[Air-l] Social networks research (was Definitions)

[John Veitch]

Thank you Michael for the link below, and thank you Danah Boyd for the 
effort in putting together the list of research work being done.

For an old guy like me this is unsettling. If people want to do research 
work on a network, places like Ryze, Ecademy, OpenBC and LinkedIn are 
stable, and have a long track record of people being active with 
continuous memberships. You can quite easily track the posts of 
individuals over several years on Ryze, Ecademy and OpenBC.

Thank you all for the hint. I've just joined Facebook. Apparently that's 
where the action is. However the way Facebook is set up it's going to 
very difficult to get any groups running where useful conversations can 
happen. I need to do a lot more exploring there. Anyone with any hint's 
might send me private mail.

I'm also a member of MySpace. That's a site where people who have 
serious purposes have actually built some very interesting pages. I've 
been spammed with porn since joining MySpace. That could never happen of 
the four networks I first mentioned. The mixture of people with first 
experimental web sites, people trying to sell something and people who 
want to communicate and converse on MySpace is intensely interesting.

Here is a link to a letter I wrote on Ryze about the observed changes in 
people there.
http://www.ryze.com/posttopic.php?topicid=738706&confid=1031

However the most interesting thing about these networks to me, is the 
failure of most of the people who join to do ANYTHING at all. As I said 
in a previous letter, my own small survey indicated that most people 
never join any networks or groups. Of those who do, (I use Ryze as an 
example) +50% of people achieve in one year what the best 10% achieve in 
one week. The failure rate, often referred to as the long tail is at 
least 80% of all members. After that depending on your criteria there is 
some success but only obvious and sometimes spectacular success in the 
top 6% or so.

You can see the same pattern everywhere.

I imagined that once people got online, that they would soon learn about 
the internet and that they would develop both knowledge and skills at 
their own pace to become self educating. A tiny number do. That's 
probably true of everyone here.

But the vast majority of people don't get it. For me the clue was spam. 
The average person in my group hardly ever, and sometimes has NEVER seen 
any spam. (3 years ago I was getting 100 plus a day.) One man who had 
been on the Internet for 5 years asked, "What's this spam, people talk 
about?" In 5 years he had exchanged mail with less than 30 people.

I found the video's on YouTube by searching for "OWD" very interesting. 
As a market researcher I can tell that the sample was not a random 
sample. The two people who were much like some of the people I've 
interviewed were remarkable for what they didn't say. The questions 
asked expected them to have knowledge (biased), and they tried to 
oblige. The assumptions behind the questions didn't expose their real 
behaviour.  

Choose some people at random. You'll only need 5 or 10 to see the 
picture. Don't say or ask anything, just watch what they do at their own 
computers in the beginning. Then ask what else they regularly do? 
Finally do some counting, emails a day, is the email personal of list 
mail or spam, look at the history file on the browser. How much activity 
is there?  People say "Gee the Internet is great!" But when you look at 
what they really do, almost nothing, you'll see why there's such a long 
tail of failure on the social networking sites.  The hype and the facts 
don't match.

Researchers are so keen to demonstrate "success" that they set up the 
research so the failure is masked. Of course the easiest way to do that 
is to start with a biased sample.

Regards
John




Michael T Zimmer wrote:
> Hi - danah boyd has collected a list of much of this work:
> http://www.zephoria.org/thoughts/archives/2006/08/19/research_on_soc.html
> (and most of them are on this list, I suspect)
>
> -michael
>
> -----
> Michael T. Zimmer