***** To join INSNA, visit http://www.sfu.ca/~insna/ *****
I would certainly love to talk to anyone who is interested in analyzing
this data. I have a scraped version of 1/3 of Friendster that i'm using
for analysis and can run queries on Tribe's data with a purpose.
The problem with traditional analysis is that there are HUGE structural
differences between this data and what sociologists typically consider.
This must be embedded in the conversation and analysis.
I am hoping to work with Jeff Boase to do a homophily analysis of this
data based on the interests and demographic data. Unfortunately, temporal
link structure is not available so there's no way to determine when the
graph connected where (or when it disconnected). This is frustrating
because it would allow for meme spread analysis.
There are huge problems with doing a structural equivalance analysis.
First, people who simply know of each other (and would normally be deemed
structurally equivalent) link because they recognize one another.
Second, most of the time there is huge structural equivalence, there's a
reason why the people don't link. In virtually all of my interviews, it
can be explained by one thing: exes. They know each other quite well and
would be considered connected in any normal social netowrk. Thus, this
kind of analysis requires a recoding of the data through qualitative work.
I've definitely been looking at the different types of ties that are
emerging, but i doubt my musings are considered a discussion...
Each site definitely has a desired outcome. I don't know how on earth you
measure "successful dates" though (and, no, marriage is not a proper
answer). Tribe is certainly doing quant analysis based on their data.
And you can bet money that Spoke is as well.
Many minds would be great... but i doubt that you're going to see publicly
available data. And besides, one thing that i'm definitely finding
invaluable in any analysis of this is the qualitative component. Aside
from myself and the companies, i don't know of anyone with such data. If
you have any, please do let me know!
On Sat, 10 Jan 2004, Don Steiny wrote:
> ***** To join INSNA, visit http://www.sfu.ca/~insna/ *****
> With all the "social networking software", Freindster, LinkedIn, Ryze,
> Spoke, Visible Path and so on I am wondering:
> - Is there any definition of the outcomes that they are trying to
> achieve in any way that could be tested?
> - Is there any discussion of the types of tie and what constitutes
> a tie?
> - Is there any attempts to use network structure like structural
> equivalance to answer questions (based on the outcome above)?
> (Note: I talked to Reid and Johnathon, from LinkedIn and Friendster
> and the answer they gave was 'no.' Spoke and Visible Path are
> keeping it closer to the chest).
> - Will any of the data become available to researchers? (I think
> here of Ron Burt's reinterpretation of Coleman et al's data on
> diffusion - many minds ...).
> Don Steiny - Central Coast Angel Network - www.ccangels.net
> Institute for Social Network Analysis of the Economy - www.isnae.org
> 125 Mission St #3 - Santa Cruz, CA 95060 - 831.471.1671 - fax: 831.471.1670
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
- - - - - - - - - - d a n a h ( d o t ) o r g - - - - - - - - - -
taken out of context i must seem so strange
(ani difranco | danah.org/ani) .. musings ..
v-season: events.vday.org (misbehaving.net)
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.sfu.ca/~insna/). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.