Tag Archives: facebook

In which I have a quarter-million friends of friends on Facebook

One of the privacy options Facebook allows is “restrict to friends of friends.”  I was discussing with Tom Scocca the question of how many people this actually amounts to.  FB doesn’t seem to offer an easy way to get a definitive accounting, so I decided to use the new Facebook Graph Search to make a quick and dirty estimate.  If you ask it to show you all the friends of your friends, it just tells you that there are more than 1000, but doesn’t supply an exact number.  If you want a count, you have to ask it something more specific, like “How many friends of my friends are named Constance?”

In my case, the answer is 25.

So what does that mean?  Well, according to the amazing NameVoyager, between 100 and 300 babies per million are named Constance, at least in the birthdate range that contains most of Facebook’s user base and, I expect, most of my friends-of-friends (herafter, FoFs) as well.  So under the assumption that my FoFs are as likely as the average American to be named Constance, there should be between 85,000 and 250,000 FoFs.

That assumption is massively unlikely, of course; name choices have strong correlations with geography, ethnicity, and socioeconomic thingamabobs.  But you can just do this redundantly to get a sense of what’s going on.  59 of my FoFs are named Marianne, a name whose frequency ranges from 150-300 parts per million; that suggests a FoF range of about 200-400K.

I did this for a few names (50 Geralds, 18 Charitys (Charities??)) and the overlaps of the ranges seemed to hump at around 250,000, so that’s my vague estimate for the number.

Bu then I remembered that there was actually a paper about this on the arXiv, “The Anatomy of the Facebook Graph,” by Ugander, Karrer, Backstrom, and Marlow, which studies exactly this question.  They found something which is, to me, rather surprising; that the number of FoFs grows approximately linearly in the number of friends.  The appropriate coefficients have surely changed since 2011, but they get a good fit with

#FoF = 355(#friends) – 15057.

For me, with 680 friends, that’s 226,343.  Good fit!

This 2012 study from Pew (on which Marlow is also an author) studies a sample in which the respondents had a mean 245 Facebook friends, and finds that the mean number of FoFs was 156,569.  Interestingly, the linear model from the earlier paper gives only 72,000, though to my eye it looks like 245 is well within the range where the fit to the line is very good.

The math question this suggests:  in the various random-graph models that people like to use to study social networks, what is the mean size of the 2-neighborhood of x (i.e. the number of FoFs) conditional on x having degree k?  Is it ever linear in k, or approximately linear over some large range of k?

Tagged , , ,

Social media benchmark

Sometimes I share my post to various social media sites.  This morning, as it happened, I shared a post almost simultaneously to Facebook, Twitter, and Google+.  On my hit counter, I now see

  • 80 hits via Facebook
  • 30 hits via Google Plus
  • 23 hits via Twitter

Wouldn’t you have expected G+ to come in a lot lower?  Despite the general sense that it’s a total failure, there really are a fair number of people using it. 

Just to reassure you the Earth is still on its axis, Google Search provided 58 hits and Bing 2.

 

Tagged , , ,

Graph search skepticism

Gary Marcus asks the right questions about Facebook’s new Graph Search:

Most search engines, including Google’s, mainly sort pages to see which come closest to some set of keywords (or their synonyms), but they do relatively little to integrate information across pages. If you want a list of all the books written by members of Congress in 2007, you can do a search, but you’ll end up lost. Unless someone has already compiled that information into a single page, you are likely to be directed to a series of individual pages, many with little relevance. It would be left to you to consolidate information across many, many pages; worse, you would have to start from scratch to get the same data for 2008.

But, in theory at least, Facebook Graph Search consolidates information over time and space (albeit in very limited ways). In effect, each user can now use Facebook as if it were a giant, custom-tailored database, not just a librarian that gives a list of documents that are most relevant to your query. Although the ideas behind Facebook Graph Search aren’t entirely original—Google can do similar things in limited domains, such as shopping, and Wolfram Alpha can do math (and draw graphs) based on data in its archives—it really is likely to change the way many people think about search…

Forget search engines. The real revolution will come when we have research engines, intelligent web helpers that can find out new things, not just what’s already been written. Facebook Graph Search isn’t anywhere near that good, but it’s a nice hint at greater things to come.

A nice hint at greater things to come?  Or, like Wolfram Alpha, another case where some of the best programmers in the world, given massive amounts of resources and time, fail to bring us appreciably closer to the dream of the research engine?  In other words, a hint that maybe there are not greater things to come, at least in this direction?

I’m not part of Facebook Graph Search’s “slow rollout,” but from the coverage I’ve read it sounds like it’s good at handling canned combinations of boolean searches.  That’s no joke, but does it really represent progress towards the goal that Marcus has in mind?

Wolfram Alpha, of course, has no idea what books members of Congress wrote in 2007, but that’s not quite fair, because Wolfram Alpha isn’t supposed to know about books.  What does Wolfram Alpha know about?  Well, a query for “Missouri Senate election 2010″ gives you the results of that election, so we know it has state-level results for those elections.  But it can’t put these together to answer “How many Republican senators were elected in 2010?”   “Senators elected in 2010″, which you might think would give you a list, doesn’t – it does, though, tell you that 24 seats went to Republicans and 10 to Democrats, along with the meaningless data of the total votes cast in the US for GOP and Democratic Senate candidates.  “List of senators elected in 2010″ gives the same result.  WA obviously has access, state by state, to the names of the Republican senators who won elections in 2010; but it apparently can’t put that information together into a single list.  Given that, I think gathering their book credits is pretty far off.

Meanwhile, a Google search for “Republican senators elected in 2010″ gives you the relevant Wikipedia page, with the complete list, state-by-state results, and much more.  And searching for “books written by members of Congress” pulls up as the first result a Roll Call article about books written by members of Congress.  Hard to complain about that.

Were any of those books written in 2007?  Who knows?  More to the point — who cares?  That’s the genius of the Google approach.  You know how they tell you, if you’re confused about something in class and you want to know the answer, you should raise your hand and ask, because probably other people have the same question?  That’s the Google principle, except they take it one step further; if you need an answer, not only do other people have the same question, but one such person has already found the answer and put it on the web.  Google can’t tell you which states that entered the Union after 1875 have public universities with animals as their mascots, or which Congressional district ranks 10th by percentage of area covered by water, which is the kind of thing Wolfram Alpha is ace at; but that’s because no one has ever asked those questions, and no one ever will.

 

 

 

Tagged , , ,

Reader survey: have you ever clicked on a Facebook ad?

Apparently Facebook is said by some to be valued at $75 billion.  Facebook’s revenue, just under $2b last year, derives mostly from online ads.

I visit Facebook most days and was only dimly aware that it had ads.  I just looked, and yes, they’re there:  not on the main newsfeed page, but on my profile page.  I’m trying to get my head around the fact that buying Facebook ad space is a good investment.

Question:  have you ever clicked on a Facebook ad?  Have you ever clicked through and then purchased?

Secondary (or maybe really prior) question:  Does it matter to advertisers whether Facebook users click?  Or is the point just to put the name of the product in the corner of people’s eyes, for brand awareness?

Anticipated objections:  1.  Yes, I know the readers of this blog are not a good sample of FB users.  2.  No, I don’t think “businesses wouldn’t buy ads if they weren’t effective” is obviously correct, though it does carry some force.

Update:  Reader Rod Carvalho points out that the same question applies to Google — so feel free to answer that too!

Tagged , ,

How many people will vote on Facebook?

As I write this, about 4.7 million Americans have clicked the “I voted” button on Facebook.  As best I can tell, there are between 125 and 150 million Facebook accounts in the US, and about 90% of these belong to people of voting age.  Of course, lots of people have Facebook accounts but don’t use them.  Facebook says half of registered users log in every day.  So let’s say there’s a pool of about 60 million people who might be expected, if they voted, to click the Facebook button.  Turnout in the election is expected to be about 50% — though of course it will be lower among 18-24s, who are disproportionately represented on Facebook.  How many voters will Facebook tally by the end of polling?

Of my friends, 13% have voted.

 

 

Tagged ,
Follow

Get every new post delivered to your Inbox.

Join 548 other followers

%d bloggers like this: