Tip:
Highlight text to annotate it
X
Mr. Jeffrey Veen is the founder of Smallbatch, Inc.
and is an internationally sought-after speaker,
author, and user-experience consultant.
He's the founding partner of Adaptive Path
and was the project lead for Measure Map,
a well-received web analytics tool acquired by Google.
He is a consultant involved in a lot of different designs including
Blogger, Typepad, Flicker, and the National Public Radio.
He also managed the look and feel
feel of Hotwired, Hotbot, Search Engine, and Lycos. com,
as well as others. In addition to lecturing and
writing on web design and development, Jeffrey has been active
with the World Wide Web Consortium, CSS Editorial Review Board,
and as an invited expert in electronic publishing.
He's also an original columnist for Webmonkey and the author
of the acclaimed the books The Art and Science of Web Design
and Hotwired Style: Principles for Building Smart Web Sites.
Please help me welcome Mr. Jeffrey Veen.
Thank you.
Have you been watching this?
This is pretty cool. You guys, this conference was the number one trend
this morning when I first logged on. Yeah, nice work, everybody. You were
doing better than both Sarah Palin and Tina Fey, which I think is pretty cool.
That says something. Alright, let's see, let's get some slides up here.
How many of you were watching the Olympics this summer?
It was really just phenomenal. I must have watched
the opening ceremonies in beautiful HD four or five times. I was just
overwhelmed by how gorgeous it was. These guys in particular, right,
there were 2008 drummers on the floor
of the big Bird's Nest Stadium that they had there.
And they're all beating in time to some music and while they're doing this,
every time they hit their drum, it lights up. And they're
counting down-3, 2, 1-until it was the 8th second of the 8th minute of
the 8th hour of the 8th day in August 2008.
Eight is a very lucky number in China. They were pounding this;
it was unbelievable to see. There was so much choreography,
so much human coordination, it just absolutely blew me away.
So many of them did it in such detail as you look at this.
Just absolutely phenomenal and really,
kind of an interesting sort of metaphor for what I wanted to talk about today,
which was how all of our collective, tiny little behaviors that we do all
the time-although probably not as choreographed as these guys-can
add up into some pretty remarkable things.
In particular, when we start to look for patterns in those things,
we find very, very impressive results.
So that's a bit about what I wanted to talk about today.
What I think is sort of coming in the future here on the web, or at least
as far out as we could possibly see, and what we should do about it
as people that kind of built the web every day.
So, let's get into this. I'd like to talk about the future
by talking about the past, and I 'd like to start in 1974
because I think 1974 is pretty remarkable year with implications
for what we do today on the web and for what we're going to be
doing for years to come. 1974, to me, is kind of conceptually the end of the 60's.
At least with our recent history. I was a history major in school,
studying history and journalism. One of the things I learned is our
chronological years and our conceptual understanding of decades in the
20th century don't always line up. And 1974 was sort of when
underneath the mainstream sort of bubbled up in everyone's lives.
For example, you can see in 1974, the environmental movement that
the hippies and the counter-culture was really adamant about really
hit home to people when the OPEC oil crisis happened. And as you can see,
the prices of gas skyrocketed to 81 cents/gallon. Can you imagine, right?
But it was a big deal. People couldn't get gas, and you could only buy gas
on odd and even days based on your license plate number.
For the first time, people are like, maybe there's something to be said about
sustainability and the environment. Likewise, the sort of
anti-authoritarian/mistrust of authority, that sort of stuff that was
brewing and bubbling down under the surface, came up in
1974 in the Watergate crisis. When President Nixon here,
with one of his trusted advisors, had to leave office, right,
that kind of shook the country in a pretty fundamental way
that it hadn't before. These changes were happening
all throughout our culture here in America.
Entertainment, think about the 60's.
1974 was the first year that these guys released an album.
Talk about a shift, right? From where music had been before and
to where it was going. Anybody here had this poster in their
bedroom like I did? Okay, wow, a bunch of you, cool.
Say that reminds me, how many of you were not born in 1974?
Alright, we've got a lot to talk about then today.
Big shifts were happening in business. 1974 was when the
House of Representatives started proceedings in breaking
apart the natural monopoly of our telecommunications system
here in this country. At that same time, as this sort of big,
centralized focus of power was being taken apart,
bubbling up from underneath was the first mention of a white
paper from Vinton Cerf and Robertv Khan, using the term
"internet" for the very first time. Big shifts, change happening.
I had a bit of an epiphany happen in my life. This is a picture of me in 1974.
That's me right there in the gray, next to my brother.
I'm pointing and laughing at his ridiculous plaid pants right there.
But I guess it was the 70's. I was six years old in 1974 and one of the things
I really enjoyed doing was going shopping with my mom.
The reason for that is that when we went to Sears,
we got to go to the restaurant inside Sears. This was before food courts
and stuff in the mall. We would go to Sears and have lunch and
it was a very big deal for me as a little kid. One time,
when we went that year, we couldn't find a seat. It was very busy
on that day, and so we sat in the back corner at this really
weird looking table with a glass tabletop. When we sat down at
that table and looked down through the glass tabletop, we could see
that there was a television pointing up at us. And on the screen was this
laughter at Pong screen
And it was the very first time in my six-year-old life that I had
ever seen a video game. And I had this tiny little six-year-old epiphany that
I could control what's on the screen. And that was amazing to me.
And we fed quarters in and we were playing with it.
Eventually I convinced my mom - this is what it looked like,
one of these deals with the knobs on it so you could control it
- we should have one at home. That Christmas, I got one of these
and my brother and I would play for hours and hours, playing Pong,
controlling what was on the screen. Participating in what was happening.
Traditionally, what had been up to that point was a consumptive media,
a media in which we were not participants, where we simply
engaged in consumerism. But now we could start to participate
and that led to more. I'll show you a brief engagement of
my history with technology. This is how, from there, sort of increasingly
more sophisticated tools for participation with what's happening
on the screen appeared. That was sort of a big deal in my life
and set the tone for the rest of my career. At the same time,
in that same year, IBM released their first hard drive.
This is the Winchester 30-30 hard drive. They called it the Winchester
after the shotgun, I guess, because it had 30 milliseconds sync time
and 30 MB of storage. Look at how productive it made those people.
They are getting stuff done. For IBM this is important because
it is the first commercially viable hard drive that they produced.
Before, it was always a custom installation,
but you can just almost buy this right off the shelf. Although they did lease it.
If you did the math for what they were leasing it for in '74 in today's dollars,
the price came out to be about $100,000 a gigabyte. So we've
come a long way I'd say, considering today you can sign up for Amazon web services,
and you can lease from there a gigabyte about 15 cents/month.
I get an e-mail all the time telling me this has gone down
a couple cents more every time. Clearly, storage has not only
become commoditized as a business, but it has become accessible
in a way that is remarkable, and what that means is there has been
a tremendous proliferation of data storage and just data in our lives.
There's,just data everywhere. It surrounds us and
we're all watching this data. And if you put these two things together
- this tremendous amount of storage with the tools for participation
- you get this roadmap for where we're heading on the web, which is:
we are in control of what is happening here but there is too much happening
to feel any sense of control. So that's what I'd like to dig into.
Particularly, I'd like to start withů I'm a designer by tradeů
I take a design and user experience focused look at these sort of things.
And so I'd like to take a look at just a whole bunch of data
and see how we might use design to help put people back in control
of the data that is kind of plugging through their lives.
There's tons of data in the world, tons of data in our lives.
I saw that when I was at Google. I spent about two years at Google.
I left in May and while I was there, I saw a scale of data
that I didn't even think was possible. I remember coming into
the office so excited that I bought a terabyte hard drive. A terabyte.
Do you know how much you can store on that?
The engineer just sort of shook his head and laughed
and said, "Yeah, we fill up a few of those every day. "
I said, "Well, I still think it's pretty cool" One of my colleagues
that worked at YouTube told me that every single minute of the day,
13 hours of video gets uploaded. I don't know what all that video
could possibly be, but I think it's awesome that we are collecting
so much of this stuff. But that also puts responsibility on those of us
who are the collectors and who are responsible for sort of caretaking
people's data to make it accessible to them and to give them some sense
of control and ownership of their data.
That's what I'm going to show you here.
This is a bunch of data. When I look at this, it makes me feel
a little dumb because I don't know what it is and I don't know
what I should do with it. But I also realize at this point in
my career that if I'm sort of feeling dumb, it's usually not my fault.
It's usually the fault of someone who is trying to communicate with me
and is not doing it very effectively. I think that since
design is communication, it can help us understand and put data
into context by doing a little visual communication,
and we can start by adding some metadata. So here, just by
adding some simple labels, I can show you that I'm displaying
on the screen the average rainfall in inches/month of
some North American cities. Now, you can start to make use of this data.
You can see that this data means something. You can sort of
wrap your mind around it. I think you can do a better job of that
using some design techniques. For example, we can add typography
and some arrangement of the typography so that we
can see what's more important. And we can recede what's less important.
We can create a bit of visual hierarchy. I can also do something
like this to make the data even more accessible and even
easier for you to use. So now I've taken the value of each one
of the table cells and changed the saturation of the color behind it,
so that you can really tell. Even if you're in the back of the room
and you couldn't see the numbers before, you can at least tell
now that if you go to Miami in the summertime,
you're going to get rained on. San Francisco is very dry in the summer.
So I've made that data more accessible, more actionable, and more efficient.
I've brought out the numbers to get meaning from that. I could
take this a step further and try this. Who thinks that's better?
Really, no matterůokay, a couple of people.
Educators like that data, I guess. Maybeit's better, right?
Saying that you like it and somebody else doesn't
means that there are differences in the audience. Maybe the data is important,
and certainly if you were all a bunch of meteorologists
you'd say let's go back to the data. Perhaps if this was a little widget
on a travel page and you're focused on getting a task done but wanted
but it's a dangerous line to stop crossing.
When you get up to that line between communication and decoration,
I think we have to really be careful, and I see
this all the time, all over the place
In fact, the USA Today, sort of notorious for this, right?
I have looked at this for months now, and I have no idea what
they're trying to communicate to me. In fact, they do this so often that
they become known for this. And The Onion, that lampoon
newspaper thing, they do this all the time, right? You look at this
and the numbers add up to like 136% or something.
They have this little quote down here and says,
"Note: The U.S. Safety Council reminds all skaters that helmets are for dorks. "
But you see examples of people making fun of the fact that it's
so easy to decorate data-decorate it and lose the meaning.
This is a pie chart I found on the Internet that
shows what percentage of the pie chart looks like PacMan. That's pretty good.
So, like I said, it's pretty easy to go too far and decorate.
I think we have a responsibility with people's data to give them tools
to make it accessible and understandable and to help
them so they don't have to think so much about the data,
but rather think about what they're trying to get done.
That was something we paid a lot of attention to when I was at Google.
We spent a ton of time-about 14 months
just in the design phase of redesigning Google Analytics when I got there.
And it was something of a scale again-a scale of data-that I had not had any
experience with before we did a lot of research into what's the
best way to approach so much data and so many different views of data.
We did some research. I went back. I was a history major so I went back.
I'm going to show you some of the stuff we looked at that inspired us
by going back to 1854. This is an engraving from a cholera outbreak
in London's neighborhood of Soho, 1854. It was vicious.
There have been cholera outbreaks for centuries before that,
but none as vicious as that one. This particular strain of cholera
would cause someone to wake up and have a bit of breakfast
and think they weren't feeling well. By lunchtime,
they were motionless in bed and by dinnertime, they were dead.
So there were literally people dying in the streets and the neighborhood.
It affected one in three people. So if you imagine a neighborhood
as dense as Soho was at the time, it was absolutely devastating
what happened there. This guy, Dr. John Snow, lived in that neighborhood.
He was, perhaps, the first anesthesiologist figuring out
how chloroform worked and things like that,
which made him very popular, as you can imagine, considering
what they usedbefore that to do surgical procedures.
But he was also kind of into numbersand kind of into math
and interested in using math to track down what was happening.
And he did. He used this map which shows the cholera deaths in the
neighborhoods stacked up by address in the neighborhood of Soho,
and he used this map to convince the city council of London that the
outbreak was not being spread by air, but being spread by water.
And nobody had understood; they hadn't discovered germs yet.
They didn't understand that disease could be waterborne.
And he used this map to convince them of that.
There had been plenty of visualizations before that about where people
were dying and from what. And as you can see, those visualizations
tended to take on some of the mythology of the day.
For example, this one shows one hundred years earlier,
during an outbreak south of the Thanes River that people
thought you could see black clouds over the city as they tried to plot
where it was breaking out. So what Snow did was track
that down. Many people have written about and believed
that he actually created the original chart.
He didn't. He took this chart, which was from a sewer
engineer who had done the marks on the chart originally. Snow took that
and what he did was edited this. He took this map and took everything out
of it that didn't actually show his point, which was that there was an
infected pump right in the center and that the people who were
drinking from that pump were the ones who were dying. So not only
was he able to use data visualization by condensing the story and
clarifying the story visually, but he was also in one of the
very first examples of using data visualization to prove a
curable path to get people to stop believing in a lot of the
mythology and superstition, and frankly classism andracism of the day.
And to say, "Look, I can show you where that's happening. "
If you go to London, you can see that they have the pump
without the handle in Soho. I went there not too long ago
and did a little data visualization geek pilgrimage.
I had pictures taken and I know, it's sad, but it's really cool to
see that there. In fact, it's a little hard to see but right
across the street is a pub called the John Snow,
and I would go have a pint. I would recommend the beer, not the water.
Another classic example that we used as inspiration when we were using
the analytics design was Charles Joseph Minard, which shows
Napoleon's march from Poland to Russia in 1812-1813.
He did this in the 1860's. A remarkable visualization and any
of you who have read Edward Tufte can read pages in his book
of how amazing this is.
But what this does is it shows different variables across the chart.
It is Napoleon's route that he took with his army.
The width of the route and the width of this line indicates the number of men
that he had with him at the time, so it shows the population of his army,
the route they took, the time, and you can see that it decreases over time.
On the route back, it also shows the temperature as they went
through one of the most severe winters that Europe had seen
in the 19th century. And you can see that pathway
get narrower, and narrower, and narrower, as more and more
men died on the way back. Just a tragic, tragic experience
that they had there. I loved this quote that I found
when I was doing research about him. It says, the "The aim of my
cart figurative isůto convey promptly to eye the relation not
given quickly by numbers requiring mental calculation. "
Another way of thinking about that is really: Don't make me think.
At least don't make me think about your interface to this data.
Let me find the story that's in there. I find that just absolutely a
remarkable way to think about this. If you think about that chart of data
that I showed you at the very beginning, and when I added the blue to it,
you were able to stop thinking about the numbers and see the
patterns that were in there much easier. The story in the data is something
that I've been increasingly inspired by as well.
There are so many stories in charts like this. In John Snow's chart,
with people dying around that pump. In this chart,
you can see all kinds of stories. For example, when Napoleon
crossed the Berezina River. Right here, where that arrow is,
you can see him crossing that river with his troops and having literally
22,000 men die that day when they crossed the river.
You can see negative 20 degrees below zero. Just a disaster.
In fact, in French today, the word Berezina is an idiom for catastrophe.
It was just a disaster. This is a fine art representation
of the same story that we see in the Nard's representation of the data.
So, finding stories in the data and using design to illustrate that to
bring that forward is a big thing there. Another is where people find inspiration.
Harry Beck worked for the London Undergrounds in the 1930's.
In 1936, he drew that map of the London Underground. I love this photo
because he's just so proud of it. But this is what he had to work with.
This is what they had beforehand, showing the tube system in London.
Harry wasn't a designer, though. He wasn't even an illustrator.
He was an electrical engineer; he was a draftsman.
So, all day long he would look at stuff like this,
look at the map, and look back at this.
Then he would say, "I think we can put this together in a way that's
going to make more sense to people, especially Londoners who only
care about two things when they're on the tube: one is what
station connects with what other station;
the other thing is what side of the river am I on?"
And this is his map that he drew to kind of illustrate that.
So effective that it is still used some 90 years later.
Talk about scalability of the designů that's pretty impressive.
I like that Harry got the inspiration from his daily work,
from those electrical drafting diagrams that he was
doing all the time. The same thing really happened to us when we were doing the
analytics for the design. This is a chart we
used that ended up in the product that's kind of the main focal point
of every one of the reports we created. This is hard to get to though.
It's easy to do a case study of design, and you, show the thing you got,
and then say Here's what we did. " But to show the way"
that we got thereůwe were trying all kinds of different things.
It's murky data; it's deceptive. I would ask if the paint was on
top of the green or rather it was behind the green.
I got 50/50 answersůclearly, this would not work.
Sometimes, we were trying to shoot different kinds of data points.
I told my team to hold off on the visualization of the information.
Let's work on something else for a couple of weeks.
Something will come to us. Let's put it on hold for now. Literally, about two
weeks later, I was watching the DVDs for the Indiana Jones Trilogy.
I was watching Raiders of the Lost Ark, and you know how
and this line on the map? I watch that, and then the next night,
I had a dream where Indiana Jones's airplane was flying over
the charts and the numbers. And I went to work the next day, and I'm like,
"We have to do it like they did in Raiders of the Lost Ark"!
I think they're used to that now. Yeah, but this was the inspiration that sort
of led us to the clean looking charts with dots.
It's interesting to see where inspiration can come from.
Historically, I looked at a bunch of things like that,
but we also looked into the real world for examples
for how we can use data visualization as a way of
understanding the context of the world around us.
I found a great website called MegaPenny Project.
Its sole purpose is to track the amount of pennies there
are in the U.S. today. It turns out, there're
about 200 billion pennies. This is their estimate of the U.S.
Treasury Department. How do you put your head
around $2 billion in pennies? What does that mean?
You've probably got a few in your pocket right now.
Using some of these ways of gaining perspective are pretty powerful.
This is $10 in pennies, a four-inch square that weighs just over six pounds.
Here're a million pennies as compared to a dentist.
This, though, is $10,000, and it weighs three tons.
This is a billion pennies, which is about the size of five school buses.
This is the number of pennies in circulation right now.
We have gained insight in a rather pointless data point
about our world. At least we have gained some insight there.
If you keep going on the website, it's fantastic.
What does a trillion pennies look like?
What does a quadrillion pennies look like?
Pretty soon you see the earth in front of pennies.
It's fun, it's Megapenny. com You should check it out.
But it also kind of illustrates a way that we can help each other
understand the world around us and understand things of
scale that we may not be able to think about without being able
The artist, Chris Jordan, does a fantastic job of this.
His work or photo renderings are computer generated renderings
that he does and then prints out at high-resolution in enormous scale.
It's about 40 feet long and 20 feet tall. These panels show statistics
about our lives around us. This, in particular,
is a very powerful one. As you walk in the room, you
could imagine this is about actual size.
As you get closer, you can see he's rendering plastic soda bottles.
You can see these are the two million plastic bottles we
use in this country every five minutes. That's consumption!
That's consumption in a tangible way,
perhaps in a way that could change behavior.
Perhaps we are not using so many of these.
He has a quote in the artist statement that I absolutely love.
He says, "Statistics can feel abstract and anesthetizing,
making it difficult to connect with and make meaning. "
That's the first thing I said when I saw the data from the rainfall.
I look at it, and I kind of gloss over. Finding a story in that data
and bringing it forward and letting people engage with the story
is something that I think is incredibly important for us to
do on the web every day. That's one of the big lessons
we've learned looking back into historical data visualizations.
We need to find that story in the data,
and then we need to tell that story in visual ways.
Assigning different visual cues to each dimension of the data,
whether it's raindrops or a path getting smaller.
in a way that really resonates with people. Then, right,
we have to remove everything that isn't telling that story.
I was a journalist for a few years before the web and that's
the thing I just hammered home. You take every single thing
out of your text that isn't doing hard work and telling the story
to make it as effective as possible. The same thing applies to design.
We take everything we can away that isn't directly
communicating to the story we are trying to tell.
So, really powerful stuff. But it also felt a little like it
only got me halfway. It applied to how we have
historically done design and data visualization.
I realize that, sort of looking back on my career,
that I've been incredibly fortunate to work with some
absolutely stunning designers. I worked at Flyer magazine in
the mid 90s, and then I was over on the digital side
working on Hotwire, and these designers were the best in the world.
I learned so much while I was working there.
One of the things that stood out while I was doing that was
the designers working on the magazine who came to the
web side were very different, in that the experience for
them was pretty dramatic. They have to kind of forget almost
everything they knew about traditional design to bring it to the web.
The most important aspect of that was control-giving up
this sense of control. We considered that as
a sort of constraint. We couldn't control the typography.
We didn't even know how big the page would be when
it was rendered. A very different experience
from obsessing over a printed magazine and sending out
the artifact to people. Now we are simply sending our
content and our recommendations for how it might be rendered
and hoping that it would work out on our users' screens.
Very difficult to give up the control there. One thing that
I understand- and saw dramatically-was the designers
who gave up the control and embraced this as an amazing
part of the web were the ones who were most successful.
They were making that transition from the print world to
the digital world. They could kind of go Zen,
and say, "I have no control but what can I do to my
content to make it as flexible as possible?" I saw this most
clearly with a designer named Dave Shey who created the
CSS Zen Garden-one single piece ,of very structured, semantic
marked-up copy. He put up some content and then held
a contest to see who could add a style sheet to it in the
most dramatic and moving and .interesting and lacking way
Literally thousands of designers created different designs with
this one piece of content, showing this incredible separation
between content and presentation that leads to giving up.
control over time. You can ;see this in designers all over the place
Jeremy Keith has written .a number of books on AJAX
A super-simple example on his website is this pull-down
menu that has different themes .that can be applied to his website
Kind of silly and pointless. You can select a Zeldman
view and it changes. He understands that his content
can be viewed in a variety .of different ways
I know most of the content that I consume on a daily basis
looks like this to me. This is how I see almost
everything that I read online, and I don't see the design
at all because I'm using a RSS aggregator to get through as
much content as efficiently as I can. That means that all the work
a designer has done on the visual presentation isn't even
coming through, which is completely appropriate, especially when you
think about devices like this. This is a Braille browser where
you put your fingertips on the keyboard here and you can listen
through your fingertips to what's coming through.
You can use the buttons above to respond.
Really, it was about a sense of control in the design work that
we do to embrace this type of change. We are trying to transform
people who go from us, as designers who find
stories in the data, to giving up control and putting the data
in user's hands so they can find their own stories.
That, I think, is the big shift for people coming online and using the web.
This is a ridiculous example of that.
I've scratched the surface now on interactivity.
What I mean by that is giving control over to users so they
can find their stories here. Now, by using an interface-rather
than simply looking at what was the interesting story-enabling
people to find their stories in their data is the big point.
Whether it's data or content, giving the control over to
them to find it is what's really, really powerful about the web today.
And increasingly moving forward as we have more and more data
about ourselves online. So, for example, looking here
at a map of relationships that we have in social networking.
This is something that someone has on their MySpace page.
Their social graph is visualized for them.
Or, for example, the music that people listen to.
This is a site called lastfm. com,
where you can put a tracker inside your iTunes.
It will report back to the website what you're listening to and the site
will make basic recommendations by using a collaboration engine.
You can see patterns emerging from data, even data that's very
important to us. This is EveryBlock, a website that takes streams of
data from government and normalizes that data and formats it
at the block level in up to six or seven metropolitan areas.
This is the stuff that's happening, so everything from crime reports to
health code violations to building permits or even news articles.
they can find a story that's interesting to them.
Think about what Nike is doing with Apple.
You put a chip in your shoe, and it talks to your iPod, and you
connect it to the web and can compete with your friends.
It's impressive that you can do this kind of stuff.
This is from rescue time, which tracks what you
are doing all day long.
This provides visualizations with your level of productivity and efficiency.
This is a nightmare to me. But it's just another example of
collecting tremendous amounts of data and making it accessible for people
It's difficult for us as designers because we don't know what
the story's going to be. Just like we didn't know what the
fonts were going to be that people had installed or the
screen resolution that they had. Likewise, we don't know what's
going to be important to people. So instead of us finding the story,
again, we have to let them do that by creating tools that let them
manipulate their own data. There's a design agency called
Staymen in San Francisco that's worked with real estate companies
to make this visualization of Real Estate. You can zoom in on a place,
and you can see it's a rather remarkable tool.
Down here, you use the data as an interface, a way of navigating.
We did a similar thing with Google and RSS where you can use
the data of your behaviors to interact with the system as a
method of navigation. Likewise, as a method for sort
of understanding how you use the content in your life what's updated.
Again, patterns emerge in the data,
creating opportunities for action for people to control the data
that's coming through their lives. That graph that I showed you of
somebody's social network is interesting, but not that useful.
This is the same thing. This is my social network and
an interface with the website called Doppler was put on top
of that network to manage what my relationships are with
people in their context. I have a sort of overall graph with
the people that I am connected to. This is permission management
for who can see what I'm doing. Likewise, I can do the same with them.
This turns into a bunch of serendipity where I find out I'm
in an airport with one of my friends at the same time.
This is an interesting interface to allow me to do that.
Likewise, this is the application that we sold to Google about
3 years ago. It's called measuremap, and it's an analytics tool for bloggers.
We used people's data as a navigational path for them
to act upon what they found there, using the charts and graphs as
the way of interacting with the system and understanding what's going on.
Finally, I think we have to provide filters to enable clarity for people
- just like I was trying to do moving from the blue squares
into the raindrops. We need to help people understand
this unbelievable amount of data by helping them whittle it down
to the data that's important to them. Let me show you a couple of examples.
This was also acquired by Google a few years ago.
We are literally looking at millions of data points right here.
I have configured this particular visualization with life expectancy
of people on earth on the vertical axis and how much they
make on the horizontal axis. Every circle is a country in the world.
The color of the circle is the continent that country is in.
The size of the circle is the population. A ton of data here. At the bottom,
it has a Play button so you can push play on this and see it over
time and watch what happens. As humans, we're basically doing better.
We can use filters to find the stories that are interesting.
If I filter something out, we can watch the pattern that happens there.
In box one, I was doing pretty well. Then we get to 1990 and everybody
starts dying from AIDS in Africa. Then we look at this chart and you
see everything in the bottom left-hand corner is blue
because that's the color of Sub-Saharan Africa, which was
devastated by AIDS in the early 1990s. Look at the story by providing filters
that the providers of data never would have imagined existed.
I see this all the time. This is from the New York Times.
I like this one because it was created by the information design team
of the New York Times in 2005, and then they let it keep running.
We don't know how long the data's going to persist or how
it's going to grow over time. We have no control over these data sets.
We need to provide tools like this one that provide that analysis for people.
The tools, the filtering, the using the data as the interaction
design as a way of understanding data.
This is something that was really hard when we started at Google.
This is a small excerpt of something that we had designed that showed
different relationships between reports and data in those reports.
There were 96 different reports that we eventually got down to
just about a dozen. This was on our wall;
it was about a twelve-foot-long diagram.
It was a ton of data and a ton of different types of data.
Eventually, we had to acknowledge that we couldn't possibly design for
every possible relationship. We had to provide the tools to
allow people to find those relationships on their own and find different
visualizations for the data that made sense to them. This is
an example of doing different things. If you think about it historically,
and where that brings us today, really, really valid principles in the
annals of information design where storytelling is really very important.
To us, we need to think about that storytelling as a sense of discovery
for our users. Rather than telling them a story, we need to enable discovery.
Rather than providing them visual cues for that story, we need to
provide interactive tools that allow people to find the stories themselves
by manipulating their own visual cues. Just like Snow edited down his
his chart and just like Mr. Beck and his map of the underground,
some things are very important. Now, we need to think about that as
filtering rather than just us doing the editing.
That's a big shift, right? That's a big shift from the data into the tools,
from the design into the interactivity. It comes from a lot of things that
people call Web 2.0-people working together with this collective
intelligence. It's the only thing that I could add to this.
All of this is important into the "what we should do" and the
"why we should" and the "how we should do it. "
But the why is also very important.
One of the things that I spent a lot of time on when
I was at Google was helping all of us on the product teams
understand why we were doing this and to whom we
were communicating with in these interphases.
Google is a very user-centered place. I was happy to be there
because of that. This is a visualization that
came from the lobby that I walked past every single day.
It shows, in nearly real-time, the queries that are happening
at all the data centers around the world.
So we can get a sense of just how many different
people are using our products every single day.
Each shaft of light is coming out where there is a data center located.
The height of the bar is how many queries and the color
are the languages. Really a lot of information.
I would wear at Google occasionally that said,
"Math is easy. Design is hard". You know what's not funny at Google?
This t-shirt.
But when we would talk to engineers and to the
product people, and we would ask what a
Google user was like, everybody would say,
"Pretty connected, uses technology upgrade, needs
a monitor upgrade. But, you know, pretty comfortable with technology. "
That's a fine strategy to get started with.
To go after early adaptors and see things they're interested in.
Google was well past that trickle down period.
They had the rest of the world now. I would say, "Yes, this
is probably very true. People use this everyday and connect
but we get people like this. In fact, 70% of the traffic
that came from Google was from outside of the U.S.
The U.S. is quickly becoming a minority in the amount of traffic
consumed and requested in the search queries.
That meant that we had to understand how to design
products that would be adaptable and understandable and fit into
context for people who have a very different notion of
how technology fits into their lives and how they use
technology to connect with their social network to meet
the information needs of people like this.
Likewise, this is a very big part of our audience.
We talked to lots of teenagers, and they told us lots of
very interesting things. For example, they use e-mail
to talk to old people and that was about it.
That's pretty significant for a company that has a very
big e-mail application. That would mean that,
over time, we are going to have to shift a lot of our
understanding and assumptions about what it means to
provide communication tools. Most kids have very flexible
and context-aware methods of communicating with one another,
depending on what they're doing. Very rarely are they sitting in
front of a computer for long periods of time sharing
documents back and forth. They do those things in
very different contexts. The other thing that we learned
from talking to teenagers is that they have such a different
notion of private versus public that it is a generational gap
as divisive as music was in the 50s and 60s. Most people,
who didn't grow up with interconnected computers and
grew up with technology later, believe everything on the
computer is private until it's shared. Whereas kids believe that
everything they do is public unless kept private.
will be used. Likewise, we also had users like this, who would
make two or three gigabytes a day of data and needed help
indexing, searching, sorting, archiving, and keeping that data safe.
All these users have multiple contexts throughout the day.
That is another thing we must understand.
So you start to take all these vectors and start
to put these together into some sort of understanding
of what an audience is.
It is virtually impossible, or maybe very difficult,
for global communication on the web every day.
I'm going to show you a website that I think absolutely
understands their audience so clearly that it is one
of my favorite examples. I've shown this site in
presentations that I've given for years because I think
they've done absolutely perfectly. It is the U.S. Department of
Agriculture's Hay Net, and they understand that
their users either need hay or have hay.
They have one variable. That's all they need.
It makes you think that it would be awesome
if we could design all of our websites that way.
But we can't. People are complicated
and diverse and rightfully so. That's awesome;
I'm glad that they are.
So we spent a lot of time at Google doing research.
That also caused a bit of crisis, or reflection,
on my part for the last year. We did so much of this
that I wondered if it started to become the point and
if it took away from what we were really trying to do.
I have always been such a proponent of user research.
We would talk to people in the field and observe and
understand what they wanted and what they needed.
We wanted to see how we could get things done.
I could look at this photo and see a dozen different product features.
That's good for us to constantly be building that
empathy to see what people are trying to do. So we would
do tons and tons of research and we would bring that
research back into the lab. Everyone at the table would
be analyzing and synthesizing everything that we did.
We would end up with walls and walls of documents full
of this kind of stuff. We would then prototype this
stuff and bring it into the lab. I feel so fortunate to have
had this kind of infrastructure at an organization like Google.
This one in particular is really cool. Stuff like that is phenomenal.
These monitors are fancy, laser monitors that shoot beams
into your eyes and track your eyeballs where they go.
I don't know how it works, but it's really cool.
We would get these videotapes of people using the products
and where their eyeballs were going.
You could see that over time, right?
You could aggregate that into visualizations, like this,
and what they were looking at. All of it is incredibly powerful but
all of it sort of became overwhelming. We did so much research that
I wondered what we were trying to do.
That got me to today, where I try to find the balance.
I had a really good conversation with Jeffrey Zeldman
who keynoted here a few years ago.
This is a picture of him at a conference recently.
If you've been doing as much as he has for the last
15 years for the web, you can wear whatever
you want to a conference. Jeffrey gave a presentation
at this conference saying that it's important that we
start with the user and we know ourselves.
You need to know yourself as well as the users.
I agree with that, with a tiny tweak.
I think it's incredibly important to know yourself, right,
and then understand users. Because if we're just doing
the research and we're just constantly looking at data
and analyzing it just from the user research,
I think it only takes us so far. We get mediocre products
that don't have any life to them and they lack passion.
I did a conference earlier this summer with a bunch of
entrepreneurs with what motivated them and what
inspired them to make their products that were
eventually successful.
I talked to Matt who created Word Press.
He's 24 years old, and has been working on Word Press
for six years. and what he said was that "Everything I built came
from the frustration that it didn't yet exist. "
This is a theme that I heard over and over again from
people who left comfortable
jobs and found themselves asking what to do.
These things bothered them enough that they had to fix it.
And that sort of started to inspire me as well.
What is inspiring me so much that I have to go fix it?
It is pretty rare to me that it would come across my desk
with the variety of jobs I've had in the past.
Occasionally it would, but it was normally the thing
that I was most proud of. We talked to a journalist
who writes GigaOmen-that's a popular Web 2.0 Omen.
Even he-jaded journalist guy-said,
"New ideas come from your heart, not your wallet. "
This is difficult for the venture capitalist group to swallow,
but it's true. The passion comes from inside.
To me, it started making a reflection on all the research
I've done in the past. It is a lot like travel and going places
that I've never been before and experiencing new cultures.
Seeing the world through other people's eyes.
It didn't make me able to make things for them,
but it decreased the distance between me and them, right?
As we go around and see the different way people live
and see the way they experience the world,
it decreased the distance between who I was and who
the rest of the world was.
That is where we can draw a lot of the empathy and inspiration.
That's why we really need to make our products so good,
because it's really hard work. I had this quote; it's long,
but I want to read it to you. It's from Steve Jobs, and he said,
"When you start looking at a problem and it seems really
simple with all these simple solutions,
you don't really understand the complexity of the problem.
And your solutions are way too oversimplified,
and they don't work. Then you get into the problem,
and you see it's really complicated.
And you come up with all these convoluted solutions.
That's sort of the middle, and that's where most
people stop, and the solutions tend to work for a while. "
He's right. That's where most products end up being:
convoluted solutions to really difficult problems.
"But the really great person will keep on going and find
the key, underlying principle of the problem, and come up with
a beautiful, elegant solution that works. "
I find that incredibly inspiring, especially considering it wasn't
this Steve Jobs that said it, but this Steve Jobs that said it.
When Jobs launched the Macintosh in 1984,
he was 24 years old. That's where we sort of need to
find the energy to keep our products better.
To find that passion and inspiration from stuff like
that in the past. We can look back on what
people have done before us. We can look. All the stuff that's
happening around us in the contemporary world and the
amazing future that we're glimpsing into with web social
networking and data visualization and stuff like that.
We need to look to users for inspiration to shorten
that gap between us and them, so that we can understand
what we can make, so the stuff we feel passionate about can be
understandable and usable for them. And it's remarkable where
we find inspiration all around us. Inspiration can come from
almost anywhere. At this educational conference,
I give you some homework. Here is your homework, if you like.
These are some of the things that brought me on this journey.
Stephen Johnson is a popular author and wrote the book
called The Ghost Map. There's a URL at the end, and
you can download this presentation. He wrote this book about
John Snow's journey through the data visualization,
and it reads something like a detective novel.
It's really interesting.
The godfather of information design is Edward R. Tufte.
He has written a bunch of books on information.
I read this about 12 years ago, and it fundamentally changed
the way I saw the world. So start with that one if
you really want to get into data visualization.
Then, if you're really technical, this book by Ben Fry-who
created the Java processing library-called Visualizing Data is
90% lines of java code and explanations for how the library
works and 10% absolutely fascinating stuff bringing
interactivity into data and maps and charts.
So check that one out. And that's what I have so
thank you very much.
We have time for a few questions.
And you can take the presentation because I have
a creative commons copyright license.
Leave my name on it and do what you want with it.
Question: With the stock market going down, how do you see
the relationship between what's going on outside
of this room and what you've been talking about here?
Answer: I am totally freaked out by the last 10 days of news.
But I'm more optimistic this time than the last time,
because I was right in the center of it last time. In 2001,
when most of my friends had to move out of San Francisco
and everyone lost their jobs, that was a bust that happened
I didn't see that this time. Although this one is much
more significant, of course, I don't see the same implications.
When I was living in San Francisco in 2001,
you couldn't even say dot com without people laughing
and sneering at you-with people saying,
"What have you done with our economy?!"
Imagine that you say your job is a sub-prime
mortgage guy. That's what it was like for us.
So I'm a little more optimistic now than I was the last
time our economy took a correction, shall we say?
But I don't know. Also, I grew up in California.
I'm used to boom and bust, because that's all we ever do back there.
So we're going to ride this one out,
but this time it won't be as bad for our industry.
That's what I think. Do we have any more questions?
Question: Why was the fumble your fault?
Answer: When I said the fumble was our fault,
I mean it was driven by speculation in our industry-the
.com industry-which was completely unrealistic expectations
for what would happen to the IPO's that were going on.
So, we took a lot of the capital that was in the markets and
moved it into something insanely risky that wasn't proved out.
When that fell apart, everyone moved into real estate and
that didn't work either. But that's essentially what happened.
I think collectively, or at least on Sandhill Road in Silicon Valley,
we could take a little responsibility for 2001.
Less so now. That's what gives me the optimism that we're
doing all right this time. We are out of time so
I'll be around for a little while, but thank you very much.