as people even more and more of their
lives on line
00:56:59.269 --> 00:57:01.749
really behind digital traces
00:57:01.749 --> 00:57:03.919
and with these digital traces
00:57:03.919 --> 00:57:07.210
became deeply into the statistic
re-energize
00:57:07.210 --> 00:57:14.210
what it means to be sure
00:57:40.689 --> 00:57:43.199
so where does all of this leave office
00:57:43.199 --> 00:57:47.889
we generate i mean mention of the
quantities of data about everything you
00:57:47.889 --> 00:57:49.289
can think of
00:57:49.289 --> 00:57:52.099
analyzing to review the at it's
00:57:52.099 --> 00:57:53.239
amount
00:57:53.239 --> 00:57:56.259
not only expert but all of us
00:57:56.259 --> 00:58:03.259
understand the stories in the mail
00:58:07.139 --> 00:58:12.619
instead of being led astray by prejudice
with statistics at our fingertips
00:58:12.619 --> 00:58:18.039
alliance can be opened for a fact based
you whatever
00:58:18.039 --> 00:58:23.479
more than every before we can become all
sorts of are all the best it's pretty
00:58:23.479 --> 00:58:30.479
exciting isn't it
00:58:30.759 --> 00:58:31.939
stay here
00:58:31.939 --> 00:58:38.939
the with
00:58:40.929 --> 00:58:47.929
he indeed
00:58:53.599 --> 00:58:53.979
seyi