Advertise here with Carbon Ads

This site is made possible by member support. ❤️

Big thanks to Arcustech for hosting the site and offering amazing tech support.

When you buy through links on kottke.org, I may earn an affiliate commission. Thanks for supporting the site!

kottke.org. home of fine hypertext products since 1998.

🍔  💀  📸  😭  🕳️  🤠  🎬  🥔

Clay Shirky on Communities, Audiences, and Scale

“Prior to the internet, the I-thou quality of mass media could be ascribed to technical limits — TV had a one-way relationship to its audience because TV was a one-way medium. The growth of two-way media, however, shows that the audience pattern re-establishes itself in one way or another.” So says Clay Shirky in his latest essay, Communities, Audiences, and Scale, and I think he’s on to something fundamental here.

At the same time, I do think there’s a middle ground that is quite useful as well, with small communities of people communicating with each other, not one-to-many or even many-to-many in the traditional sense (many individuals interacting with each other), but group-to-group. Groups don’t communicate with each other nearly as effectively as individuals do, but having those larger nodes can help stimulate keeping people in contact with each other while handling broadcast duties as well.

Reader comments

Mike DApr 04, 2002 at 7:25PM

Between networks and hierachies, the best solutions do seem to tend towards a mixture; like a network of heirachies, rather than either extreme.

I did an alife project a few years ago involving creatures controlled by networks (not neural, more like boolean networks with more functions available) generated from mutating strings. In the very few simulations that successfully developed a persistant species, the network was almost always divided into two or three smaller, interconnected networks (lobes!).

In a pure network, a mutation could have changed its operation completely; seperating into units might be the best way to dodge that. Nested networks.

nickApr 04, 2002 at 10:07PM

I actually responded to Clay with this, but I'll share here: a good example of the 'audience effect' was the 'cloudmakers' community last year following the AI game puzzles. While most people managed to make the early links themselves, which was lots of fun, it soon formalised itself into an established community (the Yahoo group and related sites) which knocked off the problems at double-quick pace. Turning many people who happened to be in a different time-zone, or outside the inner sanctum -- the chatroom -- into spectators, following the game like someone with a copy of Tomb Raider and a walkthrough. In fact, it was worse than that, because the game-writers had to make the puzzles harder to solve entirely because of the community, making an individual progress through the puzzles nigh-on impossible. For me, a lot of the fun went out of the game because of that 'audience effect'. And it reminded me that no matter what, you'll get the formation of hierarchies, which in the environment of the Web, can be even more exclusionary than in normal social interactions.

(I think this can perhaps be extended to Daypop, Blogdex and even Google, where there's an amorphous but undeniable hierarchy of leaders and followers, who in turn are leaders and have followers, with the effect that you have a rather Reaganesque trickle-down.)

Perhaps there's almost an uncertainty principle about inpromptu web communities: you can't be simultaneously made known and unaffected by that making-known. There's certainly tangible effects -- server load, Slashdot effect, calls from the New Yorker -- but there's also the intangible effect of knowing that you're being watched, and adapting to suit your watchers. And perhaps that's a pity.

jkottkeApr 05, 2002 at 1:13AM

(I think this can perhaps be extended to Daypop, Blogdex and even Google, where there's an amorphous but undeniable hierarchy of leaders and followers, who in turn are leaders and have followers, with the effect that you have a rather Reaganesque trickle-down.)

Heh. This is a bit off-topic, but I love how Dave crows about some thing that he links to being ranked highly on Daypop, like people are linking to it because it's good and not because his site is read by a crapload of Userland brown-nosers. Popularity != quality (as is often demonstrated on this very site).

there's also the intangible effect of knowing that you're being watched, and adapting to suit your watchers. And perhaps that's a pity.

It is a pity, and I envy those that can ignore that watching...or turn it into something good. A fair number of people read this site on a daily basis, and I've definitely found myself writing for them instead of for me. Lately I've been making a conscious effort to write as much for me as I can, and if people want to tag along, that's great....and even better if they leave great comments on the site.

jkottkeApr 05, 2002 at 1:20AM

Clay's argument has interesting implications for MeFi, as its membership is right around the magic 10,000 user number that he mentions. You can definitely see the audience and community aspects struggling against each other there. It's too big to be a tight community, but too community-oriented to be really effective as a broadcasting mechanism. To Matt's (and every other MeFi member's) credit, it's remarkable that it functions as well as it does without having to resort to the technological solutions of a site like /.

richard vacapintaApr 05, 2002 at 2:30AM

A contrarian comment:

Clay Shirky's footnote about the 150 limit in the size of human communities anticipates what I was also thinking about. Can this issue be solved by mere mathematical optimization methods? There is a natural rhythm and organization to human communities. We communicate warily with many and incessantly with only a few. A tightly-bound community has a common sense of purpose - its members know each other and their capabilities.

I think the word 'community' has been over-used and mis-used in the world of the web. Kottke.org and Metafilter.org are not communities, they are more like cities with a presiding mayor and a few city supervisors. Mayor Brown of San Francisco, for example, issues proclamations and sometimes elicits the opinions of his citizens but he is not the head of a "community". He is unaware of the doings of the artists and young skateboarders and the sewing clubs and the park preservationists and the goths and the VC clubs and the book restorers and the reggae musicians and the activist cyclists except when they naturally come to his attention. San Francisco is not a community - it is a network of communities with overlapping members, a hybrid organism, a network of networks.

Televison was the true anomaly, with its power to enforce cultural norms across wide swaths of society. The web, with its litany of voices, some louder than others, is I think a return to the egalitarian babble of open societies. Look around, at your friends and your friend's friends and who they know and who they listen to. This hierarchy of networks has been with us for a long time and I have found no compelling argument as to why the web has changed that. The phone is like conversation, only faster. The web is like a town hall meeting - only there are more of them and they are faster.

jkottkeApr 05, 2002 at 11:19AM

This hierarchy of networks has been with us for a long time and I have found no compelling argument as to why the web has changed that.

I think you and Clay are in agreement here. His argument is that the Web doesn't change anything with regard to community and audience. It's not the medium (TV, radio, Web, etc.) that matters, it's the size of the group. Small groups can become communities, and large groups have little choice but to become audiences, regardless of environment.

I think the word 'community' has been over-used and mis-used in the world of the web.

Agreed. Clay does define for the purposes of this essay in the first footnote: "For this analysis, community is used as a term of art to refer to groups whose members actively communicate with one another."

Kottke.org and Metafilter.org are not communities, they are more like cities with a presiding mayor and a few city supervisors.

I agree that kottke.org isn't a community...it's more me shouting at you with occasional discussion. MeFi used to be a community (when Matt was not really regarded as the mayor of MeFiVille, just another member) when it was smaller, but it's grown into more of the audience stage these days, with a few internal communities of people who know each other pretty well, outside the influence of Matt-as-mayor.

Michael HibbsApr 12, 2002 at 3:55AM

Many small groups are now harnassing the possibilities that the web offers. For example I have helped set up to flourishing chat groups with the help of Yahoo!Groups and a PHP bulletin board.

The most successful 'grouper' to me is MSN Messenger which allows many people conversations.

This thread is closed to new comments. Thanks to everyone who responded.