Wolfram Alpha Drills Deep Into Facebook Data 70
Nerval's Lobster writes "Back in January, when Wolfram Alpha launched an updated version of its Personal Analytics for Facebook module, the self-billed 'computational knowledge engine' asked users to contribute their detailed Facebook data for research purposes. The researchers at Wolfram Alpha, having crunched all that information, are now offering some data on how users interact with Facebook. For starters, the median number of 'friends' is 342, with the average number of friends peaking for those in their late teens before declining at a steady rate. Younger people also have a tendency to largely add Facebook friends around their own age — for example, someone who's 20 might have lots of friends in the twenty-something range, and comparatively few in other decades of life—while middle-aged people tend to have friends across the age spectrum. Beyond that, the Wolfram Alpha blog offers up some interesting information about friend counts (and 'friend of friend' counts), how friends' networks tend to 'cluster' around life events such as school and sports teams, and even how peoples' postings tend to evolve as they get older — as people age, for example, they tend to talk less about video games and more about politics. 'It feels like we're starting to be able to train a serious "computational telescope" on the "social universe,"' the blog concluded. 'And it's letting us discover all sorts of phenomena.'"
True Wisdom (Score:5, Interesting)
Wisdom of the crowd. [wikipedia.org]
Nick Kolakowski Strikes Again! (Score:5, Interesting)
That's not to say that the data is without merit or interest. The issue here is that Slashdot's publication of the April 24 post on Wolfram's blog [stephenwolfram.com] had to wait until after Kolakowski had offered his summary of it on April 26 [slashdot.org]. Why did slashdot readers have to wait a few days for Kolakowski to write his own summary of the blog posting? What value did he add?
Re:Politics or Video Games. (Score:5, Interesting)
This is why: In the 1930's Adolf Hitler was ELECTED into government. A couple of years later he invaded a lot of countries, including mine. The nazi's took control of all the files the government had on its citizens. People who had (for example) Rosenbaum or Levi as a family name were 'visited'. Police files on 'crimes' like homosexuality were examined as well and although the original government wasn't actively prosecuting gay people, the nazi's turned out to be slightly more active in that regard. People who had checked the box 'Jehova's Witness' also got to stare down the business end of a rifle. And the list goes on. All straight out of the paper files, with compliments of the former government.
If the nazi's had FB tough, they would have their hands on far, far, far more explicit and far more detailed information, searchable with a mouse click. And people provide those bits of information without hesitation, without complaining and out of free will. The idiots!
And for anyone thinking... nah, that 1940's business would never happen again... You are probably among the first one's rounded up.
Also, politics may change. What is legal now, might not be legal tomorrow (because the elected government puts new laws in place), and the elected government will set their constitutional instruments (aka police, intelligence agencies) on FB to monitor offenders.
Social media and politicians are as dangerous as a box of nitroglycerin. In a roller coaster. Doing 100mph. On square wheels...
No thank you, not for me!
Re:True Democracy (Score:5, Interesting)
Replace all corrupted clowns chosen by rigged popularity contests with math. Math can be trusted.
Lies, damned lies, and statistics.
I talk about open data, not manipulated results of undisclosed calculations.
The parent poster said exactly what I was going to.
If you think that the famous comment "lies, damned lies and statistics" refers only to concealed and rigged calculations, then you're wrong.
But back to what you said originally; your "math can be trusted" comment struck me as blinkered idealism meets dangerously inexperienced naivity. It's quite possible (and common) to use and abuse publicly-available data- via technically-correct use of statistics- to argue many different cases. Unemployment has gone down? Yes, because the definition of "unemployed" has been changed, and people (say) choosing to accept another benefit are no longer counted as unemployed. Or maybe lots of people have part time jobs for five hours a week, but they're not "unemployed".
Maths is perfect, in itself? Yes, we know that. It doesn't solve all the problems if you're proposing running the real world that way. Opinion is not mathematical. How exactly do you propose to correlate and translate people's opinions into a mathematical system? That in itself is subject to interpretation of some sort, whether via manual means or automatic.
Wolfram's system might be useful and insightful, but how do you translate that into democratic representation? And politicians- or their equivalents- will find a way to manipulate people's perceived or stated opinions if it suits them. And they'll find a way to manipulate the opinion-based system.
Maybe it's realised that by convincing a small number of people to feel very strongly about something, or talk about it more, that their opinion carries more weight. Let's radicalise some people and/or foster extremist opinion and behaviour to achieve our ends.
So, the current opinion-translated-into-pseudo-maths system is being manipulated and needs changed? Who decides if it's to be changed? The people? Ah... the measure of the people's opinion says that they don't want the system changed. Of course, that "measure" is via the current, manipulable system, so it supports itself. Unfortunately.
Democracy and voting are- to some extent- already attempts to translate opinion into a solid mathematical representation. The vote is- essentially- a translation of an opinion into a mathematical entity. Of course, it's not perfect, but at least the person making it gets to decide how their opinion translates.
Your system- abandoning voting for more "direct" means- would actually be less direct, because it would be imposing someone else's chosen interpretation of that person's opinion. Well, I say "your system", but actually, you didn't propose a system at all beyond using mathematics. Which we're doing anyway; the hard bit is choosing the most appropriate mathematical system to represent people as a whole, and translating people's opinions into input for that system.