![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Entry tags:
six degrees
About a year and a half ago I participated in a "six degrees" experiment that relied on email connections, and I asked to be notified when they had results. They just published those results (alternate link).
(Edit: Um, the email I received said they just published them, but I've just now noticed a rather older date on the linked article. I'm confused.)
They postulate that, unlike in some other social networks, "hubs" (people who know lots of people) are not a significant factor. While they talk a lot about drop-off from people lacking incentive to continue messaage chains, they don't seem to talk much the decision to use a hub (or not). In retrospect, I don't remember providing any data about negative decisions I made. They collected information about the people I did choose ("how do you know this person and how well?"), but they didn't ask "who did you decide against sending this to?".
I know a few people who I consider to be social hubs. I deliberately did not send all my message chains through them, because I figured that if they wanted to participate in such a study, they'd sign up for the study and start their own chains. So for any given hub-like person (who I thought inclined to participate in the first place), I sent one or two messages and then stopped. I had a total of eleven targets to reach, so I did not rely on those hubs. Mind, I also did not succeed in reaching a single target.
In other words, I was influenced by the meta-data, that this was an experiment and that I was trying to reach a bunch of different people. Also, that I wasn't personally invested in reaching these people; it was a fun game, not a matter of personal need. If there were a pressing need I would have tried the most expedient paths (using those hubs), but for a just-for-fun exercise I didn't want to bother people overly much.
(Edit: Um, the email I received said they just published them, but I've just now noticed a rather older date on the linked article. I'm confused.)
They postulate that, unlike in some other social networks, "hubs" (people who know lots of people) are not a significant factor. While they talk a lot about drop-off from people lacking incentive to continue messaage chains, they don't seem to talk much the decision to use a hub (or not). In retrospect, I don't remember providing any data about negative decisions I made. They collected information about the people I did choose ("how do you know this person and how well?"), but they didn't ask "who did you decide against sending this to?".
I know a few people who I consider to be social hubs. I deliberately did not send all my message chains through them, because I figured that if they wanted to participate in such a study, they'd sign up for the study and start their own chains. So for any given hub-like person (who I thought inclined to participate in the first place), I sent one or two messages and then stopped. I had a total of eleven targets to reach, so I did not rely on those hubs. Mind, I also did not succeed in reaching a single target.
In other words, I was influenced by the meta-data, that this was an experiment and that I was trying to reach a bunch of different people. Also, that I wasn't personally invested in reaching these people; it was a fun game, not a matter of personal need. If there were a pressing need I would have tried the most expedient paths (using those hubs), but for a just-for-fun exercise I didn't want to bother people overly much.
no subject
no subject
This apparently isn't a surprise to them; they also provided an alternate link. (It appears to be the same content, but I find their PDF difficult to read so I've made only a cursory insepection.)
no subject
no subject
They told their webmaster to give it to you. :) For a technical explanation (and the code to do it yourself!) simply type ".htaccess" into Google. :D
-- Dagonell, Webmaster Geek!
no subject
Well yeah, but it's the mechanism I wonder about. To the best of my knowledge I have never registered with Science, and certainly not since acquiring a new machine and corresponding new IP address. While I did follow a link from the researchers' site, I didn't log into that site first. So maybe they're checking IP addresses but not down to the individual machine and one of my coworkers is registered with them; I don't know. I'll test the link from home later.
no subject
-- Dagonell
no subject
another possibility
no subject
no subject
no subject
no subject
A quick search for AU: "Watts, DJ" reveals nothing more recent than ana similar article in PNAS in 2003. Maybe they've been accepted for publication somewhere else, and gave you their previous work? I'll keep an eye out and let you know. BTW, his articles on the subject have been advertised as "reprints available for the general public" but I'm still not sure how that works.