TITLE OF PAPER: What Groups Will Be URL OF PRESENTATION: PRESENTED BY: David Weinberger CONFERENCE: Emerging Tech DATE: 25th April 2003 LOCATION: Westin, Santa Clara -------------------------------------------------------------------------- REAL-TIME NOTES / ANNOTATIONS OF THE PAPER: {If you've contributed, add your name, e-mail & URL at the bottom} Learning from experience is the worst way to learn. -- Shirky // Explaining why groups make the same mistakes over and over again. Or maybe it's why do people refuse to learn from experience. Simple Definitions: Group - a set of people who know one another and know they're in a group. Groupings - groups of people that a somehow related, but not groups. Communities - a group in which the people care about each other more than they have to. Premises: Groups are high-value The net's really bad at them! Helps us make them Doesn't help us do anything with them "I don't know what groups I'm in" End-to-end principle Keep as many things as possible out of the centre Lots of room for innovation around the way groups are handled: Friendster ( http://www.friendster.com ) [My personal theory about Friendster is that it's a really detailed and elegant way of doing one tiny aspect of maintaining a community. It's like having the world's best login-process. Integrate this tech into other standard message boards and it could be really useful. TEC] Broad range of relationships get reduced to a binary choice. Interests / favourites People don't KNOW their interests and don't know what their favourites are. More importantly they don't necessarily want to show people what they actually enjoy. [this is what is cool about iTunesStatus... it shows your embarrasing but well-loved as well as your ultra-hip choices...] Generically more pessimisticc about unfocussed social networks rather than focussed ones like Friendster. We have the wrong model for the explicit and the implicit. "Making something explicit is most often an act of violence" Something that's implicit is inevitably contexual. We do violence to it when we lose that richness. Constitution building is a necessarily social act... Cannot separate the social physics from the technical physics. [Interesting stuff here - the structuring relationships are just too simplistic, perhaps? Is that what he's talking about? That the fake social physics in these environments aren't flexible enough - messy enough - primary enough?] And if we try, it's a painful thing - act of violence in making it explicit: [seeing the stage machinery - deus ex, lack of immersion, violent] Aboutness in search engines, reduction in ambiquity reduces utility (and serendipity) disambiguation is hard... messiness and confusion can be utlity - remove useful stuff by mistake. Knowledge Management: Problem with KM as social software is this failure to value this ambiguity. Grew out of two needs: 1) reduction in noisein the info Problem in identifying high-value info (treating it like a gem). Problem is that information is fully contextual - it loses identifiable value when removed FROM the noise. And computers are bad at grokking context, though humans are. Much misidentification. Example: Engineers discuss gaming, which is trash until company needs to develop graphics cards. Can't -preidentify- knowledge [The very distinction between knowledge and information isn't clear - some information looks like noise - looks annoying (kids with games) but becomes suddenly really useful when there's a need for graphic cards information and experience.] 2) loss of information through staff loss. If Marge were a database, you could extract. But human knowledge is implicit. It's hard to make the implicit knowledge explicit (technical support, writers and educators exemplify an ability to do this). Rough conclusion or model? Knowledge as piece of scrap that fits the hole in your tin roof... value of knowledge comes through context: computers very bad at this, humans are very good. Social Software and Emergence There is a connection between the two because social software allows the 'real' network to emerge. [Not sure I buy this, but maybe. TEC. Hydra is an example of the software letting the network emerge. I think it's more than that - I don't know any of you people!] The structures of the group also emerge? Social software Bof - difficult occasion by all accounts! [I buy the critique that nobody's willing to impose a structure (as it is 'violent' to a degree) having been in the room] Social network past one degree is not real knowledge. Social Software allows this second order network to emerge (Tom, did I get that right? I don't think so but I'm woolly in my mind here) Isn't he saying that this second order network isn't really your social network - you don't actually know your friends of friends until they're introduced to you. >> yes. That was what I heard. I didn't express that, however, very well. [however: value/power of weak links cf. buchanan, duncan watts, : grannovetter: "the strength of weak ties": http://citeseer.nj.nec.com/context/428117/0 most ppl get jobs through weak links etc.] [I think he's confusing emergence with the transition between implict to explicit] [REALLY HUGE SIDETRACK ON WASHING MACHINES] What will groups be? Tons of services from ecards to friendster. Maybe it's less about managing groups but managing different views or perceptions of ourselves and the relationship between ourselves and other people depend on those perspectives. Not about technology - social software - the tech is really easy [REALLY GOOD POINT THIS] [Dave Winer wrote this morning that social software is suffering a resurgence. Suffering. He's not entirely wrong - the spotlight's too fierce here. TEC] [Definitely a function of the selection process, which I [gf] was involved with, and that's an interesting meta-point.] So what now? Maybe we're coming out of our fascination with bits - binary thinking - maybe we're looking for more messy thinkings. [This is really anti the Social Software foundation (?) approach - who are all trying establish technical and programming bases for social software rather than trying to work out what's going on and what people think needs to be done. That's really badly written by the way I'm sorry if it doesn't explain itself very well. TEC] [people learning to love uncertainty... meta-theme... why messy interwingled webs are useful/humane cf. bookmarked-but-not-read-yet: http://www.edge.org/3rd_culture/gigerenzer03/gigerenzer_index.html SMART HEURISTICS: GERD GIGERENZER [3.31.03] "What interests me is the question of how humans learn to live with uncertainty. Before the scientific revolution determinism was a strong ideal. Religion brought about a denial of uncertainty, and many people knew that their kin or their race was exactly the one that God had favored. They also thought they were entitled to get rid of competing ideas and the people that propagated them. How does a society change from this condition into one in which we understand that there is this fundamental uncertainty? How do we avoid the illusion of certainty to produce the understanding that everything, whether it be a medical test or decidingon the best cure for a particular kind of cancer, has a fundamental element of uncertainty?"] Q from audience: tacit knowledge is the most succesfful form of knowledge management software, incorporating, for instance, email transmitting within a company. Meta data is crap? Then communication out of context is irrelevant. "Meta data is also crap when it is done explicity; particularly in context of building social network" Q: With reference to Netscan, isn't there a danger in dependency on ambiquous data (posting numbers not expressing individuals, etc) A: Dunno. That manner of metric is fascinating, but not the only way people make decisions. Ebay metrics are debased, yet people make some use of it as well. Q: Why isn't the promise of adhoc groups comiing together. Why isn't the future of that to have power rather than being buried in one database or another (be it database, democratic party, etc). Liberation from corporate business plans. A: Enthusiastic about threadsml because it may enable the conception of a mailing list as just a set of threads, rather than a set of members. This transfer of information is empowering (?). Q1: is agreeing to some set of rules relating to semantic web, if some basic classification is agreed to from inside social structure rather than imposed from without A1: Dunno. ------------------------------------------------------------------------- NOTES: Weird shortall - people aren't commenting in real-time any more. istha because the lag is so long now? Partly, I suspect - afraid of interleaving text again. We really need a read-only view for some people - and maybe even a people management process - it would be good to make it possible to block certain people or let some people n by default - particularly in large social situations like this. >> we should compile all this into a realworld case study for the hydra guys. Started writing my ideas (online) this AM. [www.kittyjoyce.com/eric/log/] >> commenting should be able to push out to iChat. why can't I take it there (and *koff* thread-it) - don't like the tying in to iChat... why rely on another application. (OK, I use Hydra, but not iChat). >> doesn't have to be iChat (but its there). But some way to bring these meta discussions into a new, but tied, space, would be nice. >> nod nod to more granular auth model proposed by Tom - yes, yes. >> also interested here in the role of a gardener of hydra - is that what my role is now I'm not notetaking? - Where do these documents live once the social network is broken? automatic WebDAV or other upload would be great to shared, predefined resources, persistent without intervention -- Someone should post (tom/eric?) all of the hydra notes to the etech wiki. btw, thx so much for all the note taking. There's a start here: http://trevor.smith.name/blog/archives/000108.html and trackback is your friend, as I don't have most of the notes from sessions that I didn't attend -- thx --- >> maybe we can meet up in the hallway or email to discuss next steps? Template --> feature ideas --> sharing? If nothing else, we should feed a lot of this to the Hydra creators so that they can open and share in these results. >> *nod* ----------------------------------------------------------------------- REFERENCES: {as documents / sites are referenced add them below} David P Reed. End to End Prinicple, Reed's Law: value of social network is number of groups which are able to form (steeper line than Metcalfe) (see also discusison in STLJ). ( http://www.reed.com/Papers/endtoend.html ) -------------------------------------------------------------------------- CONTRIBUTORS: {add your name, e-mail address and URL below} Eric Sinclair | esinclai@pobox.com | http://www.kittyjoyce.com/eric/log/ Tom Coates | tom@plasticbag.org | http://www.plasticbag.org Richard Gayle | richard_gayle@excite.com | http://www.corante.com/livingcode/ matt jones | matt@blackbeltjones.com | http://www.blackbeltjones.com Trevor F Smith | tfsmith@parc.com | http://trevor.smith.name/ -------------------------------------------------------------------------- E-MAIL BOUNCEBACK: {add your e-mail address separated by commas } jmay@pobox.com, tom@plasticbag.org, richard_gayle@excite.com, esinclai@pobox.com, phil@gyford.com, allen@hutchison.org, mgraham@mail.ivillage.com, matt@haughey.com, tim@lauer.com, matt@blackbeltjones, dan@inmyexperience.com, glenn@glennf.com -------------------------------------------------------------------------- NOTES ON / KEY TO THIS TEMPLATE: A headline (like a field in a database) will be CAPITALISED This differentiates from the text that follows A variable that you can change will be surrounded by _underscores_ Spaces in variables are also replaced with under_scores This allows people to select the whole variable with a simple double-click A tool-tip is lower case and surrounded by {curly brackets / parentheses} These supply helpful contextual information. -------------------------------------------------------------------------- Copyright shared between all the participants unless otherwise stated...