8nov2002 Social Software seminar [10.40, iSociety] INTRODUCTION one to look up - "social capital formation" robert putnum, bowling alone *making links that stop social exclusion* me-networks (my friends, my colleagues) social software for => social exclusion => social capital, that is capacity to earn, learn => businesses [the parallels with games/government/business are important, business is something else to chuck into gov/games talk] SHIRKY generations 1 early 70s, mail list, email. nonphysical group conversations 2 in 18 months: muds, moos, cb simulator (pre-irc), usenet 3 web then now initially web was one to many, publishing [the www is not true to itself. heh, this is tying in] [but that's just reflecting books, one to many] [what properties in the www cause this to happen?] collaborative COMMUNAL filtering id-switching, online culture -> shirky posits these aren't really around -> we don't spend our lives in moos [is real life id switching like small world? it's the opposite to abstraction and clustering] end of sparseness [online friends have to be new friends], and now ubiquity [friends are in all media] [this is like the im conversations in my front room] why was it wrong? assumption used to be that multimedia was the future -> but it's not [good enough, suboptimality.. not just that media isn't appropriate multimedia is still not RL, to be true to itself, nneds to understand semiology, like text not like tv. the future is radio] group adhesion, cohesion [i can't spell..]. [liquid bubbles] group behaviour like at parties that people don't leave groups different from sum of individuals judgements about the nature of membership scale is important [back to secret properties of technology. this is about understanding what the software OPERATES on: groups. they're different. so this isn't email and isn't the web, which is still one to one] [look, this is hyper literacy, many-to-many. maybe transhuman step 3 will be when multimedia is like RL, and we don't need radio style media that makes use of its suboptimality] metaphor of architecture: build environment where people can be some place together alternate: shipbuilding, between an environment and tool [well, yes. because people can't interact completely, the environment is just a space where people can't cohere into groups. implicit politics of software if you can't blink or turn away] => few-to-few [nice] [we can't build tools for people. that's like building properties of technology. need to find secret properties to let tools grow, but the only successful group is RL. hmm] somebody said, two groups (his name, byon): [FIND THIS] 1. sophisticated workgroup. convened for deliverables 2. basic group. secretly pursuing other goal. two types: 2a. it's there to help people find mates... ha. sex chat 2b. identifying and vilifying enemies. arguing 2c. religious worship, like star trek groups. uncritical conversation of a canon [HOW DOES THIS MAP ON TO THE GAMER TYPES, "clubs, spades, etc"] principal idea of building group structure is to protect groups from themselves, to keep soph. groups stay on track [like a language needs mechanisms to handle ambiguity. interesting!] 2 goals: - shipbuilding metaphor. half environment, half tool. measure progress, change course. software should help them, tell them where they are, if they've run aground. designed for a certain number of people [korean gamers, 80 people in a cafe. why do they HAVE to be in the same place?] - social constitutions. decision making, ways to agree on what decisions have been made, and to bind people to them RESPONSES 1 putnam vs barry welman how technology affects social capital new technology networks welman's conclusion: technology doesn't change much. people use tech to communicate with people they would anyway - your ability to exploit social/technological networks is your capital - social groups require closure and limits - tech groups require exchange and openness [in an absence of walls, a group is established by exchange bosons..] [how important is the front door at a party? could you have a party in the mall? is people not facing AT you just exchanging eye contact with each other?] - he's on about trust between strangers and reputation systems 2 - simon roberts, idea bazaar, anthropologist "imagining a new kinship of networks" - explicit social software (not implicit, like email) - "a long traidition of devices which allow us to think through, map and analyse social relationships" [hm. the media i consider, speak/books/number3... what's the analogue for groups not individuals] - political economy of information exhcange, reciprocity and flow (flows of meaning through social webs) - "as logic is to philosophy, kinship is to anthropology", nature of relatedness and organisation logic of a community [a group grammar of participation? do we have an inbuilt one, like chomsky's universal grammer? social language is as important as social.. what's hardcoded? how to investigate the switches, deduce the properties?] [IMPORTANT COMMON THEME: social grammer/language expressed in anthropology kinship, politics of social software, behaviour at parties.. same thing. implies it's hardcoded into the brain, like the spoken language thing. knowing these rules lets us establish a good, abstracted way of working in groups, like the study of rhetoric. doesn't even need chomsky! are there studies like this?] he says post industrial. - consociates, contemporaries, successors, predecessors: where are these 4 from? it's in his slides 3 - forum for the future [or something?] positive social change on the policy level move on from statistics to pulling in creative resources from people who WANT to participate [well, it should do for people who don't.. barrier to participate should be negative] [more thoughts: why should groups have to learn to operate with a social rhetoric? because it achieves things faster. if it's business, you want that. conversation is something independent. this is because conversation is PULL, but business has to be directed and production line, it has to be PUSH. the web conversation is pull. but directed groups have to be push, and that's a good thing] [it's all bloody semiotics. it all has to be signifier and signified in the brain UNTIL the technological can *be* RL. so acknowledge it. social *has* to be abstracted, directed, learn a language. make it easy, low barrier to entry, by having a grammar that is *like* the inbuilt one] 4 - matt jones, bbci/ican ecological thinking - shirky's term: social weather - architecture, big windows in social places [architecture as codified social rhetoric] - supermarkets, portable visual context. people looking in each other's baskets [THIS IS IMPORTANT barrier to entry to this visual conversation is *negative*, again. effort to covering the basket] - how can content not be textual? - "the bbc seeds conversations. how many conversations revolve around bbc news stories" [like the romanian workers one i sent round at work the other day, but that's only applicable to, say, 6 people. so it couldn't be a conversation on the news site, i have to say it to people, or shout it *with a known volume*] [you have to build a shout volume control in. that's part of social rhetoric, right there. RL has two: sound amplitude and geography] [another bit of grammar: the rule is we exchange meme bullets of sound. but you can break that rule in certain circumstances. social rhetoric is game-like rather than language-instinct rule-like] [a secret property of social software: you have to know what mode you're in so you can follow the correct rules] 5 - eCadamy [oh god, what a stupid name] - 14,000 members. what is it? - books, software and wine. or something - david weinberger, markets are conversations "love is the killer app", good book apparently - broadcast capitalism.. we're sick of that. moving to network capitalism. he says it's push vs pull. um - his idea: brands becoming causes. brands doing goodness, starting dialogue networks => nasty moneymaking ploy. don't like it [corporations can't create groups and control them. ugh] - he reckons we're moving from push to pull or pull to push, he's said both - networks stuck at a number of 10,000. hm. level 3? - his network is ruled by members, out of control [pull, in my parlance]. [it's poor environment rather than ships] - he's all about making money. and *actively* meeting people, deliberately! (like he read tipping point and did it deliberately) QUESTIONS - a group with a purpose is a team. shall we look at teams to see how they work? [that's interesting. hackman says clay. i say belbin roles, but that's only for a dozen people. what work has been done on teams of tribe size? ah, and clay mentions this: he's talking about big teams. we understand 5 as the ideal size for teams, but we need to know how to let larger groups work together] - communities based on geography (upmystreet conversations). why haven't people done this so far? can bootstrap the social rhetoric from the local shout volume. distance is a known volume control. - is society really "bowling alone"? fragmented. didn't hear the answer... - matt locke said ebay is transactional rather than being communitarian. [is this actually two ends of a spectrum, or maybe an axis, where another axis is team versus conversation? what about the term "public space"?] [okay, here's a sum-up. teams are push, they're directed, they need a grammar or politics. this has to align with an abstracted natural behaviour, so we need to investigate social rhetoric. we can't duplicate real life, that IS real life, but like radio/typography we have to figure out what bits to keep. architecture was an early stab. it overlays both gestures and social rhetoric.] - one question: says ebay *is* a social space like a market where somebody can say "oranges 20p a pound" but you can't ask about the weather - he's also saying, trust and reputation systems can be ends in themselves. are they part of the activity or meta to it? [in games like everquest, reputation is something people pay for in time, but it's not something real], ie gaming the trust system - more questions: 1- are networks actually a good thing, given people escaped them deliberately by moving from countryside to city 2- social exclusion, people without computers clay says to 2: using a cellphone in bangladesh, one person in a village has it, a personal piece of equipment becomes a social one anthropolist says to 1: networks are powerful social tools. but if people don't want to be in them, that has to be fine. [closing post offices made the barrier to entry to geographic communities positive rather than negative. that's the social breakpoint, tipping point] homogeniety [sp?] - clay: some people are in groups like echochambers where you here your own place. [groupthink] his question: can we design software that enforces heterogeniety of views? yes, encourage, but not enforce. there need to be incentives to encourge hetero.. so that when you climb out of your echochamber it doesn't dissolve into a flamewar. new question: people push their "me-brand" in a group. isn't this going to be inevitable? steve bowbrick: aren't these networks just meaning people who connect in the real world will dominate online? [me and tomc say to ourselves: no. the top end of the weblogger powerscale are instapundit, andrew sullivan, and yes they are, but they're connectors to the Real World. but in the middle scale there are connectors in *communities*, and they don't have to be connectors in the real world. weblogs allow people to come in.] clay says: all business groupwares are unsuccessful apart from email anthropologist: email corresponds with the organagram of business as network ==== rumblings: - welcome to the strongly connected centre [or whatever scc is] - computer.org - a meaningful slice of the www is similar to the www.. so why the blogdex problem? lowest tier *isn't* similar, because subgroups are just people - Problem Solving ML: needs to be autocatalytic write itself, write more psml needs counterfactuals needs to be pull not push, just expand, record everything that is hard to calculate and make it available as resource for other stuff so we need a good way to find, say, the list of all descriptions of all rss files, once something has calculated diameter of the web, like diameter of society max size of society, diameters, complexity exchange limit?