“Search requires that you know what you’re look for, but maybe you just want to see what the world is paying attention to.”

bitly:

Dear developers and data lovers,

We’re incredibly excited to announce the public release of a new social data API! Every day millions of people shorten, share, and click on links via bitly’s services. This API gives you direct access to the best available content shared by people across all social networks.

There are three kinds of functionality you can now access:

True Realtime Search

Run a query and get back the top URLs and stories for that query right now. Queries can be specific phrases, like “obama”, or filtering criteria, like “stories about food being read by people in Brooklyn”.

Attention Spikes

Search requires that you know what you’re look for, but maybe you just want to see what the world is paying attention to. Our bursts API returns the current phrases that are receiving a burst in attention beyond what we would expect. For example, “giant squid” is bursting today because of this story: Giant Squid Captured on Film. Bursts automatically aggregates multiple articles about the same thing together, which you can see on the realtime story pages.

Metadata about URLs

Finally, we do quite a bit of analysis on the content of each URL. You can now query on a URL basis for keywords, topics, content, language, and location relevance. Generating this kind of metadata is difficult problem faced by anyone who wants to build an application on links. We’ve solved it, now there’s no need for you to!

Full documentation is available on our dev site and in our Python library, and if you’re in New York stay tuned for our API LAUNCH HACKATHON next week in our office. Send your questions, comments and ideas to api@bitly.com!

A$AP Rocky hit a Twitter Trends special character bug.

pnpulse:

As of 5:20 pm EST Halo 4, which was released in the U.S. today, was mentioned more times on Twitter than #ivoted.  

• Halo 4: 204,999 Mentions
• #ivoted: 57,149 Mentions

*Both topics were listed as actively trending on Twitter when data was pulled.

Hm.

1) I’m gettin’ something different.

2) When you query for #election2012 instead of #ivoted, it’s not even close.

What tool are you using? I’m using Topsy.

You’ve seen it in action, but maybe haven’t quite understood what exactly was going on.

Trending Topics on Twitter like:

Thirst Re-imagines Twitter

Yahoo CEO

Music on iPad again

They’re Mashable headlines! Or at least pieces of them. And due to their frequency/velocity/quantity, they trend on Twitter for many folks, pretty much every day.

Here’s why it happens: Spammers. More specifically: spray + pray bot-marketers. We’ve all seen these bots on Twitter telling us about free iPads and weight loss, etc. But they need filler content, too. They just can’t blast 24-7 about their CPA offer.  And the most popular go-to filler content is Mashable.

These folks plug Mashable’s RSS feed into tools like Twitterfeed, Ifttt, Buffer, Hootsuite, Dlvr.it, etc. Then whenever Mashable posts a new story, so many of these accounts push out the story around the same time, that it often becomes a Trending Topic. 

This isn’t Mashable’s fault, but they definitely benefit from it. Also, Mashable isn’t the only site that benefits from spammers’ hunger for “filler” feeds. But they are the most popular one I’ve seen. 

Anyway. Case closed. That’s why Mashable trends every dang day. (Yes, I realize that Twitter’s recent tweak to personalized Trending Topics alters this a little.)

If you dig solving social media X-Files like this, you should probably sign up for my mailing list. I’m such a Scully. (In a universe where Scully’s deductive skills matter more than the supernatural.)

(This blog post was based off of some Tweets I made earlier today, Storified below.)

Let’s talking about finding influencers.

Just. Kidding. 

While I’m not an expert in the science of finding influencers, I hear the word often enough that I can’t totally avoid it.

I try to never say “influencer.” Not just coz it has icky connotations, but coz it’s part of the larger science of network analysys/social network analysis, and I prefer to borrow more specific terms from that body of work.

But I’m sympathetic to why people are drawn to the word and the idea, influencer.

It’s an easy concept for our brains to grab. Lotsa animals, humans included, seem wired to be aware of status. Who’s important? From there it’s not a big leap to “who’s influential?”

It’s natural to want to keep track of a small number of important folks.
Who’s in the VIP? Who’s A-List?
(Or in social-network-analysis-speak: “Who is the important node in the network?” If you want to sound really smart, definitely talk about “nodes” whenever people bring up Klout scores :D)

But in flailing for influencers, it’s easy to forget why it even matters- to get people to consider a new idea/song/recycling habit/brand of laundry detergent. It’d be pretty cool if Kim Kardashian tweeted about your lipstick, but you can waste a bit of energy going after that scenario. Plus… it can be demeaning for all involved.

The good news

But what if there’s a workaround? And in the spirit of the egalitarian bias of the social web, what if it’s waaaaay less icky than courting influencers? What if you could interact with more normal folks, get the same results, and have way more dignity?

Both my experiences and the math say it works.

What’s the catch?

You need to interact with more “nodes.” But unlike the process of courting influencers, encountering a disinterested node is not an issue. Just move on. Just work the process.

The amount of people you need to engage is dependent on the size of the network. Is it “everybody who follows $competitor?” Is it “everybody who Likes $MusicArtistAlbum?”

If you’re interested in this stuff, I just read a great paper [pdf] from West Point (!) about how to calculate your “seed set.” (The amount/body of people selected from larger network to engage with.)

But here’s a summary from their blog.

The spreading of a trend or behavior in a social network is a very active area of research. One very important model of trend spreading is the “tipping” model. With tipping, an individual in a network adopts a trend if at least half (or some other proportion) of his or her friends have previously done so. An important problem in viral marketing is to find a “seed set” of individuals in the social network. If all members of a “seed set” in a social network initially adopt a certain trend, then a cascade initiates through the tipping model which results in the entire population adopting that trend.

…seed sets in real-world online social networks can be very small. For instance…we found a seed set that consisted of only 0.8% of the population.