Tip:
Highlight text to annotate it
X
(music)
What I want to talk about today can be maybe best
expressed as a kind of a syllogism
in educational environments, we are affected by the way we communicate with each other.
Who says what to whom how? What tools we use and how we use them?
Shapes our interactions. The corollary
is the changes in the way we communicate with each other
change the educational environment and digital
tools, the Internet, mobile phones, the applications built on top of them, represent
a big change. We have a medium that's global
social, ubiquitous, and cheap, and
the way we do what we do, academic work
educational work in the kinds of institutions we're all members of
is changing as a result. Now, in the academy we always like to think we're
masters of our own destiny, that we get to choose whether or not to do this,
but in many cases we are just being borne along by the tide of larger
changes that are affecting society as a whole. So, rather than thinking
how can we sit down and decide what we're going to
do, what I want to talk about today are a bunch of stories in which
we're being asked to react to things that are obviously shifting
in the environment and we have the opportunity to react to them in ways that make the
educational environment better. So, rather than talk about that in the
abstract, I'll start by talking, by just telling a story that I think
illustrates some of what's possible. It's a story that is not from
within the academic environment, but it is a story about education.
So, this is Arthur Gugick. Gugick is a math teacher
in Cleveland, Ohio, and an adult fan of
Legos, they call themselves. And one day in his
attic he found the plans to his house.
This is the early part of this decade, he found the plans to his house, and he looked at the plans and said, "I can build that in Legos."
So, he went out and he got a bunch of pieces and he built the little Lego model of his house.
And he liked it. It was the first time he'd really worked in 3D, he said, "That was interesting."
So, he decided for his next project he was going to do the Taj Mahal, which is a
little bit like taking a jog around the block, and then deciding you like it then saying you're going to run a marathon.
So, Gugick goes and gets several thousand pieces
from various different kits, studies the Taj Mahal and figures out how to
assemble this, and then, in a move
that test the boundary between labor of love, and maybe you should get some fresh air,
he does the reflecting pool. Right? (audience laughter)
which involves finding flat Lego pieces that represent the color
of the 3D model as reflected in water. Right? It's a famously
hard 2 dimensional problem. And then
he does it again 3 more times
in various different scales. Some of these are macro some of these are micro.
Alright? But, he has over the course of the last
decade built and then published the plans
for building a Taj Mahal out of Legos from this
size to this size 4 times.
So, I called him up and I said, "Why are you doing this?" Right?
(audience laughter) And he's a math teacher, it's an interesting
2D problem, it's an interesting 3D problem, but he said, "You know what really made the
difference was MOC pages." I'd never heard
the word MOC pages before, so I went and looked it up. MOC pages stands for "My Own Creation".
And it is a place for people who build Lego
models to show them off to each other. Right?
And Gugick had always built with Legos. He said he'd never had
what's called the "dark ages" among adult fans of Lego when you stop around
the age of 12 and you pick up again at the age of 24. He had just been
building with them straight through. But when he came across MOC pages
he had that sense that we all have finding a niche that likes what we like which
is "Oh my god, I'm not the only one!" And so
he said something that has resonated in my ear
about education ever since. He said when I'm doing new
project, I take care to document it, I take photographs of it, I tell people how to build it,
and so forth. And I will periodically update
gugick.com, I'll periodically update my own site,
but I always update mocpages right away.
Right? It turns out that being embedded in the community
is a better motivation than simply showing off
from an ego-centric way, in an ego-centric way. Right?
Now Legos would seem to be, Lego building would seem to be the classic solo activity
and we'd expect the usual intrinsic motivations for solo activity
to be the thing driving Gugick, right? We'd expect him to be driven by
a desire for autonomy and for competence. I did this myself
and I'm getting better at it. And that is indeed part of this.
But it turns out that when you harness autonomy and competence and you
attach them to social goals, to the goals
of membership and generosity, you actually
get more participation than if you just
than if you're just working by yourself. So, I saw him do
this and I changed my own practice in the classroom.
I rearranged my syllabus in my Spring seminar, I gave
up two weeks of content in order to make the student
paper due two weeks before the end of the term instead of
on the last day of the term. Because you think I'm the person
the students are supposed to be worried about impressing, but what I
told them instead was, you bring your papers in two weeks before the
end of term and bring them in in multiple copies, please, because I'm
assigning them to the other students. And the readings and discussions
for the final two weeks of the class are reading each others' work
and discussing it. Then I put the students together in thematically related
panels. And the level of quality
when they found out that it actually wasn't just me reading them,
their own peers were going to be looking in on their work, I got a higher
bump in quality than any other thing I have ever done in a
dozen years of teaching, right? So, what we're seeing consistently
is that the embedding of the kind of personal actions that
we want people to engage in can actually be
charged by embedding them in a more explicit social context.
So here's that same observation
made again entirely diagrammatically, right? This is what
we're used to, right? This is canonical sage on the stage
there is some information that we have and we want to broadcast it to you, we will send it
to the edges of the network. Here are this week's readings. Here are the assignments and so forth.
And we're getting used to the idea
that in a communications environment where the students all have these
tools on their desks, in their pockets, and so on, that they can talk back.
Right? That this channel is potentially two way. But we all have these
same issues, which is the faculty member doesn't scale, that if it's a big class you can't
have a two way interaction with everybody. And so while
we are wrestling with that, the world is changing in another way, because the
freakouts, the real change in my view in the
academic environment isn't coming from the red lines.
It's coming from the green lines. It's coming from the fact that any
two people plugged into this communications matrix can now talk directly
to each other, no professionals anywhere in sight.
And there are way more green lines than red lines. And that's not a
function of the technology, that's just a function of math. That's how networks work, right?
And it is this social context that's always
been there, but has been much more salient and much
more visible and much more flexible in this environment that gives
us a choice as academics and educators, and in fact
forces the choice on us as academics and educators about what to do
about this. So, what I want to talk about this morning are
three big dilemmas we
all face as a result of this change.
Places where we don't get to pick whether
we change, how we do business, we only get to pick how.
And the first of those dilemmas I want to talk about is the tension between individual
and collaborative work.
So this is a picture of Chris Avenir. This is a story from a
couple years ago at Reyerson University. Avenir was a freshman
at that college, and in his first semester there he took Chemistry
101, and as freshmen have done since time immemorial, he
found that Chemistry 101 was hard, and he decided that he was going to turn to
a study group to work on it. But Avenir's 19 years
old in 2008, right? The Web is as normal as the
telephone, right? Google has existed since he was five
social networks have existed since he was ten, Facebook
has existed since he was 15 years old, now he's 19, he wants a
study group. Where does he go? He goes to Facebook. He joins a study group
on Facebook called "The Dungeon - Reyerson
Chemistry 101" and joins a number of other
Reyerson students talking about the class. And then one day
he gets a call from the Dean for Academic Affairs.
And he's been called up on charges and threatened with expulsion.
How many charges? 147 charges!
One for being a member of this Facebook group, or being an administrator rather
of this Facebook group and 146 additional charges for
every other member of that group, right?
And so, from Reyerson's point of view,
what he's done is obviously cheating, right? There's no
simple alternate explanation here.
What Avenir's done is obviously cheating and
he has engaged in something which is contrary to
the spirit of individual work. And yet they
called the Facebook group "The Dungeon" because
that's the name of the room on the Reyerson campus where real
world study groups meet, right? So here is what Reyerson
says about this case, right?
Reyerson says this is absolutely clear cut, and here
is Avenir's reply.
And so the question that's raised by these two statements, by this
comparison is "Is Facebook
more like traditional media, just in a new form
it's just like a mimeograph machine, or is
Facebook a study hall, just
moved up online? Which of those two models
are we to use when thinking about Facebook? And the answer
is that's the wrong question. Because it turns
out that the thing Facebook is most like is Facebook, right?
It does different things than any of the previous models in fact
if it didn't, it wouldn't feel new, right?
There is no way to analyze our way out of this.
Because there is no previous
practice we had that would allow this kind of thing to happen.
So here I think is the dilemma, that
Avenir surfaced and Reyerson reacted to.
We have two different messages we send
to the world about our students. To the outside we
say: We do quality control of individual minds.
We bring them in, we educate them, we put a diploma on their foreheads and we
ship them off to you. And inside the four walls of this
institution we say to the students: Welcome. We're glad you're here
you've joined the community of scholars, conversation is how we advance learning.
We tell them that they should want to be in small seminars.
That this is, in a way, the end point of the education, when you get down into
a room of 15 or 20 people hashing out the ideas around the table.
We give them lots and lots of spaces so that they can get together in
the real world. We have special places on campus where study
groups can meet. But, we've never
had to have both of those messages happen in the same
place at the same time. One message is for the outside world,
the other message is for inside these four walls. And those message don't get
messed up, because we're in charge and we get to decide which messages go where.
And the students aren't going to mess that up for us because it's not like 18 year olds are all global
publishers or anything, except that now it is exactly like that.
Now it is exactly like that, and the ability to
keep those two messages separate
is now gone. So, this is the first of the three big dilemmas.
which is how do we
manage a world in which we do have to concern ourselves
with individual minds and we also know
collaborative work is very often how people learn
best, right? And I think what the Reyerson example shows is
there is no set of existing principles that we can pull out and
lay down and say there, that is how we've always done it and this isn't changing.
Because what is practically possible
is so different from what used to be possible. Avenir's
argument, very clear, which is we've alway
done it this way, the university asked us to do it this way, and yet the
argument I think could have made, they didn't unfortunately make it,
but they could have made is, if there were 146 people on that Facebook
group, there's a good chance that somebody was free-riding, right?
Because digital media works not by resisting free-riders
but by being tolerant of them. If you're
at a table of six people studying chemistry
and somebody comes along and says, "Hey, can I just copy the answers from you guys?"
It's easy enough to say, "No, get out of here. We're actually trying to learn!"
Little study groups are self-policing. There is no
table that can seat 146 people having a
conversation. You can only get to that scale online.
And if there's 146 people there, it's a good chance that
many of them are lurkers, and it may be that the lurkers are just trying to
spider the answers. Unfortunately we don't know, because as a condition of Reyerson's
settlement with Avenir, he was graded down for the class but not expelled, they also deleted
the Facebook group. So, what could have been a key
moment of study for how is this working
in our environment, we lost the ability to know what was happening.
But we can't just let all of this stuff flow up
to the network, because we also have to make sure we're not creating a situation
where people are being free-riders, right? We're not asking
you how many hydrogen molecules are in water because we don't know
we're asking because we want to see you figure it out.
So, very often we have to give students challenges where getting the answers
is more important than having the answer, and in that
case we also have to police against this. So, it's not like you can
clamp down on it and it's also not like you can just say whatever, it's going to be
fine. We really do have to go back to first principles, what kind of
education are we trying to produce, and deal with the fact that the personal
and social context that we've always had going on at the same
time in the university are much more overlapped and much more visible
to all the participants in the system. Now,
this change, right, this change in the more salient, more visible
more workable social context isn't just a change in the
academy, this is a change in the world, and because we're part of the world
we're changing as well. My name for this
ability to deal with
to build new value out of coordinated voluntary participation
my label for that has been "Cognitive Surplus".
The ability of people to commit some of their time and talents
to tasks in voluntary ways, right? And
the ability to take that free time and those talents
and to coordinate them at very, very
large scale. And so in
researching the most recent book, I needed some way to talk about that, to talk about
what was happening, and I needed some unit to say
this is what we're talking about. And so the unit I took was
Wikipedia. Wikipedia is our largest
most public participatory
effort. There's obviously things like open source software as well, but for something almost
anyone can be invited into participating in, Wikipedia plainly the largest.
And I needed some way to describe how large it is.
So, with Mark Wantanburg, who studies Wikipedia,
first at IBM, now at Google, did his back-of-the-envelope calculations.
How big is Wikipedia as a coordinated effort?
And the answer is, it took about 100 million hours to get Wikipedia to roughly
the state it's in today. It took about 100 million hours of human effort.
Every edit on every page, every top page, the source code, all the rest of it.
That's a big number. Obviously anything that takes 100 million hours to do is a big
deal. But, how big is it compared to free time
and talents, the available free time
that Wikipedia is drawing from? So I needed another metric and I used
television - number one use of free time worldwide.
So, imagine a graphic representation of
the amount of time we spend consuming television. How big is that? Well it's 200 billion
hours in the United State alone every
year. Or, put another way, Wikipedia
is there. The entire
ten year effort of Wikipedia is a
little corner of the television watching of one country
in one year. Put still another way, in the United
States, we spend a Wikipedia project's worth of time
every year just watching
I'm sorry, every weekend. We spend a Wikipedia project's worth of time every weekend
just watching ads. So this is a very big number.
This is the resource from which these surprising
new projects are being drawn. And
using Cognitive Surplus, taking advantage of this surplus, we
can see incredible scientific, literary
artistic, and political innovation
spreading everywhere. (audience laughter)
We also get a lot of LOLCAT.
LOLCATs as you can see are cute
pictures of cats, made cuter with the addition of cute captions, and
this also is a use of people's free time, right?
And so one of the constant critiques of this medium is
this, you cannot possibly be serious
about the Internet, mobile phones having any kind of important social effects.
Just look around at the
cultural product of this medium. It is frivolous
uses done by frivolous people.
But here's the thing. That ALWAYS happens. And I don't
mean that happens with Internet tools. I mean that always
happens full stop in the history of media.
The printing press gets invented, middle of the 1400s, it doesn't
take long before the first *** novel appears.
The Venetians turn this into a commercial
business and shortly thereafter, someone says "Hey,
you know what people would pay for, I bet?" And here
come the *** novels. It takes another 100
years for anybody to even think
of scientific journals, right? So,
there are two things I think
can be drawn from this example. One, the desire to declare
an entire medium or an entire user population
serious or silly isn't going to hold historical water.
And, two, ***
novels are the kind of thing we got because we are the kind of people we are and the printing press is
the kind of tool it is. But the scientific journal took
real cultural work. Philosophical Investigations,
the first scientific journal in English was produced by a group of people who
called themselves the "Invisible College". And they said
We're not going to believe anything that's not true.
This is a fairly remarkable attempt, because human beings are
not well provisioned with the ability to subject our own ideas
to the kind of withering scrutiny that would root out error. We do, however,
have a related skill, which is we're very good at subjecting
other people's beliefs to the kind of withering scrutiny that would root out error.
And so the scientific journal became
the platform for making believable assertions. It was
the essence of peer review. It became
an academic norm, but it didn't just happen
the way *** novels just happened. It happened because
a group of people said we have this capability and we have this cultural
norm we want to support and we're going to fuse them. We get
LOLCATs because we're the kind of people we are and the Web is the kind
of thing it is, but the value we're going to get out of
this medium isn't going to come as a side effect of
cute pictures of cats, it's going to come because people like us decide
we're going to take this medium and we're going to press it into service for cultural norms
that we want to support. And this mixing of
the serious and the frivolous is something we're
having to get used to, and I think it's something that we need to
say when we're looking at this medium just because the use
today is frivolous, doesn't mean it can't be serious tomorrow. I'll tell you a story that I think illustrates
that. This is a screenshot from a Web log called GnarlyKitty.
Alisara Chirapongsed, who is a fashion blogger in
Bangkok, and at the time of this story is back in
2006, at the time of this story, she was a fashion student. She
was in the university there. And this was her blog and here is one of her blog
posts, it's called Gone Fishing, you probably can't read it because of the font, but
the post reads, "No, for real. These past few days my fingers have been glued to my
iPhone, thumbing my way through tournaments on flick fishing."
That's it. That's the entire post. Why
would anybody post that, right? And the answer is
she's not talking to us. Any day of the week you could go down to the
mall and find a group of teenagers saying things to each other that are at least this benign.
But in that case it's clear that you are the weird one, because what are you doing spying on the
teenagers, right? (audience laughter) We're not
used to things being in public, but not FOR the public.
Right? The logic of the media environment that I grew up in,
that many of you grew up in, is filter, then
publish, right? Decide whether something is worthwhile and then
and only then do you put it out. Which means that if I can see something that has been published
I feel like: A. It's for me, and B. It's already been filtered.
for quality. That is not the logic of this medium. The logic of this
medium is publish, then filter. Because the costs are so low the
imperative to decide in advance has gone away.
And much of the filtering we see is being applied post hoc.
Whether it's search or tagging or recommendations from friends
it's about finding the good stuff after the fact rather than before.
It's a huge shift in the internal logic. And what that means
is that we get a lot of things that are in public, but not for the public. She's actually just talking
to her friends. She's just doing it in the cheapest way possible
which happens to let us look in as a side effect. So then you can say Oh I get it, this
is all just silly, right? This is, most of this stuff in fact, this
stuff is just frivolous. And that's true
to the 90th percentile, but 90 is not the same as 100, because
here's what happened next. There was a coup in Thailand.
And Chirapongse took her camera phone, and she went down
she went down to see what was going on, and as the coup was
happening, the heads of the military went to the heads of the media and said please don't report on the coup.
The heads of the media? Okay, right?
Here goes, da da da da da, music on the radio, nobody's saying anything
about the coup. Chirapongse gets down there, she takes the first
photos of tanks parked in front of government house in downtown Bangkok.
And she blogs them. And it gets picked up by a site called GlobalVoices
Rebecca McKinnin, Ethan Zuckerman's international blogging crossroads.
And suddenly, a fashion blogger in Bangkok
has become the go-to person for news of the coup.
Because the Thai military, not rolling so much with the blogosphere in 2006 for
forgot to shut down the Internet. And so people are pouring in
from all over the world, her commenters are saying they're saying this on CNN, that on the BBC
this is what we're seeing on the streets, first person reports and
she has not become a journalist, but she has committed an act of
journalism. So you can't say that gnarlykitty is
frivolous, because some days it's not.
So this goes on for a while
and then she posts this. She gets a new phone, it's
in that kind of manish gray black that mobile phones
tend to be in, and she says well that's not the kind of phone I want. She gets out her copy of Photoshop,
and she gets an image of the phone and then she makes it pink
and I don't know if you can see this in the back, but she makes a little Hello Kitty
a little Hello Kitty figure on the back, says you know that's the phone I want.
And you can imagine the comments on this post - GET BACK TO THE COUP!
(audience laughter) Right? And in
response she posts
one of the great posts in the history of explaining
blogging to the rest of the world. It's called Life LIke
This. The first sentence is "Okay let me clarify some stuff." which
is clearly the windup to a rant, and quite a rant it is, and she says
in essence, this is my blog. I'm here
to blog about things happening in my life and some of what's happening in my life
is that there's been a coup in my country, and some of what's happening in my life is that I want a new phone
and if you're just here to read about the coup, that's fine.
But if you want me to stop talking about this other stuff, I'd rather you leave.
(audience laughter) There is no professional
media outlet in the world that would see its audience grow by three
orders of magnitude and tell the new readers to buzz off.
(audience laughter) But she did it and she
did it because she's not motivated by money, right?
She's not trying to be, amateurs are not little biddy professional
she's not trying to be a little professional outlet, she's
trying to be her on the Internet. And one of the things we're
having to get used to is that as we know from our own
lives, our personal life and our social life and our
intellectual life are actually all mixed up around the edges. And you can certainly
find centers for all of those things that are clearly isolatable. But at the
edges they're all mixed up and now those edges are much more visible.
So any attempt to say either this medium or this person
is either a serious medium or serious person or a silly medium or a silly person
is going to break down, because you're going to see serious people do
silly things, and silly people do serious things. And we're going to have to as a community
get used to the idea of understanding what people are doing
across larger parts of the context of their lives, because we're going to be able to see
those larger contexts.
So this, this change in
voluntary public behavior brings me to the second
big dilemma I want to talk about and I think it's an easy enough one to understand.
Anybody who's done participatory work will see it.
Which is the dilemma not between individual work and
collaborative work, as with the Avenir story, but the dilemma about dealing
with the gap in degrees of individual participation.
So, this is a screenshot from
the Wikipedia article on Pluto, and
Pluto, I've been studying the Pluto article over several years, and it's actually quite interesting. You may remember
in the middle of the decade when Pluto was getting kicked out of the planet club? There was
a huge amount of attention that came in on this particular
article. How do we characterize this change? What's going on? And so forth and so on.
And if you had a long skinny browser, the Pluto page would look like that.
It is as you can see, it is
illustrated throughout, the bottom third of it is all
links and external references. This is a remarkable piece of work. If you want
to see Wikipedia at its best, this is an article I can recommend without
qualification. The time of this screenshot had been edited
over 5000 times by over 2 and a half thousand users.
So, you think, this bottom up construction, it's kind of an ant hill model, people bring in
their grain of edit and the whole Pluto article kind of builds up.
And it turns out that that's not how it happened.
at all, right? If you graph
if you put the users along the bottom, and you graph
the number of edits they contributed along the Y axis, that
red line shows you what participation actually looks like.
Which is the say that of the
2500 users,
roughly half of them contributed only one edit
once ever and never returned. They're not members of the community in
any sense, they're not collaborators in any sense.
Meanwhile, over at the head end there, Serendipodous, the dean of the Pluto article,
has contributed hundreds of edits
by himself and the curve
as you can see falls off very quickly, and in fact it follows what's called a power law distribution
which is to say roughly an order of magnitude decrease in participation
for roughly an order of magnitude increase in rank. The tenth user
has contributed only about a tenth as much as the most active, the 100th user
only about 100th as much and so forth.
This is the normal case for participatory
systems. It is edits on Wikipedia articles, it is checkins
to the Linux kernel, it is tags on Flickr photos, it is views
on YouTube, this distribution shows up over and over and
over again. And anyone who does participatory work
in their classes sees that the larger the group being asked to participate
the greater the range in participation between the most active student
and the least active student. And so in addition
to just the dilemma of individual vs. collaborative work, why are they important?
We have a second dilemma, which is the bigger we make the
collaborative system, the bigger the gap is going to be.
And how do we reward the people at the head end and encourage the people
at the tail to do more work? There are lots of methods.
But the one thing we can't do is open up a participatory systems
and hope that everybody will be getting the same thing out of it.
One of the things I do is to put students in small
groups, multiple small groups when I ask them to do this work, but I
have the luxury of mainly teaching seminars, right? And so there are different techniques
that are going to be required for large lecture classes. But the one thing we have to know
is that when we open up participatory logic, we also invite
and in fact all but force this imbalance to express
itself and there needs to be some way of addressing that.
Both rewarding the people at the head end, but also making sure the people at the tail end
are not just engaging in social loafing, not just checking their, doing their one thing
so their name shows up in the system and then doing nothing else again.
And this problem comes about
because we're starting to have to treat our classes
as ecosystems, not just as collections of individual
minds, but as ecosystems, and you see it already in discussion
classes, but really once you move to more collaborative, more participatory
work, the ecosystem aspect of this
expresses itself really clearly, which brings me to the third big
tension I think we face. And this is not now about our relationship to our students
but about our relationship to each other, which is a key
part of the academic ecosystem, the educational ecosystem is
the disciplinary matrix, right? We think that
anthropology is one thing and that sociology is another thing.
For no good reason, but we do. We think that
we should put psychology over here and then you gotta walk all the way over the quad
to get to sociology. And we encourage
people by doing that, we encourage in particular our students by doing that to believe that those are
actually different things in the real world, instead of
simply the way that we manage our own careers. And I don't think
you can see this tension and what's happening because of these tools anywhere
more clearly than in the library, right? Because the library is where you really
have to carve things up. So here for example
just as an example of the kind of thing that can happen. This is the Dewey Decimal System
categorization method, this is the 200s.
Ruh-ro!
(audience laughter) So it's easy enough to say well
poor Melville Dewey, right? Working a the dawn of these kind of classification systems
of course you get these kind of benighted British attitudes. We have much better
systems now. And we do indeed have better systems. This is D from
the LIbrary of Congress's History of the essentially Old World History, non-American
history. So a much more sophisticated
much more refined cataloging system. But I want to call your attention to the items in red.
And, just you know, to review
the geography, I
Africa, Asia,
Balkan Peninsula, right? What is the same
about those categories? It's not square miles,
it's not headcount, it's not GDP, it's books!
At the time of the categorization system, the Library of Congress had about
the same number of books about the Balkan Peninsula and Africa.
The categorization system is not designed to
leave nature at the joints, it's designed to minimize seek time on shelves.
And so the shelf turns out to be the secret
organizing principle. So, no matter how hard you
think I have a theory that will allow me to understand the world,
you get these kinds of exceptions. And so essentially what
is implicit in the decategorization Library of Congress
is we believe that history can be best understood by looking at geography.
And so you find these nations and you do
the history of those nations. But you can almost, if you read that
list you can almost put yourself in that meeting where it's Friday
afternoon, it's 4:30, we've done an amazing job, thank you everybody, we're
just about ready to wrap, and somebody goes "Aw, the gypsies!"
(audience laughter) And so they get put at the end
as punishment for not being geographically
stable, right? So, even within these
systems that are designed to help us manage
intellectual life, there are these
breaks. In the world we're in now there is
no shelf. There is no, this has
to be in one place, so it can't be in another place. The essence of a database is
that things can be every place all at once. And so we're starting
to see experiments with that possibility. The Smithsonian
took several thousand images and just dumped them on Flickr,
the photo-sharing service and just said "Let's just see what happens."
We're going to give up saying we are the sole arbiters
of value for these photos and we're just going to turn it over and see what happens.
Here's what happened. People tagged them like crazy.
They create all kinds of tags and these tags
these are the most popular tags. Out of an enormous list of tags those
are just the A's, and if you render the entire list it's a huge
number of tags. People coming in and labeling these
images in all kinds of ways. And what
any individual's label going to mean on any given image?
Not much, but when you do it in aggregate,
you start to see where there are communities forming around
photos that couldn't find each other any other way. So I want to go
back to this image, which are the most popular tags and I will call your attention to three different
labels. Cyanotype, mustache, and
steampunk. So cyanotype is photographic
historians that don't actually care about the documentary content
of the photo, they care about the technical medium.
These are media theorists. There's no one who if you had
to have a shelf would organize photos based on substrate.
And yet why shouldn't that community be able to get as much value out of
these photos as possible? Mustache, right? Self-explanatory.
Perhaps it's a costume designer, perhaps it's a scholar
of fashion. We don't know. It doesn't matter. And most
speculatively, steampunk. Steampunk is a
kind of retro-science fiction in which the idea that the inventions
of the 19th century created 20th century
values, albeit with 19th century technology. Essentially starts with the idea
that babbage is different and the mechanical computer works
and moves on from there, and so here are the steampunk images. And these are
essentially inspiration for potential steampunk authors.
Not at all serious, but this
is the Smithsonian, these are American citizens, why shouldn't they be able
to get this kind of value out of the system as well. So what the Smithsonian discovered
is you don't release these materials to
see what the public does for them because you know what kind of value you're
going to create. You release the materials because you don't know what kind of value
you're going to create. And because you accept that people are going to do all kinds
of things with your images that you might not even thought of. One of the images
they released was a very nice picture of a Puddingwife Wrasse,
and if you go to the comments on the image, the first comment is form someone
who runs a database of fishes and wants to add this picture, but
because it's a good illustration. And so this image gets
imported into an academic corpus without the Smithsonian even knowing
that that corpus existed, and then the next image is "Hey, I put this on a purse!"
(audience laughter)
And that range of value
no one could plan for that, and even if you could, no one could execute it
because going to your executive director and asking for extra money to
work with the textile designers? Not happening!
And yet here it is. It's being pulled into place.
So, I want to end with
a story that just happened last August. And
it's been the most interesting thing I've seen happen in the academic community
in its scholarly form
and I think it illustrates all three of these dilemmas.
The dilemma between individual work and group work, the dilemma
between the degree of participation and participatory ecosystems,
and the dilemma of academic self-conception,
the difference between what we think of ourselves and what's possible
in the world, right? And the story starts here.
Last August, early August, a researcher at HP
Labs, Vinay Deolalikar showed up and he said I have solved
the most important problem in computer science, the problem
is called P=NP or P does not equal NP, and
subject to violent compression, the P does not equal NP problem is basically the
observation that computers are really good at some kinds of numerical tasks
like adding huge columns of figures and they're really bad at other tasks, like
factoring huge numbers. And those two kinds of problems
seem to be different, but nobody can say for sure. Maybe they're the
the same kind of problem and we just haven't found a good way to factor big numbers really fast.
Or maybe there really is something fundamental going on that
keeps those problems separate. Now it sounds pretty abstract. It is pretty abstract, but
if P=NP every encryption system in the world is broken
tomorrow, because if we figure out a way to solve
hard problems fast, the entire edifice of digital
security has been built on sand. So this is a problem we would like to know
the answer to. So Vinay Deolalikar pops up and says "Hey, I have a
proof that P does not equal NP, and he
shares this proof with a few friends, who share it
with a few friends, stop me if you've heard this story before, and lo and behold
a few days later, somebody says "Oh, there's this very interesting proof
going around. I wonder what *** Lipton thinks?"
*** Lipton runs Gödel's Lost Letter. It's the best blog
that you and I will never read because I'm certainly not smart enough to read
it and unless you're a serious computational scientist
and mathematician you probably either it is the premiere blog
of the P does not equal NP problem. Because if it's a subject there's a blog about it and P
does not equal NP as a subject therefore we have Gšdel's Lost Letter. So,
Liption gets this proof, and he turns around and says to his
audience, this is very interesting. This is a serious attempt. It's worth some discussion
and some discussion is what it gets. He posts about it eery single day for
a week, getting dozens and in some cases even hundreds
of comments, and then they take those comment threads
and they movie it over to something called the polymath wiki. And the polymath wiki
was Nick Gower's proposal from 2008. Which is
framed around a single question: Is massively
collaborative mathematics possible? Is it possible
to have an enormous number of contributors
and this is essentially the topic paragraph from
Gower's original post proposing this, and the key
part of this I think is, in other words what you would
not tend to do, at least if you wanted to keep in the spirit of the thing,
is spend a month thinking about the problem then come back and write ten pages about it, rather you contribute ideas
even if they were undeveloped and/or likely to be wrong.
drawing a link between that observation and the undergraduate population I will leave as an exercise for
the reader, but this collaborative model
not the norm in the mathematical world. So they drag it over to the polymath wiki
and they disassemble the Deolalikar's proof and they say
interesting, serious, something here, fatally flawed in its current form
here are the issues. This process takes
eight days. If I was in publishing
and I heard about anything happening in eight days I would be
worried. If I heard of it happening over the course of eight days
in August I would be terrified.
And it gets worse. Polymath doesn't just
take apart existing proposals. Polymath also
generates new proofs. They
have a proposal, this is about finding primes, it's called polymath four, which is
just the number of the problem they took on. They set out to think about
finding primes, they got an interesting result, they turned it into a paper
and they sent that paper off for editorial review.
And the review came back, I'd like to accept this paper it's a good paper
it's in the field. Here are some changes I'd like to make
Let's take that space out. We've got an extra "is"
an extra "that". Now I'm a big fan of copy editing.
But in terms of mathematical value, this was pretty
weak tea. And then, my favorite sentence in the whole
thing. Finally, the paper does not clearly say who the authors are!
The person reviewing this paper
cannot conceive
that it's actually unclear who the authors are. They assume
that it's just been left out. It must be clear who the authors are. And to
say "kinda" is not an acceptable
answer, but that's the truth. It was done on a wiki, so
lots and lots of people participated and as we know their participation
looked like this. And the journal can't
accept it. In either sense of the word accept.
Academic journals, which at the beginning at their inception were the core
of the academic enterprise. We're the thing that that let us
look in at and review each others work. Have
many cases have been reduced to a rubber stamp. The
final validation phase that allows the tenure committee to consume
the work and to consider it for promotion, which is not nothing.
But it's not helping much on the core
mission of advancing knowledge. And we're
in a world where that divergence is only increasing.
This happened last August. It's not like
polymath is going to shut down anytime soon, so there's a
tension here between individual and group work, a tension between the range of participation
and a tension around academic self-conception, they all show up
in the P does not equal NP story. And they're all going to show up in our lives
at some point over the next months. So to answer
what do we do about this? If you want
to integrate these tools, how do you do it? Now I've been teaching
theory and practice of social media at the Interactive Telecommunications Program
for about ten years and in that time the average age of my
students, mid to late 20s. The average age of my students has stayed about the same.
My average age on the other hand has been growing
at the rather alarming rate of about one year per year.
And so I find myself in the strange position of teaching my own youth as
ancient history. Students who have no adult conception
of the 20th century. I have to tell them about
what it was like when the Web got invented, and the hardest thing to convey
to them is how crazily contingent it all was.
They think well Wikipedia has always been a good idea. Obviously
Linux would own the cloud. How else would you do it?
So I've got a slide I put up for them and I'll put it up for you. It's a sentence from Linus Torvalds
first public statement about what became Linux and Larry Sanger's first
public statement about what became Wikipedia and what I think comes through
both of these is how incredibly modest they were.
Let's just try this. It's just a hobby.
It's not going to take you more than ten minutes, right? These were not enormous
world-beating claims. This was just I bet this would be
interesting. Let's see. Even now
Twitter, right? The piece of software that's most recently crossed into
the population scale, which is like 100 million users,
this is the phrase that got Twitter started. Nobody said I want to drive the Iranian
crazy. They said I want
to have a dispatch service that connects to our phones using text. That's where it got
started. So, to experiment with
this stuff, it turns out that the key is to find the person
in your organization who's got the big vision, who knows how to transform
the organization in one big organized push and
lock them out of the building and don't let
them back in until they come back with 10 medium sized
ideas or a hundred little ideas, because most of this stuff
doesn't work. There's no point in having a
perfectly good little experiment thats going to become
the thing that changed the world. Nobody knows.
On that day in 1991 Linus Torvalds certainly didn't know it was coming and
had all the information it was possible to have. If I'd
handed you a mail spool, if I'd handed you everything
literally every public email from 2001 and said
find the one is going to change the world sent on that day. No one
would have said Oh yeah, those people making the encyclopedia
that's going to be the one. You would never have picked that message right?
And so it turns out that lots of low cost
experimentation, not hundreds and hundreds of things because you get distracted
but also not one big push. Try six or seven things. Get
rid of the stuff that obviously doesn't work, double down on the stuff that obviously works
but it's the stuff in the middle, the stuff that kinda works and maybe you can change
it and learn from it. That's where the real
change is coming. So I'll end with this image. This is John Fitch's
steamboat, the first working
practical steamboat, it plied the waters around Philadelphia, and
Fitch committed himself with the idea of taking a steam engine and putting it on a boat to make
to create new motive power. But this isn't John Fitch's
first steamboat, it's just his first working steamboat.
This is his first steamboat. He took a regular boat and he said let's
make the steam engine row the oars. We'll just take a boat
and add some steam and then we're good.
This did not work. The next thing didn't
work either, the thing after that didn't work. It was the fourth
steamboat that worked. We've come through a
phase where we've said let's take the university and add some internet.
That'll go great, and what we've discovered is
some things go great, but a lot of stuff doesn't fit.
Because it's not just about adding new
tools to an existing ecosystem, it's about building a new ecosystem.
The way we communicate with each other creates the environment
that we share. Changes in the way we communicate
with each other change the environment we share. And we're in the middle of a big
set of changes. Thank you.