Tip:
Highlight text to annotate it
X
REIHAN SALAM: OK.
Urs Gasser.
URS GASSER: So first of all, let me clarify neither Vint
Cerf nor I are Boston city officials,
despite wearing a tie.
[LAUGHTER]
VINT CERF: I'm sorry.
It's all the clothes I got.
It's just the way it is.
URS GASSER: I'm delighted to be here.
Thank you so much for the invitation.
It's a fantastic, very rich day.
What do we need for a responsible
and innovative web?
Cover that in half an hour?
Well, we'll see.
I would like to share three quick observations, hopefully
building up on some of the points that we touched upon
throughout the day, and those are with an eye towards
policymaking.
My first point is as users, parents, teachers, corporate
decision-makers, policymakers, we need a balanced
understanding and view on both the opportunities and
challenges associated with digital technologies,
especially when it comes to youth online.
Today's discussion was fantastic and a very balanced
exploration of both risks as well as these tremendous
promises of the internet.
In my conversations with parents and teachers, as well
as policymakers sometimes, the debate is much more one-sided.
To make things worse, it's often fueled and driven by
fear and headlines rather than by empirical
evidence and data.
For the future of the web and for responsible and innovative
policymaking for an innovative web, we really need to
emphasize this point: that it is important to craft policies
that are driven by data.
Amanda Lenhart was on this panel earlier today.
She and her team are doing tremendous work, for instance,
when it comes to youth policymaking, where the data
they gather can really inform us.
Of course, sometimes you have to act under conditions of
uncertainty.
We acknowledge also today there are certain things we
don't know, and we still have to make decisions, whether as
consumers or as policymakers.
In such situations, we really need to become more
innovative, how we incorporate mechanisms of learning into
our decisions and into our organizations
and into our policies.
We often don't get it right at the first chance.
How do we ensure learning?
The second point I would like to make it is more about the
approaches or the solution space.
We've identified a long list of challenges and potential
problems and risks earlier today, ranging from privacy,
security, safety issues to concerns of information
quality, gaps in skills and literacy.
I think it's a fair summary to say one of the consensuses
that emerged today is that there is certainly no
silver-bullet solution to any of these hugely complex
phenomenon challenges.
As we move forward, I would argue much of the success of
the use of tools as we explored them today a little
bit, whether it's educational interventions, whether it's
laws and regulation, whether it's reputation systems-- you
will talk more about that--
or other institutions that we use to deal with some of these
challenges, we need to better understand the tools
themselves and how they play out in the digital
environment.
We talked a little bit about the question of how certain
solutions scale, especially when dealing with
huge amounts of data.
The example of YouTube this morning was very, very
illustrative, where you see that this amount of videos at
some point need to be reviewed by human beings.
The algorithms are good at flagging and prioritizing but
not addressing the core issue.
There you see this kind of bottleneck situation.
Moreover, still in this category of approaches, we
need not only to understand which tool in the tool box is
useful to address which sort of problem that you face
online, but also to better understand the interplay among
the different instruments that we have available and that we
talked about today.
In the same cluster, just two quick notes.
One, on education.
I'm thrilled that we spent so much time and energy today in
discussing the promise of education as a potential
strategy to cope with some of the thorny challenges.
As I said, we didn't only talk about challenges
but also the promises.
Education is really important.
At the Berkman Center, for instance, we have currently an
experiment on the way.
That's what I like about education, you can also run
experiments and learn from them.
The experiment that we call the Youth and Media Lab, where
we bring together high school students and college students
with experts in the field on internet policy and create a
research and development lab for policymaking as well as
critical development and tool development.
We've heard many other examples from Esther and Nancy
on this power of education.
To see it as a potential of innovation is very exciting.
The second quick remark in this category is about law.
Law, of course, is also a tool.
David Drummond earlier this morning started with a
cautionary note.
Don't jump immediately to law and think law will solve all
the problems, such as cyberbullying.
Of course the world is much more complex.
But one point seems important.
We should not mischaracterize law.
Law is not only a constraint on behavior.
Law can also be an enabler.
Many of the tools and platforms and applications
we've seen today have been enabled by laws, some of them
internet-specific laws, fostering innovation and
protecting innovation.
That's another core pillar, I think, as we move forward, to
think carefully about such laws as an infrastructure for
a responsible and innovative web in the future.
Last point quickly--
and that's the third statement or observation--
we should be increasing
interoperability across all layers.
Of course interoperability is a concept many of you in this
room here are deeply familiar with at the technological and
data layers.
Many of the opportunities characterized today, and the
web as such as we presented it obviously, is all about
interoperability.
The social web is about the flow of data across systems
and the meaningful exchange of information.
Not only that level of interoperability is important
as we move forward, but I would argue it's the
organizational interoperability, the working
together of different stakeholders to address some
of the challenges and harness the opportunities.
Parents, teachers, schools, private companies such as
Google and many others need to work hand-in-hand to create
interfaces of collaboration among institutions.
Similarly, at the top layer, we also need to think about
legal and policy interoperability.
We talked a little bit about legislation, the context of
cyberbullying, where you have different state laws that are
not interoperable, putting companies such as Facebook in
a tricky position if they want to comply with all the laws.
International, they mentioned again, is to be acknowledged
where privacy laws-- that was mentioned as well-- create all
sorts of interoperability problems for
global internet companies.
These are my three things.
REIHAN SALAM: Thanks very much, Urs.
And Daniel Kent of Net Literacy, would
you like to go next?
DANIEL KENT: Sure.
I've prepared some remarks just so that I'm cognizant of
the five-minute remark.
REIHAN SALAM: Terrific.
DANIEL KENT: I'm honored to be here with two of the most
forward-thinking internet visionaries.
I'd like to add my congratulations to Google for
convening this discussion.
While these gentlemen were busy inventing and crafting
the infrastructure and shaping the policy that would come to
define the internet, I was growing up as an end user.
I'm a little bit embarrassed to say that it took me until
the fourth grade to open up my beige box and upgrade my RAM
to what I think was 256.
That issue of confidence is still a large barrier to
individuals around the world in adopting broadband and
computers and access devices and learning how to navigate
through the net.
As a digital literacy practitioner, it was
especially poignant when senior citizens told me that
they felt that they were too old to learn.
Sort of switching gears, organizations that provide
digital inclusion data like Pew and the International
Telecommunication Union have highlighted that digital
inclusion is a serious barrier to computer and broadband use.
Today those that are offline are being left behind.
Tomorrow those without access to computers and the internet
will be in a very sense disabled and unable to access
the increasing number of services that are migrating to
the net and not available to those offline.
Those that are disconnected in North America and abroad will
become poorer, detached from an increasing torrent of rich
information, and less able to compete in school, in the
workforce, and in life.
As we progress into the 21st century, those that are not
net savvy will be increasingly relegated to an underclass
that may be subject to an ugliness of a new form of
discrimination.
Four years ago, I was a teenager and I can tell you
from very recent experience that to my generation, Web 2.0
and social media is a double-edged sword.
As Dora said earlier this morning, while it has
empowered us, it has also created new responsibilities.
It's social to tweet and plus 1 or like pages and posts and
photos, but unfortunately, members of the digital
generation sometimes make poor decisions and post
inappropriate pictures or comments.
Five years ago, few fully realized the impact of their
future net reputations.
But today, scholarship programs, colleges, and
employers are regularly googling applicants.
It's been recently reported in the news that some employers
are now asking for their employees to turn over their
Facebook logins and passwords.
Without addressing any of the privacy and other issues, it's
clear that what we're doing online today will definitely
impact our future well-being.
Users need to become more aware and better educated
about their net reputation.
I commend Google for creating Me on the Web, a web
application that integrates with your
current Google account.
It helps users manage and control what others can find
about them on the web and know what is published about them
online, not only what you publish but also what others
publish about you.
Google and the Berkman Center's initiatives to inform
educators and innovators and policymakers and end users is
truly the way to build the character of our community of
digital citizens.
As we discuss what's on the horizon on this panel, I'll
try to have my comments reflect a digital inclusion
practitioner's prospective.
I'm mindful that without digital inclusion and
literacy, some of us will not enjoy the benefits of a
responsible and innovative web.
REIHAN SALAM: Thanks very much, Daniel.
And Mr. Cerf, with the home team, would you'd like to
share with us your thoughts?
VINT CERF: I didn't know we were in a competition.
This was to be a cooperative view of what's
actually going on.
Let me start out by making a couple of observations.
I want to focus on
responsibility for just a moment.
Among the various things we enjoy on the internet, at
least many of us, is the freedom to speak and to hear
and to interact and to share and to
collaborate and cooperate.
But I think there's a freedom that is often not articulated,
and that's freedom from harm.
We are not free from harm.
We know that the internet is a place where harms can occur.
It's a very complicated environment because the
perpetrator of a harm may be in one jurisdiction or one
country and the victim might be in another.
This gets to your interoperability question,
where it's not clear that jurisdictional boundaries and,
let's say, reciprocity is available.
We have a global problem when it comes to coping with
improving safety on the internet.
I've often concluded that there are only three kinds of
things that you can do in order to cope
with the safety problem.
One, you can create technical means to prevent the harm from
happening: smarter operating systems, paranoid browsers
that don't let themselves get infected, and so on.
But frequently the technical solution doesn't work here.
It's not adequate.
The next thing you do is you try to figure out how to
detect that something bad has happened and
figure out who did it.
This is sort of detection and punishment.
That's what we do when it comes to drunk driving.
We can't stop you, but we pass laws saying if we catch you,
there will be consequences.
That's the second kind of response that you can have.
The third one is moral assuasion.
I don't mean to suggest that's a weak response.
In some cases, saying don't do that, it's bad, it's harmful,
it hurts, what if they did it to you is
a pretty good argument.
Those are three elements that I think are very important.
I think also what you've been hearing today on the part that
I was able to participate in underscores another really
important point: the power and importance of infrastructure.
That's really what a lot of this is about.
The internet and the World Wide Web and the apps that sit
on top of mobiles are a consequence of enabling
infrastructure.
I would not ever want to discount the power that that
kind of enabling infrastructure has.
I want to come back to the responsibility question,
though, because regardless of those three mechanisms that I
mentioned, taking responsibility for the way we
behave on the network is a very important thing.
What we don't have, though, on a global scale is a sense of
what the social norm should be in this online environment.
It's come up several times in the earlier conversations.
I don't think that we can make this up.
I think we are literally going to end up having to live
through this sort of thing.
Let me give you a trivial example.
When we're walking towards each other down the street,
it's generally common that we try to figure out how not to
run into each other.
It's just socially more convenient that way than
smashing into each other and having an argument over who
gets out of the way first.
That's a trivial example of a social norm, but it's the kind
of thing that we need to find in this cyberspace
environment.
The people who are engaged that are going to have to
figure out what that is--
Parry Aftab has the teenangels activity that's been going on
for 14 years now.
These are kids trying to help other kids to figure out
what's the right kind of behavior, what
makes sense to them.
They are, those kids, are going to be a lot more
credible having opinions and suggesting ideas than any
adult or any other authority figure is likely to have.
Let me give you another example of participatory
involvement in trying to create a safer environment.
Google and Paypal and Qualys and other several companies
have gotten together to form something called StopBadware,
which was originally started in the Berkman Center and then
was exported out as a not-for-profit.
What it does is respond to the problem of infected websites.
When Google does an index of the World Wide Web, we have a
crawler that's trying to look at every single web page and
makes notes about the ones that it
thinks might be infected.
We download the web pages and a piece of
software looks at it.
If it thinks that there is a piece of malware on the
website, it makes a little note saying this
site might be dangerous.
When somebody does a Google search and one of the results
is a link to one of those marked sites, if you click on
the link, before we let you go there, we pop up this
interstitial web page.
It's bright red.
It says don't go there; there might be malware on the site.
If you happen to be the site owner, and somebody comes you
and says ha, Google says you're infected.
Of course, most of these people don't know that.
They think they've been insulted.
They make all kinds of noise and say, I never did anything.
Of course, they probably didn't.
They probably had a website that
wasn't adequately secured.
Somebody else infected it for them.
We send them to StopBadware.
StopBadware goes through the website with a fine-toothed
comb and helps them clean it out.
That's an example of a kind of responsible behavior in this
environment, which we know is not purely safe, not
absolutely secure.
This is a continuous process.
It's not something you just do once.
It's like brushing your teeth.
If you brush your teeth once in your life, it doesn't do
you any good.
You have to brush them every day.
It's like cyber hygiene, same thing.
I realize I don't want to go on and on and on.
Let me pick one another point, then get into some discussion.
Again this is responsibility.
Personal responsibility for us in my view
includes critical thinking.
What in heck am I talking about?
How many times have you gotten a message from somebody that
says, the post office is going to start charging a penny for
every email.
Go look it up on Snopes first. Take some responsibility
before you propagate nonsense.
It's true that there's misinformation on the net, and
we're never going to get rid of that.
But one thing we can do is be a little bit more thoughtful
about deciding whether we should believe this
information or not.
That's called critical thinking.
One of the best lessons that we can teach our kids is how
to think critically about the information that they get.
They're not going to get
misinformation just on the net.
They're going to get it from movies, television, their
friends, their parents might be misinformed.
There are a variety of sources of misinformation.
Kids need to learn how to sort that out as best they can.
If we teach them how to do that, then they will be able
to defend against misinformation in all of its
forms, not just on the internet.
Mr. Chairman, I have one other anecdote to tell you.
This is Kissinger's complaint.
I had lunch with Henry Kissinger
several years ago now.
The first thing he said was he hated the internet.
I thought, well, that's a great way to start out lunch.
The reason he said he didn't like it was that people had
become satisfied with too little information.
A snippet will do.
You understand that Dr. Kissinger
writes 700-page books.
He wants people to read all 700 pages of the books.
He doesn't want them to be satisfied with a paragraph.
So he was unhappy about that.
On the other hand, the thing that he liked was Google News,
because he could get a sense for what everybody was talking
about in a very rapid way.
But he did point out that he's very upset that his
grandchildren do not know how to read cursive writing
because the only thing they ever see is print.
All those letters that he has from famous people in history,
the kids, his grandchildren, can't read them, because they
don't read cursive.
I told him I didn't know what to do about that.
That's his problem.
Mr. Chairman, I turn this over to you, and thank you for the
opportunity to--
REIHAN SALAM: Thank you
VINT CERF: --hold forth this way.
REIHAN SALAM: Urs, I wonder, when Vint was talking about
freedom from harm, the third element he mentioned was this
idea of moral assuasion.
You mention very briefly during your remarks reputation
systems. I wonder if you have any thoughts about reputation
systems and whether we might see the emergence of global
reputation systems that could help us maintain a more
responsible web and also serve to help people and users
navigate the web in a different kind of way?
URS GASSER: Great question.
I truly believe that a reputation system as a
governance mechanism has tremendous potential.
But at the same time at least from what I've seen, which is
of course only part of the discussion, it's extremely
hard to think about the design of such systems because you
run into all sorts of issues.
VINT CERF: Well, what if somebody lies?
I mean, what if you tell everybody that your competitor
makes bad material and it isn't true and now we're back
to critical thinking.
URS GASSER: I've seen it with dentists and so forth.
And in the professional space with teachers as well, what
are appeal mechanisms, correction mechanisms. But
then perhaps more importantly also, thinking about again the
global context-- and Jill could obviously comment on
that much better--
for reputation.
You need some sort of identification.
That's not always desirable that you have full
identification on the web, and that every action, every
expression, every comment can be attributed to
a particular speaker.
There are mechanisms in computer science where you can
have persistent identities, build reputation systems but
not lose anonymity.
That's a hugely complex as a design challenge.
VINT CERF: Can we keep going on this just a little bit?
First of all, I think
anonymity is so very important.
We all understand there are situations where non-anonymous
speech is fatal.
So we need to preserve in this environment the ability to
speak pseudonymously.
Sometimes it's important to be able to speak in an
authoritative way with your name and make sure other
people can believe that it's you and not somebody else
pretending to be you.
On this side of reputation systems, though, it is
possible in this space to have brands that do not necessarily
identify an individual but which by their repeated value,
their content will establish brand and therefore establish
reputation.
There are people on the net who make comments on blogs.
I don't know who they are.
They have a handle.
But if the handles tend to be thoughtful, then you want to
give more credibility to that.
I am always nervous about the mechanics of reputation
systems for the reasons that you described.
Any engineer who's trying to design something needs to
think not only how is it going to work, but how is it going
to be abused, and how do I deal with that?
REIHAN SALAM: One thing that strikes me, and this is a
theme that's come up in a number of our panels so far,
is this idea that many of the problems that arise in the
real world arise in the web as well.
For example, when we're talking about reputation
systems, credit scores don't work terribly well.
But it's a fairly lawless thing where you have several
companies that provide them.
It's something that's emerged over time.
It's kludgy.
So when you think rigorously about it, people are providing
this public good, but in a for-profit context.
It's just out there, and we use it in an imperfect way.
That actually brings me to a remark that
Vint had made earlier.
You talked about much of what Google does in the context of
identifying sites with malware.
That work could be understood as a kind of provision of a
public good.
Yet, of course, Google is an organization that is
constrained.
It's also an organization that has been very much shaped by
the recent retreat of many websites from the open web to
the closed web.
And I wonder how you think about that evolution and the
implications of that for a generative
and responsible web?
If you're a sheriff of this terrain that is shrinking, as
more and more of the web is closed, that presumably
introduces all kinds of dilemmas.
VINT CERF: Can I make two observations?
First of all, we've had this happen before.
AOL, for example, started out as a very much closed system.
It was forced by the users to go outward.
My honest belief here is that the pendulum may be swinging.
If the users find that a closed environment gives them
a poorer-quality experience--
the limitation in the amount of information we have
available, more difficulty in finding things that are
relevant, or ability to see across a broader spectrum of
content, then they're going to push back.
That's one possibility.
The second thing is that, to bring back the responsibility
point, if companies like Google can't see those things
and therefore we can help protect people against, that
has an opportunity for someone to design systems that go in
the browser that are a lot more paranoid
about what they download.
The browsers will see the content
that's inside that wall.
They now become the operative opportunity for detecting that
there's something wrong.
That means better browser design than we have today.
That's why we've done Chrome.
REIHAN SALAM: Urs, I wonder, because you're at this
intersection of the academic world and also working on
public policy questions, I want to press a little bit
further on that idea of the public goods provider.
When that public goods provider is a for-profit firm
that faces its own larger strategic context, what do you
think about what other civil society actors can do to help
buttress that public goods role?
Or perhaps see to it that that role is carried out when this
firm or that firm fades from the scene?
URS GASSER: It goes a little bit back to my
interoperability point.
I truly believe that some of the challenges we identified
and addressed, as well as the possibilities, we can only
deal with them successfully if we work together.
Of course that includes--
that working together--
includes the development of a strategy of working together
and public policy discussions about checks and balances.
I share your concern that of course it's troublesome.
Some pieces of core infrastructure, and Vint made
this excellent point, how important infrastructure is.
It could not only be the fiber optics or the cables in the
ground, but also search and platforms and communication.
If that core infrastructure is in the hands of private
companies, we certainly have an institutional challenge.
We need to have a conversation as a society, how do we build
in checks and balances?
How do we reallocate mechanisms of control and
supervision?
These are tricky questions, and we don't have
good answers yet.
REIHAN SALAM: Daniel, I wonder given that your mission is to
increase net literacy and to bring people and digital
inclusion, it occurs to me that there is a sense in which
having a more cognizable, accessible space that is
moving away from an anarchic web to a web that consists of
walled gardens, might actually be advantageous from your
perspective.
Does that make sense to you?
DANIEL KENT: No, it really does.
One thing to take in consideration is it's probably
going to be a balance going forward.
One thing that we do is use a lot of
education for the end user.
In that respect, net literacy is a student-volunteer,
completely student-organized and student-run nonprofit.
What we do is teach how individuals should be safe
online, and how to navigate the web in
a responsible manner.
Working on both fronts is really the key to addressing
the situation.
There are so many stakeholders.
There probably won't be one, true solution.
REIHAN SALAM: Vint, I wonder there is this--
and this is taking off from Daniel's observation.
There is this great anxiety that the so-called
appliancization of the web is changing the way users think
about how they use the web.
You express this confidence, it seemed to me, that
eventually users might decide, wait a second, we don't like
these walled gardens.
We prefer a more open environment.
Do you fear that the way we use the web is shaping the way
we think about it?
What do we imagine to be the generative
potential of the web?
Do you worry that a new generation of web users who
are being weaned in the context of these walled
gardens might not demand a more open web?
That in fact, they might actually be trained to stay
between the lines, as it were?
VINT CERF: Possible.
Let me suggest a couple of metaphors here
that might be useful.
First of all, you hear about the World Wild Web
occasionally.
People think about the settling of the West. You have
the pioneers, the explorers and the pioneers come.
And they have to shoot the Indians or whatever it is in
order to survive.
There's all this conflict and lawlessness
and everything else.
When you finally settle, you find people
demanding more safety.
That's what governments are about.
Governments, in part, are supposed to be there to act in
the public interest to provide a safer
environment for everyone.
My belief here is that as we settle this environment, we'll
be looking for more certainty and predictability about what
things can happen to us, and what we can do about things
that we don't want to have happen.
It's conceivable that there will be some sense of safety
arising out of these walled environments.
But I have to tell you that when the World Wide Web first
showed up, there were two really important lessons that
are relevant to something you said earlier, and something
that I think is important to all of us.
The first one is the enormous amount of content that flowed
onto the net.
People had a way of expressing themselves.
They wanted to share what they knew.
This was not driven by a desire for remuneration.
It was driven by a desire that someone find what you knew to
be interesting and useful.
The second thing that's very important, which goes to your
point about how we learn to be safer on the net, do you
remember that when web pages first showed up, the browsers
allowed you then, and still allow you, to see how that web
page was constructed?
Show source or view source.
The webmasters taught each other.
You were allowed to copy some other webmaster's technique.
You could try out new ideas and other people
could learn from that.
What I hope, and I don't know how to do this yet, but what I
hope is that we have a similar way of making transparent what
people choose to do to make themselves feel safer and be
safer in the internet environment.
If we can find a way of making the viral learning happen,
that would be a good thing.
To come back to the primary question, I believe that there
is so much pent-up desire to share information with other
people that there will be resistance to
staying inside the walls.
The 60 hours a minute going into YouTube--
I don't know what the number is now, but that's what I've
heard recently--
per minute is evidence of this great continuing desire to
share what you know and share what you have. I don't see
walled gardens supplying you necessarily with that
capability.
REIHAN SALAM: Thanks so much you guys.
I'm afraid we're already over time.
I wonder perhaps we could address a couple of questions
before we wrap up?
VINT CERF: Did you say wrap up or crap out?
REIHAN SALAM: Wrap up.
VINT CERF: Wrap up.
REIHAN SALAM: Same thing.
AUDIENCE: Good afternoon.
My name is William Clements and thank you
for the robust panel.
Given that we see that the internet is just a reflection
of the people who use it, what are the best practices or
mechanisms that can be leveraged to increase
user-level accountability and responsibility without
sacrificing openness and creativity?
What are some of the best practice with that?
What we do now to implement that vision?
What is the vision of the ideal web in which we can be
innovative yet protected?
REIHAN SALAM: That is an extraordinarily broad question
you've asked, and I'm glad you asked it.
Daniel, you want to take a first crack at it?
DANIEL KENT: Sure.
It's about empowering the users and that through
education and through a bottom-up approach, users will
be able to grasp the concepts that they're not just one
individual in a morass of everything that's
going on in the net.
They're a citizen in a community.
They have a responsibility to do unto others as you'd want
others to do unto you.
REIHAN SALAM: Anyone else?
URS GASSER: I would share that view and argue it's a
concentric circle model, potentially.
It starts with every single individual user, this notion
of self responsibility and being a good
citizen on the web.
Then the social norm dynamics kicks in, in the family
context, with your friends, with your
online and offline friends.
It goes on and on.
At the outer circles, you have companies such as Facebook or
Google but also then
ultimately policy and lawmakers.
It's again a shared responsibility.
You start with the individual and the individual user has to
role to play here.
REIHAN SALAM: We have time for one more--
VINT CERF: Can I get in on that one?
I may wipe out the next question as a result.
I want to give an engineer's response to that question.
I'm talking to the-- if there any engineers in the audience
or who see this video later, listen.
You have a responsibility.
When you build systems, you need to think about how
they're going to be abused.
You have to think about how to build in better protection.
You need to give tools to people who want to be
responsible but don't necessarily have any time or
ability to go down into the details.
When you put automatic braking systems into the car, you do
that to help people be more responsible because they don't
necessarily know how to go pump the brakes whenever
they're skidding on ice.
The whole point here is that we don't give people enough
tools to be responsible citizens.
This business about changing your password every 30 seconds
and keeping a list of 700 of them so none of them are the
same, that's silly.
We really need to do a better job of providing tools that
let people be responsible without so much crazy effort
because they can't.
It's not possible.
The point here is that when a civil engineer gets a license
as a civil engineer and builds a bridge and the bridge
collapses, if it's demonstrable the civil
engineer's design was faulty, as opposed to the
implementation or the materials were faulty, he
bears liability.
I'm sure I've just scared every programmer in the
country by suggesting that there's liability for bugs in
the program.
But we should be thinking along those terms, not
necessarily taking legal liability.
Bugs in programs are going to be with us till
the cows come home.
Being more thoughtful and feeling responsibility for
dealing with that problem is absolutely essential.
REIHAN SALAM: We have time for one concise question.
URS GASSER: We have one question.
VINT CERF: There's one way over there, in the dark.
Would you like to come and be visible so the
camera can catch you?
AUDIENCE: Should I come forward?
REIHAN SALAM: Please do.
VINT CERF: Enter into the light.
AUDIENCE: So there's been a craving for fact checking on
the internet.
A lot of people see things which are obviously wrong.
Then they say, this is not the fact.
We need to make sure every blogger
writes the actual fact.
You can imagine situations where companies like Google,
as your type your blog post, will suggest things that are
facts to you in the future.
This is what Google is trying to do, trying to move towards
helping the user.
My question is, who decides what fact is?
Do you let these companies decide what fact is?
Or do you encourage users to write things are not factual,
but might be imaginative or creative, which will expand
user's understanding?
REIHAN SALAM: It occurs to me there are some nested
assumptions in that question.
Urs, would you like to field that question?
URS GASSER: I think it's--
REIHAN SALAM: Or, Vint?
VINT CERF: I would actually like some help, because in
spite of the fact that the question came into the light,
I couldn't hear all of it.
I can't lip read him from the dark.
You need to help me understand what the question was.
REIHAN SALAM: I believe that the question was about the
notion is that we ought to encourage facts rather than
the proliferation of false statements.
And that we might use technology somehow to
encourage the propagation of information that is true
rather than false.
VINT CERF: I wish that I had Harry Potter's magic wand to
make that happen.
I don't think I can.
In fact, if anything, responsible journalism is
evaporating on us.
We are getting, instead of reportage and fact, we're
getting opinion.
The Op Ed page has bled off into the news page.
If I've offended any newspaper journalists
here, it was on purpose.
I actually don't know of a technology that will
necessarily do anything.
But the reputation system notion that came up here feels
like that might be a place there.
Lots of eyes looking on something can react.
In the earlier comments that Tiffany made about the Kony
situation and the response of all those eyes looking at that
story and saying, wait a minute, that might turn out to
be the path to which we can deal with the problem you're
describing.
It's very clear that it's a big problem.
We have to do something.
Critical thinking is part of it.
Reacting if you know better, is important to share.
REIHAN SALAM: One thing I'll say to push back against that
slightly is that another way to think about
it is tension blindness.
There's this notion that you could say that
x or y isn't true.
Another way of saying it is that I'm looking at a
different angle on the situation.
Because I'm focusing on this particular aspect of a given
large, complex, unruly phenomenon, it's necessarily
going to be in tension with someone else's
characterization of it.
One thing that worries me is that when you have the idea of
a technological solution, the underlying premise is that
there is some true interpretation.
The trouble is that when you have a much more small-d
democratic way, you have far more people are able to offer
their narratives concerning some situation.
Then necessarily it's going to look as though
there's more chaos.
There are more things that are said that are not true.
But in fact another way of looking at it is that, we have
the wherewithal, the ability, to have more
angles on a given situation.
That is going to look chaotic.
You do need to be a responsible user to be able to
navigate that world.
But I wouldn't say that we're necessarily seeing a
diminution of what is true in terms of
our information diets.
VINT CERF: You're channeling Esther Dyson.
Esther has this wonderful quote.
She says, the antidote for bad information is more
information, not censorship--
not that I'm accusing you of censorship.
It's a really solid observation.
We have the responsibility to try to find out, as best we
can, what the real answers or what the real truth is.
If we don't do that, then we harm ourselves.
REIHAN SALAM: Everyone, we have a reception now at which
you can pepper these guys with even more questions.
Thank you so much for joining us.
This was a great fun for me.
I hope it was fun for you as well and informative, more to
the point, and not full of lies.
VINT CERF: I hope not, anyway.