Apigee’s Sam Ramji pointed out to me today that Twitter did a first when communicating with developers yesterday about its new platform policies. For the first time in recent memory Twitter has used a graphic to actually illustrate what it sees as the acceptable and not so acceptable ways to use its API. The graphic is clear and to the point.
When you look at the graphic, one thing is for sure, Twitter wants nothing to do with the enterprise.
Here it is:
Here’s Twitter’s spin. Do you want to develop a consumer facing Twitter client? Well, you will be under some strict guidelines if so. But if you want to use Twitter for business, you are pretty much free and clear. So that means if you are analyzing tweets for sales people to get new leads then I’d say you are okay. If you want to develop a service that helps marketers target influencers then yes, go for it.
Twitter gives a few examples of what it sees as fine uses of its API:
In the upper-left quadrant are providers of tools that help businesses engage with Twitter including social CRM providers like Sprinklr, HootSuite and Radian6 (acquired by salesforce.com), and integration companies like Mass Relevance, which aggregates and filters Tweets for display on TV.
I’ll add a few others that I think are just fine:
SocialPandas launched this week. It will use data from Twitter and other sources to give sales people better tools for building relationships.
SugarCRM integrated Twitter last year.
Datasift is a powerful Twitter data analysis and business intelligence platform.
Twitter has no interest in the enterprise. For that matter, neither does Facebook.
But as we well know, enterprise startups do very well when using Twitter and Facebooks as models. Case in point? Yammer sold to Microsoft for $1.2 billion. Considering that success, I’d say it’s a good thing for the enterprise entrepreneur that Twitter has no interest in the enterprise side of APIs.
DataSift, the UK startup that allows developers and businesses to do incredibly smart things with Twitter’s firehose and other social data, is today rolling out its Historics service to every company with a DataSift subscription.
Historics is a product that allows DataSift customers to “aggregate, filter and unlock insights from public Tweets, both in real-time or going back to January 2010, to get a long view into their brand, business and markets.”
While most Twitter users can only search back through a few days’ worth of tweets, DataSift’s firehose deal with Twitter means that it has a much deeper well of data to draw from, and the company’s platform is designed to allow users to run complex, specialized searches, filters and analysis that can be of real use to businesses.
In an email sent to users today, the company says that it has been “overwhelmed by the response and demand” since it announced Historics earlier in the year. Now the news organizations, brands, and researchers that subscribe to DataSift will be able to get some real historical context to their social analysis.
While some Twitter developers have reason to be worried today, DataSift is safely in the bottom-left quadrant. Earlier this month, it announced the opening of a New York office as it aims to make a serious play in the financial sector.
Image credit: Dave Stokes
Without question, there are quite a few spammers on Twitter. These fake accounts make you wonder how many active people truly do follow you on the service. Ever since people started talking about Facebook’s fake account issue, the spotlight has now been turned to Twitter. Is it full of “fakers?”
An app called “Fakers” from StatusPeople will scan your followers and let you know for sure:
Here’s how StatusPeople explains the methodology behind the service:
We take a sample of your follower data. Up to 500 records depending on how ‘popular’ you are and assess them against a number of simple spam criteria.
On a very basic level spam accounts tend to have few or no followers and few or no tweets. But in contrast they tend to follow a lot of other accounts.
Those numbers seemed kind of low to me, and I’m still not entirely sure what StatusPeople considers to be a “faker”, but I will say that if this is close to being true, I’m impressed. Good job, good effort Twitter.
UPDATE: I fully expect the “phony” functionality any day now:
2% of my followers are fake, and the other 98% of you are just phony. How many fake followers do you have? thenextweb.com/socialmedia/20…
— Gabe Rivera (@gaberivera) August 14, 2012
Twitter just announced the launch of its Twitter Political Index. This index, says Twitter, is “a daily measurement of Twitter users’ feelings towards the candidates as expressed in nearly two million Tweets each week.” Every day, twitter will evaluate and weigh the sentiment of tweets mentioning both Barack Obama and Mitt Romney relative to every other message that passes through Twitter’s systems. The system then calculates a score for both candidates. Currently Obama‘s score is 34 (and trending down) and Romney, whose follower count on Twitter may be somewhat inflated, is at 25 (and trending up a bit). The Twitter Political Index will be updated every day at 8pm ET.
Twitter, of course, has been used to study political sentiment for quite a while not and the company even has its own Government & Politics team to analyze this data and assist third parties with their own efforts. For this project, however, Twitter also partnered with the data analysts at Topsy, as well as polling firms The Mellman Group and North Star Opinion Research. USA Today will also use Twitter’s election meter in its coverage of the ongoing presidential campaign.
According to Twitter, its own data generally shows the same trends as Gallup‘s approval ratings surveys and frequently “hints” at “where the poll numbers are headed.” There is some evidence, however, that “the predictive power of Twitter regarding elections has been greatly exaggerated.” That study, however, mostly looked as message volume while Twitter’s barometer focuses on sentiment analysis.
No doubt you’re going to hear the phrase ‘Twitter election’ about a million times between now and the US election date. Not least on the Twitter blog itself which has announced the launch of its Political Index for tracking sentiment around the forthcoming vote.
Use the index to find out what people on Twitter are feeling about the various candidates and see if you think you can predict the outcome from the changes in mood as the days pass.
According to the blog post, “Each day, the Index evaluates and weighs the sentiment of Tweets mentioning Obama or Romney relative to the more than 400 million Tweets sent on all other topics. For example, a score of 73 for a candidate indicates that Tweets containing their name or account name are on average more positive than 73 percent of all Tweets.”
There are of course alternatives for tracking elections if you need your political fix. Electionista from Tweetminster tracks elections around the world and should provide some insight for those who like to follow the crowd opinion.
It’s still uncertain if elections can be predicted using social media sentiment or tracking of any sort, but you can be sure that with so many people using the short form messaging platform that it will be a place to find out the gossip, news and rumours as the political machine switches up gears and Americans get their vote on.
Image Credit: Cristian_RH7