WEBVTT

00:00:13.136 --> 00:00:15.244
Well good afternoon everybody.

00:00:15.244 --> 00:00:18.204
Thank you all very much for coming along today.

00:00:18.204 --> 00:00:21.901
I'd like to begin my talk with a story.

00:00:21.901 --> 00:00:24.453
It's a story that begins but does not end

00:00:24.453 --> 00:00:27.544
with a mathematician named Tim Gowers.

00:00:27.544 --> 00:00:30.945
Gowers is one of the world's
most renowned mathematicians,

00:00:30.945 --> 00:00:34.843
he is a professor at Cambridge University
and the recipient

00:00:34.843 --> 00:00:39.441
of the Fields Medal, often called
the Nobel Prize of Mathematics.

00:00:39.441 --> 00:00:42.833
Gowers is also a blogger

00:00:42.833 --> 00:00:46.536
and in January of 2009
he used his blog

00:00:46.536 --> 00:00:50.025
to pose a very striking question:

00:00:50.025 --> 00:00:55.353
Is massively collaborative mathematics possible?

00:00:55.353 --> 00:00:58.894
So what he was proposing in this post
was to use his

00:00:58.894 --> 00:01:03.139
blog to attack a difficult unsolved
mathematical problem,

00:01:03.139 --> 00:01:06.488
a problem which he said
he "would love to solve"

00:01:06.488 --> 00:01:09.492
completely in the open,
using his blog to post

00:01:09.492 --> 00:01:13.097
his ideas and his partial progress.

00:01:13.097 --> 00:01:16.545
What's more, he issued an open invitation

00:01:16.545 --> 00:01:18.832
inviting anybody in the world who thought that

00:01:18.832 --> 00:01:23.116
they had an idea to contribute to post their idea

00:01:23.116 --> 00:01:26.195
in the comment section of the blog.

00:01:26.195 --> 00:01:30.504
His hope was that by combining the ideas
of many minds

00:01:30.504 --> 00:01:34.161
he could make easy work
of his hard mathematical problem.

00:01:34.161 --> 00:01:38.280
He called this experiment the Polymath Project.

00:01:38.280 --> 00:01:40.315
Well, the Polymath Project got off to a slow start.

00:01:40.315 --> 00:01:43.587
The first 7 hours nobody posted any comments.

00:01:43.587 --> 00:01:46.995
But then, a mathematician from
the University of British Columbia

00:01:46.995 --> 00:01:50.317
named Joseph Somolosie
posted a short comment.

00:01:50.317 --> 00:01:52.604
And he seemed to break the ice
because a few minutes later

00:01:52.604 --> 00:01:56.330
a high school teacher
named Jason Dyer posted

00:01:56.330 --> 00:01:58.869
a suggestion.
And a few minutes after that

00:01:58.869 --> 00:02:01.080
another mathematician
named Terence Tao,

00:02:01.080 --> 00:02:04.329
also a "Fields" medalist,
posted an idea.

00:02:04.329 --> 00:02:08.051
And things really started to move
quickly at this point.

00:02:08.051 --> 00:02:12.792
Over the next 37 days
27 different people would post

00:02:12.792 --> 00:02:19.971
800 substantive comments
containing 170,000 words.

00:02:19.971 --> 00:02:23.542
I was not a serious participant
but I was following along closely from the start.

00:02:23.542 --> 00:02:25.764
And it was just amazing.

00:02:25.764 --> 00:02:27.998
The speed with which an idea
would be tentatively proposed

00:02:27.998 --> 00:02:31.017
and then really rapidly developed
by other people

00:02:31.017 --> 00:02:35.150
and improved sometimes discarded.
It's just amazing.

00:02:35.150 --> 00:02:37.833
Gowers described the process as being

00:02:37.833 --> 00:02:44.079
to ordinary research as driving is to pushing a car.

00:02:44.079 --> 00:02:47.614
At the end of the 37 days,
Gowers used his blog to

00:02:47.614 --> 00:02:51.355
announce that they had solved
the core problem, in fact,

00:02:51.355 --> 00:02:54.589
they had solved a harder
generalization of the problem.

00:02:54.589 --> 00:02:57.854
The Polymath Project had succeeded.

00:02:57.854 --> 00:03:00.824
So what the Polymath Project suggests,
at least to me,

00:03:00.824 --> 00:03:04.839
is that we can use the internet to build tools

00:03:04.839 --> 00:03:08.614
that actually expand our ability to solve

00:03:08.614 --> 00:03:12.731
the most challenging intellectual problems.

00:03:12.731 --> 00:03:16.525
Or to put it in another way,
we can build tools which actively

00:03:16.525 --> 00:03:20.680
amplify our collective intelligence
in much the same way

00:03:20.680 --> 00:03:24.321
as for millennia we've used
physical tools to amplify

00:03:24.321 --> 00:03:28.637
our strength. OK?

00:03:28.637 --> 00:03:31.307
So, what I'd like to talk about today,
what I'd like

00:03:31.307 --> 00:03:35.047
to explore today,
is what this means for science.

00:03:35.047 --> 00:03:38.300
It's much more important than just
solving a single mathematical problem.

00:03:38.300 --> 00:03:41.193
It means an expansion in the range of scientific

00:03:41.193 --> 00:03:44.392
problems we can hope to attack at all.

00:03:44.392 --> 00:03:47.945
It means potentially an acceleration
in the rate of scientific discovery.

00:03:47.945 --> 00:03:52.614
It means a change in the way
we construct knowledge itself.

00:03:52.614 --> 00:03:55.819
So, before I get too over-excited, however,

00:03:55.819 --> 00:03:58.568
I would like to talk about some of the challenges,

00:03:58.568 --> 00:03:59.782
some of the problems.

00:03:59.782 --> 00:04:03.964
Particularly, I'd like to describe
a failure of this approach.

00:04:03.964 --> 00:04:09.035
So it occurred in 2005, or started in 2005,
a grad student

00:04:09.035 --> 00:04:13.128
at Caltech, named John Stockton,
had a very good idea

00:04:13.128 --> 00:04:16.255
for what he called the "Quantum Wiki"

00:04:16.255 --> 00:04:20.729
or "Qwiki" for short. OK?
It's a great idea.

00:04:20.729 --> 00:04:23.360
What he did with the Qwiki, was --

00:04:23.360 --> 00:04:26.290
The idea of the Qwiki was that it was going to be

00:04:26.290 --> 00:04:28.138
a great repository of human knowledge.

00:04:28.138 --> 00:04:31.102
Much like Wikipedia.
But instead of being focused

00:04:31.102 --> 00:04:34.373
on general knowledge
it was gonna be focused on

00:04:34.373 --> 00:04:37.729
specialist knowledge
in quantum computing.

00:04:37.729 --> 00:04:40.420
It was gonna be a kind
of a super textbook for the field,

00:04:40.420 --> 00:04:42.971
with information about
all the latest research,

00:04:42.971 --> 00:04:46.017
about what the big open problems
in the field were,

00:04:46.017 --> 00:04:50.590
people's speculation about how to solve
the problems and so on.

00:04:50.590 --> 00:04:54.028
Like Wikipedia, the intention was
that it would be written

00:04:54.028 --> 00:04:58.929
by the users, in this case,
by experts in quantum computing.

00:04:58.929 --> 00:05:02.249
I was present at the conference
of Caltech in 2005

00:05:02.249 --> 00:05:05.225
when it was announced
and some of the people who I spoke to

00:05:05.225 --> 00:05:07.911
were very skeptical,
but some of the people

00:05:07.911 --> 00:05:09.846
were very excited about the idea.

00:05:09.846 --> 00:05:11.596
They were impressed
by the implementation,

00:05:11.596 --> 00:05:13.549
they were impressed
by the amount of initial

00:05:13.549 --> 00:05:15.764
seed material
which had been put on the site

00:05:15.764 --> 00:05:18.669
and most of all
they were excited by the vision.

00:05:18.669 --> 00:05:20.619
But just because they were excited

00:05:20.619 --> 00:05:23.971
didn't mean they wanted
to take the time themselves to contribute.

00:05:23.971 --> 00:05:27.132
They hoped that other people would do so.

00:05:27.132 --> 00:05:29.880
And in the end,
nobody, essentially,

00:05:29.880 --> 00:05:33.667
was really all that interested
in contributing.

00:05:33.667 --> 00:05:36.475
If you look today,
except in a few small corners,

00:05:36.475 --> 00:05:39.065
the Qwiki is essentially dead.

00:05:39.065 --> 00:05:41.720
And sad to say,
this is quite a common story.

00:05:41.720 --> 00:05:44.749
Many scientists,
in fields ranging from genetics

00:05:44.749 --> 00:05:47.619
to String Theory,
have tried to start science-wikis

00:05:47.619 --> 00:05:51.298
along very similar lines.
And typically they've failed,

00:05:51.298 --> 00:05:54.081
for essentially the same reason.

00:05:54.081 --> 00:05:55.852
It's not just science-wikis either.

00:05:55.852 --> 00:05:59.080
Inspired by Facebook,
many organizations have tried

00:05:59.080 --> 00:06:02.515
to create social networks
for scientists which will

00:06:02.515 --> 00:06:07.351
connect scientists to other people
with similar interest.

00:06:07.351 --> 00:06:12.073
So they can share things like data
or code their ideas and so on.

00:06:12.073 --> 00:06:14.169
Again, it sounds like a good idea.

00:06:14.169 --> 00:06:16.539
But if you join one of these sites,

00:06:16.539 --> 00:06:19.342
you will quickly discover
that they are essentially empty.

00:06:19.342 --> 00:06:22.868
They are virtual ghost towns.

00:06:22.868 --> 00:06:25.230
So what's going on?
What's the problem here?

00:06:25.230 --> 00:06:27.866
Why are these promising sites failing?

00:06:27.866 --> 00:06:31.765
Well, imagine that you are
an ambitious young scientist.

00:06:31.765 --> 00:06:35.239
In fact, I know some of you here are
ambitious young scientists.

00:06:35.239 --> 00:06:37.925
Imagine you are
an ambitious young scientist.

00:06:37.925 --> 00:06:40.498
You would really like to get a good job,
a permanent job,

00:06:40.498 --> 00:06:44.025
a good job, doing the work that you love.

00:06:44.025 --> 00:06:46.081
But it's incredibly competitive to get such jobs.

00:06:46.081 --> 00:06:47.619
Often there will be hundreds

00:06:47.619 --> 00:06:52.325
of very highly qualified applicants for positions.

00:06:52.325 --> 00:06:56.442
And so you find yourself working,
60, 70, 80 hours a week

00:06:56.442 --> 00:07:00.950
doing the one thing that you know
will get you such a job.

00:07:00.950 --> 00:07:03.580
And that is writing scientific papers.

00:07:03.580 --> 00:07:07.128
You might think that the Qwiki
is a wonderful idea in principle,

00:07:07.128 --> 00:07:10.483
but you also know that writing
a single mediocre paper

00:07:10.483 --> 00:07:13.604
would yield much more
for your career in your job prospects

00:07:13.604 --> 00:07:17.735
than a long series of brilliant contributions
to such a site.

00:07:17.735 --> 00:07:19.391
So even though you may like the idea,
you may think

00:07:19.391 --> 00:07:23.475
that it will advance science more quickly,
you just can't

00:07:23.475 --> 00:07:28.060
conceive of it as being part of your job.
It's not.

00:07:28.060 --> 00:07:32.512
The only things which can succeed
in this kind of environment

00:07:32.512 --> 00:07:35.684
are projects like the Polymath Project,
which even though

00:07:35.684 --> 00:07:40.389
they employ an unconventional means to an end

00:07:40.389 --> 00:07:42.612
they have an essential conservatism about them.

00:07:42.612 --> 00:07:44.576
The end product of the Polymath Project

00:07:44.576 --> 00:07:46.762
was still a scientific paper.

00:07:46.762 --> 00:07:49.072
In fact, it was several papers.
Right?

00:07:49.072 --> 00:07:52.347
So unconventional means
but conventional ends.

00:07:52.347 --> 00:07:54.530
So there is a kind of conservatism about it.

00:07:54.530 --> 00:07:58.311
Don't get me wrong,
the Polymath Project is terrific

00:07:58.312 --> 00:08:01.196
but it is a pity that scientists can only

00:08:01.196 --> 00:08:05.707
use tools which have
this kind of conservative nature.

00:08:05.707 --> 00:08:07.742
So let me tell you a story about an instance

00:08:07.742 --> 00:08:10.860
where we moved away from this conservatism.

00:08:10.860 --> 00:08:14.443
So it's a rare story where the conservatism
has been broken.

00:08:14.443 --> 00:08:18.076
It occurred in the 1990s when,
as you know,

00:08:18.076 --> 00:08:21.069
for the first time biologists
were taking large amounts

00:08:21.069 --> 00:08:25.236
of genetic data to collect
in the Human Genome Project.

00:08:25.236 --> 00:08:28.705
And there were sites online
which would allow biologists

00:08:28.705 --> 00:08:31.790
to upload that data so it can be shared
with other people

00:08:31.790 --> 00:08:35.968
around the world
and analyzed by other people.

00:08:35.968 --> 00:08:38.716
Probably the best one of these
is the site GenBank

00:08:38.716 --> 00:08:42.130
which some of you may have heard of or used.

00:08:42.130 --> 00:08:45.576
And these sites, like GenBank,
had the problem in common

00:08:45.576 --> 00:08:52.424
with Qwiki that scientists, they're not paid
or rewarded for sharing their data.

00:08:52.424 --> 00:08:56.064
It's all about publishing papers.
So there was

00:08:56.064 --> 00:09:00.631
a considerable reluctance
to actually upload the data.

00:09:00.631 --> 00:09:03.561
Everybody could see
that this was silly but it was

00:09:03.561 --> 00:09:05.851
obvious that this was
the right thing to do.

00:09:05.851 --> 00:09:07.667
But just because this was obvious
didn't mean

00:09:07.667 --> 00:09:09.461
that people were actually doing it.

00:09:09.461 --> 00:09:12.829
So a meeting was convened
in Bermuda in 1996

00:09:12.829 --> 00:09:15.471
of many of the world's
leading molecular biologists.

00:09:15.471 --> 00:09:17.787
And they sat and they discussed
the problem

00:09:17.787 --> 00:09:20.083
for several days and they came up with,

00:09:20.083 --> 00:09:21.579
what are now called the Bermuda Principles,

00:09:21.579 --> 00:09:26.097
which state that:
first -- once human genetic data

00:09:26.097 --> 00:09:29.324
is taken in the lab,
it should be immediately uploaded

00:09:29.324 --> 00:09:31.880
to a site like Gene Bank,
and two --

00:09:31.880 --> 00:09:35.107
that the data would be in the public domain.

00:09:35.107 --> 00:09:36.874
And these principles were given teeth because

00:09:36.874 --> 00:09:39.884
they were taken by the big
scientific grant agencies,

00:09:39.884 --> 00:09:43.442
the US National Institutes of Health,
the UK Wellcome Trust

00:09:43.442 --> 00:09:47.471
actually baked into policy.

00:09:47.471 --> 00:09:50.483
So it meant that if you were a scientist

00:09:50.483 --> 00:09:52.307
who wanted to work on the Human Genome,

00:09:52.307 --> 00:09:54.321
you had to agree to abide by these principles,

00:09:54.321 --> 00:09:57.304
And today, I'm very pleased to say, as a result,

00:09:57.304 --> 00:10:00.133
you can go online -- anybody here -- and download

00:10:00.133 --> 00:10:03.920
the human genome.
So that's a terrific story.

00:10:03.920 --> 00:10:06.777
But the Human Genome
is just a tiny fraction

00:10:06.777 --> 00:10:10.192
of all scientific knowledge.
Right?

00:10:10.192 --> 00:10:12.896
Even just in other parts of genetics,

00:10:12.896 --> 00:10:15.438
there is so much knowledge
that is still locked up.

00:10:15.438 --> 00:10:18.045
I spoke with one bioinformatician who told me

00:10:18.045 --> 00:10:22.880
that he'd been "sitting on the genome
of an entire species

00:10:22.880 --> 00:10:27.830
for more than a year."
An entire species --

00:10:27.830 --> 00:10:31.473
And in other parts of science,
it is routine

00:10:31.473 --> 00:10:34.914
that scientists hoard their data,
they hoard the computer code

00:10:34.914 --> 00:10:37.053
that they write,
that could be useful potentially

00:10:37.053 --> 00:10:40.616
to other people,
they hoard their best ideas and they often

00:10:40.616 --> 00:10:42.554
hoard even the descriptions of the problems

00:10:42.554 --> 00:10:45.738
that they think are most interesting.

00:10:45.738 --> 00:10:49.184
And so what I and other people
in the Open Science Movement

00:10:49.184 --> 00:10:52.043
would like to do is
we'd like to change this situation.

00:10:52.043 --> 00:10:54.430
We would like to change
the culture of science

00:10:54.430 --> 00:10:57.269
so that scientists become
much more strongly motivated

00:10:57.269 --> 00:10:59.808
to share all of these
different kinds of knowledge.

00:10:59.808 --> 00:11:04.340
We want to change the values
of individual scientists

00:11:04.340 --> 00:11:08.285
so they start to see it
as part of their job to be sharing

00:11:08.285 --> 00:11:09.914
their data, to be sharing their code.

00:11:09.914 --> 00:11:15.290
To be sharing their best ideas
and their problems.

00:11:15.290 --> 00:11:20.622
So, if we can do this,
this kind of change in values,

00:11:20.622 --> 00:11:24.541
then we will indeed start
to see these individual scientists

00:11:24.541 --> 00:11:26.325
rewarded for doing these things.

00:11:26.325 --> 00:11:28.638
They will be incentives to do them.

00:11:28.638 --> 00:11:32.108
It's a difficult thing to do, however.

00:11:32.108 --> 00:11:37.472
We're talking about changing the culture
of entire large parts of science.

00:11:37.472 --> 00:11:40.799
But it has happened before once in history.

00:11:40.799 --> 00:11:43.528
Right back at the dawn of science,

00:11:43.528 --> 00:11:48.879
Galileo, 1609, he points his telescope
up at the sky

00:11:48.879 --> 00:11:52.745
towards Saturn, and he sees
for the first time in history

00:11:52.745 --> 00:11:55.432
what we now know are the rings of Saturn.

00:11:55.432 --> 00:11:57.557
Does he tell everybody in the world?
No.

00:11:57.557 --> 00:11:58.758
He doesn't do that.

00:11:58.758 --> 00:12:02.116
He writes down a description, privately,
and then he scrambles

00:12:02.116 --> 00:12:06.997
the letters in the description
into an anagram and he sends

00:12:06.997 --> 00:12:11.594
that anagram to several
of his astronomer rivals.

00:12:11.594 --> 00:12:18.410
And what this ensures is that
if they later make the same discovery,

00:12:18.410 --> 00:12:21.500
he can reveal the anagram and get the credit

00:12:21.500 --> 00:12:24.943
but in the meantime he hasn't given up
any knowledge at all.

00:12:24.943 --> 00:12:28.122
And I'm sad to say that he was
no uncommon at the time.

00:12:28.122 --> 00:12:35.366
Newton, Huygens, Hooke, Leonardo,
they all used similar devices.

00:12:35.366 --> 00:12:40.412
The printing press had been around
for 150 years by this time.

00:12:40.412 --> 00:12:43.978
And yet there was a great battle
in the 17th and 18th centuries

00:12:43.978 --> 00:12:47.503
to change the culture of science
so that it became

00:12:47.503 --> 00:12:50.377
expected that when a scientist
made a discovery

00:12:50.377 --> 00:12:54.041
they would would reveal it in a journal.

00:12:54.041 --> 00:12:57.544
And that's great. That change has happened.
Terrific!

00:12:57.544 --> 00:12:59.798
But today we have new technologies,

00:12:59.798 --> 00:13:04.077
new opportunities to share our knowledge
in new ways

00:13:04.077 --> 00:13:07.130
and the ability to create
tools that actually allow us

00:13:07.130 --> 00:13:11.759
to solve problems in entirely new ways.

00:13:11.759 --> 00:13:15.693
So we need to have a second
Open Science Revolution.

00:13:15.693 --> 00:13:18.958
It is my belief that any publicly funded science

00:13:18.958 --> 00:13:22.179
should be open science.

00:13:22.179 --> 00:13:25.163
How can we achieve this change?

00:13:25.163 --> 00:13:28.178
Well, if you are a scientist
-- and I know many of you

00:13:28.178 --> 00:13:30.855
are not scientists
-- but if you are a scientist,

00:13:30.855 --> 00:13:32.975
then there are things that you can do.

00:13:32.975 --> 00:13:36.694
You can get involved
in an open science project

00:13:36.694 --> 00:13:40.125
even if it's just for a small fraction
of your time.

00:13:40.125 --> 00:13:43.714
You can find forums online
where you can share

00:13:43.714 --> 00:13:45.966
your knowledge in new ways,
ways that allow

00:13:45.966 --> 00:13:49.809
other people to build on that knowledge.

00:13:49.809 --> 00:13:52.649
You can also,
if you are more ambitious,

00:13:52.649 --> 00:13:55.147
start an open science project of your own.

00:13:55.147 --> 00:13:56.794
If you're really bold you may wish
to experiment with

00:13:56.794 --> 00:14:00.148
entirely new ways of collaborating in much

00:14:00.148 --> 00:14:03.778
the same way as the Polymath Project did.

00:14:03.778 --> 00:14:06.130
But above all, what you should do,

00:14:06.130 --> 00:14:10.124
is be very generous in giving credit
to those of your colleagues

00:14:10.124 --> 00:14:14.888
who are practicing science in the open
and to promote their work.

00:14:14.888 --> 00:14:19.292
These only conservative
scientific values that look down

00:14:19.292 --> 00:14:22.644
on these activities --
the sharing of data, the blogging,

00:14:22.644 --> 00:14:25.662
or using the wikis and so on --

00:14:25.662 --> 00:14:29.056
you can reject those
conservative values and engage

00:14:29.056 --> 00:14:31.418
your scientific colleagues in conversation

00:14:31.418 --> 00:14:34.617
to promote the value of these
new ways of working,

00:14:34.617 --> 00:14:37.828
to emphasize that it takes
bravery to do these things

00:14:37.828 --> 00:14:39.591
particularly by young scientists.

00:14:39.591 --> 00:14:43.326
It's through such conversation
that the culture of science

00:14:43.326 --> 00:14:47.930
can be changed.
So if you are not a scientist,

00:14:47.930 --> 00:14:50.764
there are also things that you can do.

00:14:50.764 --> 00:14:54.514
My belief is that the single most important thing

00:14:54.514 --> 00:14:58.320
that we can do to give impetus
to open science,

00:14:58.320 --> 00:15:03.591
is to create a general awareness
amongst the population

00:15:03.591 --> 00:15:08.514
of the issue of open science
and of its critical importance.

00:15:08.514 --> 00:15:11.410
If there is that general awareness,
then the scientific

00:15:11.410 --> 00:15:15.446
community will inevitably find,
it will be dragged

00:15:15.446 --> 00:15:19.063
by the population at large
in the right direction.

00:15:19.063 --> 00:15:20.499
There are simple things you can do.

00:15:20.499 --> 00:15:22.375
You can talk to your friends and acquaintances

00:15:22.375 --> 00:15:24.263
who are scientists and just ask them

00:15:24.263 --> 00:15:27.145
what are they doing to work more openly.

00:15:27.145 --> 00:15:30.045
Or you can use your imagination
and your personal power

00:15:30.045 --> 00:15:32.629
to raise awareness in other ways.

00:15:32.629 --> 00:15:35.291
We're talking about changing
not just what scientists do

00:15:35.291 --> 00:15:37.095
but what grant agencies do,

00:15:37.095 --> 00:15:39.915
what universities do
and what governments do.

00:15:39.915 --> 00:15:43.774
And you can influence all of those things.

00:15:43.774 --> 00:15:48.460
Our society faces a fundamental question:

00:15:48.460 --> 00:15:51.599
What kinds of knowledge are we going to expect

00:15:51.599 --> 00:15:55.346
and incentivise our scientists to share?

00:15:55.346 --> 00:15:58.493
Will we continue as we have done in the past

00:15:58.493 --> 00:16:02.666
or will we embrace new kinds
of sharing which lead

00:16:02.666 --> 00:16:06.645
to new methods for solving problems
and an acceleration

00:16:06.645 --> 00:16:10.121
in the process of science
entirely across the board.

00:16:10.121 --> 00:16:13.911
My hope is that we will embrace open science

00:16:13.911 --> 00:16:16.196
and really seize this opportunity

00:16:16.196 --> 00:16:20.520
that we have to reinvent discovery itself.

00:16:20.520 --> 00:16:24.283
Thank you.
(Applause)