Clay Shirky on 'mass internet collaboration' at London's ICA

There are my rough notes from Clay Shirky's talk on 'mass internet collaboration' and the question and answer session afterwards at the ICA on February 4 – I haven't had time to tidy them, so they really are pretty much 'as is', though I have checked my note-taking against my digital voice recorder. I have cheekily highlighted bits I found particularly interesting. I asked a question about museums at the very end, just keep scrolling.

I've put some photos (mostly of slides) up on Flickr.

Introduction – it's a speculative talk, looking at recent events, lessons from the Obama campaign and his early government.

Technology is having profound effect on the social environment. The recession means fairly significant choices this year.

The five word summary of his book 'Here Comes Everybody': group action just got easier. Devices, tools and the applications built on them have provided an antidote to some of the hassle factor when organising groups of people to do something.

What happens when you lower the cost of socialising, whether for amusement or to achieve something?

First example – no pants day. The group Improv Everywhere provided a place and time where people would show up. There was no technology in the event. The event could only happen, make sense, have the effect it had, in the real world. And yet it's the kind of event that couldn't have happened in any other time. To coordinate something like this globally, could only work if event organisers had access to a medium that's global, ubiquitous and social.

'No pants day' only works if a lot of people show up, it has to be a social event. Network technology isn't just another slice of the pie, it's the medium. The technology makes it possible but is not embedded in event itself, just the coordinating mechanism. He's used an intentionally trivial example [to introduce] the 'third sector'.

It's tough to get people to do anything, and there are two mechanisms for doing it. If you ask, can you get more revenue than it costs to do it, then it's the private sector. Or is there significantly high social value? Then it's the state, public sector.

That dichotomy has been a 'universal truth'. But the triviality of something like no pants day means no social cause or revenue. So it falls into the 'social sector'. We now have such low cost abilities to get groups of people doing things together that question is not about profit or social value, but 'would it be fun or interesting?'. The logic of social sector is 'why not do it?'.

Pressure on traditional institutions is growing because presence of social sector means the previous monopoly on group action is being challenged.

New models of production are going to challenge… things considerably more important than whether people are wearing their pants on the subway.

Chris Avenir's study group example. It caused a clash between world views, metaphors – arguments about what Facebook is – is it like a newspaper, or an extension of real world social life. The problem with metaphor is that it flattens whatever is most important about the question being asked. It turns out that Facebook is like Facebook.

It's not a question which of our old behaviours do we layer over this possibility, how do we update academic culture to take account of this. The physical limit imposed by the space where real world study group took place prevented freeriders because they were easy to detect and kick out. 'Small groups defend themselves quite well against freeriders. Larger groups don't'. A network works because it's freerider-tolerant, not freerider-resistant. This is pedagogically problematic in a study group.

Two messages – one to students about progress through discourse, one to outside world saying you get a student filled with knowledge. Facebook caused those two messages to collapse. There's no easy solution to this.

The capability for large-scale geographically unbound learning is something academe can't integrate without changing dramatically, but it's also something they can't forgo. [I'd be interested to know what the UK's Open University feels about this.]

Power to engage is not with the academy, it's with the students. The change being brought to the institution isn't being brought by the people who run the institution, but by the members of the institutions acting as individuals.

Next example – Gnarly Kitty, a student in Bangkok with a personal blog which was suddenly swamped with attention during the Thai coup. "We're not used to seeing things that are in public but not for the public". Change in logic – why not publish it. When the Thai coup happened, the media didn't report it, so this blogger was the first person to get pictures of tanks in Bangkok out of Thailand. Her blog became conduit of news and images from in and outside Thailand. When she went back to talking about phones she'd like, commentators didn't like it. Her response – "This blog is my personal blog where I usually write things concerning my life and things I like. … I'll continue posting about the Coup whenever there are crucial updates that need attention but I will not make any political comment or turn the whole blog into a politic-centric one."

There's no way to square that kind of amateur motivation with what we're used to. Journalism is moving from a profession to an activity. The blogger committed an act of journalism, and it mattered on a global scale. There will be occasionally people like Gnarly Kitty who commit occasional acts of journalism. "Occasional times a billion is a lot".

The infrastructure where journalism operates has changed, not because of choices made by existing journalists, but the environment in which journalistic institutions has changed because of things like this.

Next example – the Obama campaign and win. It's has transformed not just what's possible, it's also changed people's sense of what's possible. In 2006 you could not have found a bookie in the States who would have taken your money on a bet for a black president. The mainstream media couldn't report Obama as a significant possibility because then they would be seen to be shilling for him. There was a sense of 'noble but doomed' in their reporting in 2006/07.

A few things happened to change this. Will-I-AM's video had 16 million views on YouTube by early 2007. "It made Obama seem possible". It shifted perception so people thought 'maybe that could happen'. Because in politics, perception is reality, so that change of making Obama seem possible had the circular effect of making Obama be possible.

The Obama campaign did not commission or vet the video. Will-I-Am did not need permission. Obama is "the first platform candidate". First to send a message to the public that said, this is my message and this is how it's conveyed. They gave implicit permission and the materials were all re-use and re-mix friendly. Sometimes the friendliness was quite explicit – Creative Commons licenced – sometimes it was just that you could see how to do it. It was easy to imagine how to make campaign media from it. Not easy to see how to do that for McCain, partly because his campaign was in terror of loss of control. The McCain campaign's idea of outreach was making copy and paste comments available. People who weren't political professionals were able to participate.

'Sing for change' video – a school teacher had kids sing a song she'd written. "It was a horror" to see people who weren't old enough to vote repeat words in unison that an adult had put into their mouth. Teacher clearly thought she was doing the right thing. The reaction was instant.

You can take down a URL but you can't take back a video that's out. Copies were annotated and remixed by Republican commentators (this was the height of remix in McCain's campaign).

But – no-one blamed Obama. There wasn't the implicit sense that 'if your name is on it, you must control it', if we don't like it it's your fault. Old media rules no longer applied to new media landscape. ACLU (American Civil Liberties Union) meetup case. "If it's got our name on it, we have to police it as tightly as if it came from central office".

Those days have passed. The adoption of these tools by people repurposing… [?]

"You can't give up control without that control going somewhere." There's no real way to embrace these tools without embracing a kind of two-way responsibility to the people who make most passionate use of it.

MyBarackObama (myBO) – they didn't use Facebook because Facebook is too satisfying. They didn't want to create a satisfying experience because they wanted to drive their users to do hard work like hold meetings, go out and get votes.

Facebook groups [are practically] 'shout outs for ending world hunger'. It's not clear how the link between joining and the nominal effect is actually happening, but the act of joining is satisfying and means people don't actually take any further action. MyBO was utilitarian about putting people together, forming groups to do things.

When Obama reversed his position on a particular bill, the principal challenge to that behaviour came from MyBO. The ingenious trick was to make the act of joining the group against signing the bill be a vote for the message of the group. Obama was forced to make a public response – he took the flack for it but didn't change his support for the bill.

This holding to account of a politician by their own supporters was a watershed moment. There was a cynicism that this was propaganda, until people saw that his own supporters were willing to call him out in his own forum, and that the Obama people didn't shut it down or try to hide the response.

Will he govern like he campaigned? He ran his candidature as a platform, the user-generated media was significantly value, created useful feedback loops. Because he's engaging the passion of the supporters rather than a 'managerial relationship', he has to respond publicly when they're angry. How much of this is going to carry over into his administration? was the site of the transition team
which asked, 'what should our administration be concentrating on?'. You could make a suggestion or vote on other people's suggestion. First issue to the top, and it stayed that way the whole time, was 'please legalise medical marijuana'.

That has been a real comeuppance [for Shirky]. Democratic legitimacy of participation over the internet, lowered barriers to political participation seemed to be a good thing. Yet when he sees this, in the context of economic crisis, wars, etc – how can they think the marijuana thing is what he wants the first day administration to focus on.

[I thought about this later – could some of this be because the mass of ordinary people don't feel capable of making or voting on suggestions for complicated specialist issues like the economy? Was there a long tail of more 'serious', nuanced suggestions focused on pressing economic, foreign policy, social and environmental issues? It'd be interesting to find out.]

Democracies don't just have votes because they allow the group to come to some kind of wisdom of crowds, they also have votes to legitimate the results of a decision. We want to rely on voting to legitimate the outcome, but when we see medical marijuana at the top of, we can't.

[It took me a while to get this – that votes on could be regarded as being as real as votes cast in the ballot box, or in a town hall meeting. Before this, I'd assumed everyone thought of internet votes as not being worth the paper they were printed on. Some of it's probably a cultural issue, having grown up with compulsory voting (well, you had to turn up, what you did with your paper before you put it in the box was up to you). Of course it's different in the UK and the US.]

Three options for dealing with this – 1) when we get broad national participation in digital plebiscites, it would be like directing democracy into the veins of the country. He's not willing to legitimate medical marijuana as the choice of the American people, it's clearly just another pressure group.

2) treat it as a PR exercise. But if you cherry pick what you take seriously, you haven't altered the political [landscape?].

3) figure out how to drag this kind of participation into the world of checks and balances. Isometric tension among competing interests. Can't currently say on the internet that are sure that everyone voted and that those votes around counted accurately, then can't yet integrate into democratic processes.

These tools don't gradually spread into a culture, they gradually spread under radar then are cemented in a time of crisis [?]. Some things (the kinds of surveillance) we relied on journalists for have been remanded to camera-phone carrying public.

2009 [will see] considerable integration, adoption, of these tools around the fact that people are groping for new models… and processes? Look back on a time when made decisions considerably more momentous [than medical marijuana?]

How do we open this up to new participation without opening up to dramatic system gaming or significant problems of delegitimisation?

The end!

Qu: what happens to organisations in this time of change? When you only have a choice of hiring old style command-and-control journalists but want to change.
Ans: depends on organisation. Laws on Obama's desk to sign are published online five days before they're signed, five long days with press hounding you if you're trying to sneak through a 'bridge to nowhere' bill. The risk is that Americans, not having much knowledge of government, won't know where to put the blame if they see a bill they don't like. Pressure should be put back onto Congress but risk is that Obama will be blamed for signing bills people don't like. Lesson is – there's no way to change institutions in a low stakes way.

We hope Obama goes the whole hog and adopts the 'fail fast, learn more' model rather than betting whole farm on single institutional change. Institutions are homeostatic. A lot of it is going to be about process rather than personality.

Qu: should we be setting up a series of codes of practice for deployment of social media in public spaces [good question].
Ans: Essentially yes. The question is, to what degree should lawyers be involved? He's been doing research on internet and generosity, the effect of culture on that. The Invisible College – an attempt in England to internalise scientific practice. 'This is how you write end results, share them, how the conversation should go'. They went after alchemists for failing to be informative when they were wrong, it was ok to be wrong but they wanted them to think about and share what went wrong. It wasn't a government thing but the benefit to England has been extraordinary. Today's event flyer said 'we encourage you to take pictures, etc', but if went to lawyer, got a series of waivers, processes, announcements, etc – would make less progress in slower time to smaller effect. Difference between institutions that encourage photo taking, blogging, etc, and those that don't, should start playing out.

Maybe we don't need to write down code of conduct, just make it a social norm at events with things like back of program. [But does that allow for negotiation of different needs? I have a friend who blogs semi-anonymously and photos at events would blow that for her. I just hate being photographed, and especially being photographed and tagged or otherwise identified – how does that work for me? I guess we're working it out on places like Facebook, where one tag-happy friend has gradually learnt that we'll all freeze her out when she approaches with her camera because no-one wants to spend the next morning un-tagging photos. But I can't un-tag a photo someone else has labelled on Flickr.]

Qu (Danny) – Reagan's war on drugs had significant effect in US, so why does medical marijuana delegitimate
Ans – in terms of issue, the war on drugs was catastrophe, would like to see it solved. But in it's the difference between the number of people who are interested and the degree to which a small group of people is interested. The history of democracy is figuring out how to balance relatively small, well organised groups with interests of large, relatively disorganised groups. I don't trust that in a general poll that result would be in top ten of concerns of American public. The intensity of people who believe in that issue, doesn't translate into 'this is the thing that legitimates the administration spending its time on that issue'.

Digg is rife with system gaming, but it doesn't matter because it's a self-contained media outlet. The benefit of market is that if you don't like Digg you can switch to another service. We can't switch governments. The things we use to legitimate stuff like Digg isn't the same as legitimating an internet plebiscite.

We're at the point where majority of people in highly developed countries have internet, but that doesn't rise to level of serious voting.

Qu: three options on medical marijuana. Healthy communities seem to be comfortable with having 'thousand pound gorilla' moderation – why can't take a leaf out of that (book)?
Ans: the mechanism that works best on internet and open source communities is 'benevolent dictatorship'. Linus Torvalds, even Jimmy Wales. Benevolent dictatorship works in internet environment and not in real world because of the threat of both switching and forking. Switching – benevolent dictatorship is mediated cos if your population doesn't like what you're doing, they can up sticks and move to another project. Or they take entirely of your project and start a completely new version (forking). It's only happened a few times but all benevolent dictators are aware of it.

You can't switch or fork real estate. People who are dissatisfied can't easily move. The things that keep open source projects working can't be trivially ported into real world environment. So everybody who lives in geographic range lives in particular regime – it's a different set of problems. Do-ocracy model (do more, get more benefit than people who just talk about code). The last step is bigger gap than imagined.

Qu: kinds of new models coming out of current crisis? Revolutionary new business models, give us a clue what's next?
Ans: he tries not to use word revolutionary. Linus Torvalds and Jimmy Wales's first message to world about wikipedia and linux were incredibly modest. No claims about altering the world, just 'give it a try'. Claims to revolution are orthogonal, inversely proportional to the likelihood of revolution.

With that caveat, he's watching logic of peer-to-peer networking apply to other things where there's a high degree of centrality and resource that actually exists at the edges. In a recession, most important thing with that characteristic is money. Mutualisation in US, re-mutualisation in Britain. – peer-to-peer lending. Suddenly have pool of people watching you to make sure doing right thing with whatever you borrowed the money for. People willing to go through that, emotional connection better guarantor than model of risk? But seen how other models of risk have played out. Body shutting that down have missed three huge things within their charter – the SEC is no good at detecting challenges within the status quo, but they're very good at detecting challenges to the status quo.

So to what degree will mutualisation happen? To what degree will the government get in or stay out of the way?

Qu: contrast between marijuana e.g. and myBO FISA bill issue?
Ans: this gets to concerns about When campaigning, Obama was answerable to supporters. When president, he's answerable to everyone, including people who aren't part of his community. The FISA telecomms bill was an internal argument that doesn't have national ramifications. Once you govern, you have to govern everybody – that's when legitimation concerns kick in.

Qu: [me, sounding like a complete dork. I hate asking questions in public.] In a post-Smithsonian 2.0, post-Digital Britain world, what messages for holders of cultural content, (e.g. museums, television stations), how we can engage with third, social sector, and generally, what are our responsibilities?

An: it's funny, in US, museums are more privatised. So he thinks less in terms of language of responsibility and more in terms of language of opportunity. Smithsonian on Flickr Commons. They were excited and astonished to see that people were saying 'hey I really like this photo' but also 'this is a mailbox from 1840, here's a link to additional material'. The curation of material, not just appreciation, was broadened and deepened by that [outreach? ]. And yet, of the total holdings of SI, they've only got c6000 images out. This is in part because curatorial imperative is challenged by exactly this. As evidence that this is not worth doing, people point to 'oh, that's a nice picture' comments and say it's insipid, ridiculous. 'Go stand next to someone in a museum sometime!' It's not that people are saying these things, it's just you can (now) hear them, and you're desperate for earplugs because the curators have not had to hear them.

A little like the analogy – if nothing new were invented new tomorrow, we have all the technology [technological mechanisms] we need to treat holders of cultural content not as just repositories but as conveyers. Of conversation, additional curation, of re-use – the framework is all there. Legal, technological. If I was running one of these institutions, I'd spend more time worrying about the institutional change than the technological platform.

Saw something that might spread – room where group of designers were watching a webcam that showed a user trying to use a system. Meetup make someone who works in the design department watch someone trying to use their system *every day*. Can you imagine if Microsoft did that? 'You can't work here very long without encountering an actual user'. If I wanted to change an institution in the direction of thinking of yourselves as a convenor as well as a repository, I'd work on ways to get the encounter between the public and the professionals to happen. Not in a big conference, but just – for fifteen minutes you have to go to the gallery and you have to talk to somebody about what they just saw. Easy technically, hard institutionally. That institutional transformation is going to be the next big platform. [or ? that's coming]

And that was the end of questions.

At the time, I made a note of Need to Know's old slogan: They've stolen our revolution, we're stealing it back, but now I can't remember why.

5 thoughts on “Clay Shirky on 'mass internet collaboration' at London's ICA”

  1. Quick little comment on Clay's Smithsonian comment; he was one of the keynote speakers at the Smithsonian 2.0 event in January. There was some misconception about number of digital Smithsonian objects; SI estimates 137 million physical objects; there are 1500 on Flickr, couple thousand more on some photographic sites; but probably a couple million total digital images online

  2. Thanks for the clarification. Are they mostly at

    I was hoping for an answer that incorporated any feedback and responses from the Smithsonian 2.0 event – the tweets had a lot of comments about 'curators are this, that', but I didn't get a sense of whether that was based on comments at the event or a general impression of what curators are like.

    But I think my question was too specific for the event audience, too much background would be necessary for a good answer.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.