Tag Archives: social-networking

How Trends Actually Spread; or, Six Degrees but No Connectors

The small sample size of Stan­ley Mil­gram’s small world exper­i­ment means that the the­ory of ‘six degrees of sep­ar­a­tion’ and the con­clu­sion drawn from it–primarily, the Influ­en­tial’s the­ory pop­ular­ised by Mal­colm Glad­well in The Tip­ping Point–could be deeply flawed. That was the start­ing point for Duncan Watts’ research that led him to say “the Tip­ping Point is toast”.

So to research how ideas and trends spread vir­ally, Watts (who is author of Everything is Obvi­ous, prin­cip­al research sci­ent­ist at Yahoo! Research (he dir­ects their Human Social Dynam­ics group), and found­ing dir­ect­or of Columbia Uni­versity’s Col­lect­ive Dynam­ics Group) ran large-scale repro­duc­tions of the small world exper­i­ment and hun­dreds of com­puter sim­u­la­tions that brought for­ward two con­clu­sions: the six degrees of sep­ar­a­tion the­ory is cor­rect, but there is no evid­ence for super-con­nec­ted ‘trend gate­keep­ers’ (such as Glad­well­’s ‘Con­nect­ors’):

But Watts, for one, did­n’t think the gate­keep­er mod­el was true. It cer­tainly did­n’t match what he’d found study­ing net­works. So he decided to test it in the real world by remount­ing the Mil­gram exper­i­ment on a massive scale. In 2001, Watts used a Web site to recruit about 61,000 people, then asked them to ferry mes­sages to 18 tar­gets world­wide. Sure enough, he found that Mil­gram was right: The aver­age length of the chain was roughly six links. But when he examined these path­ways, he found that “hubs”–highly con­nec­ted people–weren’t cru­cial. Sure, they exis­ted. But only 5% of the email mes­sages passed through one of these super­con­nect­ors. The rest of the mes­sages moved through soci­ety in much more demo­crat­ic paths, zip­ping from one weakly con­nec­ted indi­vidu­al to anoth­er, until they arrived at the tar­get. […]

[His com­puter sim­u­la­tion] res­ults were deeply coun­ter­in­tu­it­ive. The exper­i­ment did pro­duce sev­er­al hun­dred soci­ety­wide infec­tions. But in the large major­ity of cases, the cas­cade began with an aver­age Joe (although in cases where an Influ­en­tial touched off the trend, it spread much fur­ther). To stack the deck in favor of Influ­en­tials, Watts changed the sim­u­la­tion, mak­ing them 10 times more con­nec­ted. Now they could infect 40 times more people than the aver­age cit­izen (and again, when they kicked off a cas­cade, it was sub­stan­tially lar­ger). But the rank-and-file cit­izen was still far more likely to start a con­ta­gion.

I can­’t help but find it some­what iron­ic that, writ­ten almost four years ago, this argu­ment has­n’t really gained much trac­tion and Glad­well­’s ideas are still dis­cussed ad nauseam.

Inventive Ways to Control Trolls

To keep the peace on the ever-expand­ing Stack Exchange Net­work of online com­munit­ies, own­ers Joel Spol­sky and Jeff Atwood intro­duced the timed sus­pen­sion of dis­rupt­ive users’ accounts. Over time the trans­par­ency of the timed sus­pen­sion pro­cess proved to be occa­sion­ally inef­fi­cient when dis­cus­sions arose regard­ing the mer­its of cer­tain sus­pen­sions. This led the admin­is­trat­ors of the com­munit­ies to invest­ig­ate oth­er ways of mod­er­at­ing prob­lem­at­ic users.

What they found were three fant­ast­ic­ally devi­ous secret ways to effect­ively con­trol trolls and oth­er abus­ive users on online com­munit­ies: the hell­ban, slow­ban, and errorb­an:

A hell­banned user is invis­ible to all oth­er users, but cru­cially, not him­self. From their per­spect­ive, they are par­ti­cip­at­ing nor­mally in the com­munity but nobody ever responds to them. They can no longer dis­rupt the com­munity because they are effect­ively a ghost. It’s a clev­er way of enfor­cing the “don’t feed the troll” rule in the com­munity. When noth­ing they post ever gets a response, a hell­banned user is likely to get bored or frus­trated and leave. I believe it, too; if I learned any­thing from read­ing The Great Brain as a child, it’s that the silent treat­ment is the cruelest pun­ish­ment of them all. […]

(There is one addi­tion­al form of hell­ban­ning that I feel com­pelled to men­tion because it is par­tic­u­larly cruel – when hell­banned users can see only them­selves and oth­er hell­banned users. Brrr. I’m pretty sure Dante wrote a chapter about that, some­where.)

A slow­banned user has delays for­cibly intro­duced into every page they vis­it. From their per­spect­ive, your site has just got­ten ter­ribly, hor­ribly slow. And stays that way. They can hardly dis­rupt the com­munity when they’re strug­gling to get web pages to load. There’s also sci­ence behind this one, because per research from Google and Amazon, every page load delay dir­ectly reduces par­ti­cip­a­tion. Get slow enough, for long enough, and a slow­banned user is likely to seek out green­er and speedi­er pas­tures else­where on the inter­net.

An errorb­anned user has errors inser­ted at ran­dom into pages they vis­it. You might con­sider this a more severe exten­sion of slow­ban­ning – instead of pages load­ing slowly, they might not load at all, return cryptic HTTP errors, return the wrong page alto­geth­er, fail to load key depend­en­cies like JavaS­cript and images and CSS, and so forth. I’m sure your devi­ous little brains can ima­gine dozens of ways things could go “wrong” for an errorb­anned user. This one is a bit more eso­ter­ic, but it isn’t the­or­et­ic­al; an exist­ing imple­ment­a­tion exists in the form of the Drupal Misery mod­ule.

A Summary of Happiness Research

Dav­id Brooks brings ‘hap­pi­ness research’ back to the wider pub­lic’s atten­tion with a suc­cinct sum­mary of research into what does and does not make us happy:

Would you exchange a tre­mend­ous pro­fes­sion­al tri­umph for a severe per­son­al blow? […]

If you had to take more than three seconds to think about this ques­tion, you are abso­lutely crazy. Mar­it­al hap­pi­ness is far more import­ant than any­thing else in determ­in­ing per­son­al well-being. If you have a suc­cess­ful mar­riage, it does­n’t mat­ter how many pro­fes­sion­al set­backs you endure, you will be reas­on­ably happy. If you have an unsuc­cess­ful mar­riage, it does­n’t mat­ter how many career tri­umphs you record, you will remain sig­ni­fic­antly unful­filled.

Brooks goes on to look at the con­fus­ing cor­rel­a­tions between hap­pi­ness and wealth before dis­cuss­ing the wider “cor­res­pond­ence between per­son­al rela­tion­ships and hap­pi­ness”:

The daily activ­it­ies most asso­ci­ated with hap­pi­ness are sex, social­iz­ing after work and hav­ing din­ner with oth­ers. The daily activ­ity most injur­i­ous to hap­pi­ness is com­mut­ing. Accord­ing to one study, join­ing a group that meets even just once a month pro­duces the same hap­pi­ness gain as doub­ling your income. Accord­ing to anoth­er, being mar­ried pro­duces a psych­ic gain equi­val­ent to more than $100,000 a year.

If you want to find a good place to live, just ask people if they trust their neigh­bors. Levels of social trust vary enorm­ously, but coun­tries with high social trust have hap­pi­er people, bet­ter health, more effi­cient gov­ern­ment, more eco­nom­ic growth, and less fear of crime (regard­less of wheth­er actu­al crime rates are increas­ing or decreas­ing).

via Fred Wilson

I dis­cussed the ‘com­muters para­dox’ last year, not­ing that “a per­son with a one-hour com­mute has to earn 40 per­cent more money to be as sat­is­fied with life as someone who walks to the office”.

Social Networks and Their Far-Reaching Influence

In a short and bal­anced review of Con­nec­ted–“a sci­entif­ic look at the ties that bind us together”–we are treated to some inter­est­ing find­ings on social net­works and their myri­ad extern­al effects–including how far these effects ‘travel’ through said net­works.

Con­trolling for envir­on­ment­al factors and the tend­ency of birds of a feath­er to flock togeth­er […] Chris­ta­kis and Fowl­er found that we really do emu­late those we care about, wheth­er we mean to or not. Being con­nec­ted to a happy per­son, for instance, makes you 15 per­cent more likely to be happy your­self. “And the spread of hap­pi­ness does­n’t stop there,” they note. It radi­ates out for three degrees of sep­ar­a­tion, so that, say, your sis­ter­’s best friend’s hus­band’s mood exerts a great­er influ­ence on your per­son­al hap­pi­ness than an extra $10,000 in income would. If he gains 50 pounds, it will be that much harder for you to stay slim, as the frame of ref­er­ence for what’s “nor­mal” changes through your net­work. Or, on the pos­it­ive side, if he quits smoking, your chances of kick­ing the habit improve, too, even if you’ve nev­er met him. […]

Pub­lic health work­ers can more effect­ively stop the spread of sexu­ally trans­mit­ted dis­eases if they know what kind of net­work they’re deal­ing with: a hub and spoke (e.g., a pros­ti­tute with many cli­ents) or a more trans­it­ive “ring” net­work where people have few part­ners, but many of these part­ners over­lap (which could hap­pen at a small high school). On anoth­er front, they point out that vot­ing makes little sense for an individual—one vote nev­er decides an election—but is far more ration­al in a net­work con­text. As with hap­pi­ness and obesity, the decision to vote has reper­cus­sions through three degrees of con­nec­tions. […] Since lib­er­als and con­ser­vat­ives tend to form their own social net­works, this means that your decision to vote can increase the like­li­hood of hun­dreds of oth­er people vot­ing for the same can­did­ate.

I do won­der if these degrees of sep­ar­a­tion that exert influ­ence on us fluc­tu­ate with the size of each ‘degree’?

Making Applications Viral, Without Spam

Vir­al­ity isn’t an indis­pens­able fea­ture of all suc­cess­ful applic­a­tions, but for those where it can be hugely bene­fi­cial there are four core prin­ciples that help the vir­al­ity of an applic­a­tion, says Daniel Tan­ner:

  • Invit­a­tion should be a core pro­cess, that is essen­tial to using the applic­a­tion – this will max­im­ise the chances that your users do invite new users.
  • Keep pulling people back in, rather than let­ting them for­get you after the ini­tial invit­a­tion, and make this “remind­er” pro­cess also be cent­ral to the use of the applic­a­tion.
  • Be use­ful even to the lone user, because that lone user is the source of all your oth­er users.
  • Remove arti­fi­cial invit­a­tion lim­its, to recog­nise the real­ity that most invit­a­tions come from a few very act­ive users, and help those users spread the word.

Ten­ner also notes–in passing–the concept of the vir­al loop. Andrew Chen’s take on the loop is the best I’ve read on the top­ic.