Corante

Authors

Clay Shirky
( Archive | Home )

Liz Lawley
( Archive | Home )

Ross Mayfield
( Archive | Home )

Sébastien Paquet
( Archive | Home )

David Weinberger
( Archive | Home )

danah boyd
( Archive | Home )

Guest Authors
Site Search
Monthly Archives
Syndication
RSS 1.0
RSS 2.0
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

Many-to-Many

« One from the Old Skool: Godwin on ASCII, from 1994 | Main | Shannon Clark on Dean and Social Software »

February 2, 2004

Clear, precise...and problematic

Email This Entry

Posted by David Weinberger

Orkut embodies two of the weaknesses inherent in artificial social networks: it requires us to be clear and precise. Those are virtues when it comes to invoices and jury verdicts, but they are how real social networks are not built. The precision shows up in the digital choices we're given: Is Phil your friend or not? If he is, is he one-star, two-star or three-star sexy? Choices you are not given include: (i) Sort of sexy. (ii) Could be sexy if he dressed better. (iii) If I were a woman, I think I'd find him sort of sexy if I went for that type and if he dressed better. So, exactly how many stars does that work out to? Ah, but as several commenters on a previous blog entry pointed out, Orkut lets us write testimonials precisely to get around the over-precision of the yes-no rating system: We can write what we want and say what we can't say with 1-3 stars. But, while testimonials need not be precise, they do try to make explicit something important about a relationship. Sometimes, of course, that's exactly what we need to do. And, if the testimonial system is working for you, fine. For some people in some situations it's going to be exactly what they need, Nevertheless, you can only build a real social network by overcoming clarity and precision. Groups form by creating messy darkness. A team "bonds" as the relationships among the members become so tangly and ambiguous that the members can no longer sum one another up in a few words, much less by reference to their official roles. A mailing list becomes more than just a distribution channel when, over time, the participants learn enough about one another through the implicit body language of messages that their off-hand descriptions -- "She's a curmudgeon" "He's a total geek" -- feel inadequate. Our most important relationships -- our family, for example -- we can't fathom fully much less explain clearly. Groups become real through ambiguity, messiness, the implicit and the unspoken. We can be somewhat precise and somewhat explicit about these real relationships, but there's a price to pay: Any clear and explicit description I gave you of my daughter would obscure more than it showed, and would have an effect on my relationship with her if she were to read it here. Artificial Social Networks like Orkut get it backwards. They are built on explicit and precise declarations of relationship. Does this mean they're worthless and doomed? Not at all, although I personally am finding Orkut to be all maintenance and no value. Humans are so doggedly social (hmm, something wrong with that sentence!) that we take every instance of proximity as an opportunity for relationship, and we overcome every obstacle to find someone else to care about: A line for tickets becomes a nonce encounter group if the movie is sold out, and even prisoners in solitary will tap on the walls to talk with someone they may never see. (BTW, what exactly is the baud rate for cell-wall tapping?) So, connect millions of us by digital lines that are clear and precise, and we'll figure out some way to overcome the system's limitations and bring it into genuine sociality. Something will emerge. We just can't tell what yet.

Comments (4) + TrackBacks (0) | Category: guests


COMMENTS

1. Tim Keller on February 2, 2004 11:52 AM writes...

The first useful approximation is already here & implemented in a number of applications. Instead of front-loading explicit relationships and ratings, a much better way is to grow them naturally through implicit in-band data discovery, aka traffic analysis. In the security world we call it "sniffing". Mapping the network & measuring the strength of the connections is easy, we're already doing that(To: & From: lines in email for instance). Next comes figuring out the harder stuff, like reputation; that'll come when we add semantic analysis to the picture. Way down the road I'm guessing we'll add things like out-of-band cognitive analysis, reading our facial expressions as we read. And finally we'll end up with something that actually scans our brains and watches the patterns to see who we like & respect & who we don't. Horribly invasive, I know. But that's the path we're on, more or less. Obviously we'll need better security than we have now, & a lot more trust in each other as well. But ultimately the only way we'll really be able to get it right is to let the computer into our heads.

Tim

Permalink to Comment

2. Prentiss Riddle on February 2, 2004 2:55 PM writes...

Well... sort of. In fact our human preference for imprecision trumps orkut's rigid categories right out of the gate. How many of our orkut "friends" are friends, despite the screen asking us to confirm that Joe Bloe is "really your friend"? I don't know, say, Mark Canter but I find it hard to believe that he has 281 friends (not counting his offline ones).

I'm as guilty of this as anyone: I've accepted invitations to "friendship" from vague acquaintances-of-acquaintances just because I didn't want to offend anybody. But anyone wanting to use orkut with any precision would need to be aware that its model of social networks suffers from massive GIGO problems.

Permalink to Comment

3. Prentiss Riddle on February 2, 2004 2:58 PM writes...

(Oops -- I see that Mark Canter's prolific friendships have made him the subject of controversy and hence a bad example. Substitute anybody else with friendships in the three digits and my point is the same.)

Permalink to Comment

4. ZbigniewLukasiak on February 2, 2004 3:26 PM writes...

Seems like my old rant to make the Social Networking Software more implicite: http://zby.aster.net.pl/kwiki/kwiki.cgi?CritisizmOfExpliciteSocialNetworkingSoftware
In http://zby.aster.net.pl/kwiki/kwiki.cgi?SocialRouting
I propose a thinking framework for designing and analyzing implicite Social Networking Software.

Permalink to Comment

TRACKBACKS

TrackBack URL:
http://www.corante.com/cgi-bin/mt/teriore.fcgi/1373.

Listed below are links to weblogs that reference Clear, precise...and problematic:


EMAIL THIS ENTRY TO A FRIEND

Email this entry to:

Your email address:

Message (optional):




RELATED ENTRIES
Spolsky on Blog Comments: Scale matters
"The internet's output is data, but its product is freedom"
Andrew Keen: Rescuing 'Luddite' from the Luddites
knowledge access as a public good
viewing American class divisions through Facebook and MySpace
Gorman, redux: The Siren Song of the Internet
Mis-understanding Fred Wilson's 'Age and Entrepreneurship' argument
The Future Belongs to Those Who Take The Present For Granted: A return to Fred Wilson's "age question"