[Air-L] Archiving Past Tweets

Pablo Garaizar Sagarminaga garaizar at deusto.es
Tue Feb 28 00:47:46 PST 2012

Hi again,

  BTW, "DataSift Unlocks Access To Historical Twitter Data Dating Back
To January 2010":


DataSift, one of Twitter’s data partners which currently provides
developers and third parties with access to the full Twitter firehose
in realtime, is about to unlock a whole new set of Twitter data to the
ecosystem. The social data platform has launched Historics, a
cloud-computing platform that enables entrepreneurs and enterprises to
extract business insights from Twitter’s public Tweets dating back to
January 2010 (we originally reported on the pending launch here).

Developers, businesses and organizations can essentially use DataSift
to mine the Twitter firehose of social data. But what makes DataSift
special (besides the premier access to Twitter data) is that it can
then filter this social media data for demographic information, online
influence and sentiment, either positive or negative. As we’ve reported
in the past, DataSift does not limit searches based on keywords and
allows companies of any size to define extremely complex filters,
including location, gender, sentiment, language, and even influence
based on Klout score, to provide quick and very specific insight and

Currently, DataSift is the only company to offer historical Tweets from
as far back as two years ago. As AllThingsD’s Liz Gannes reported
recently, Gnip was the first authorized reseller of historical data
from Twitter, Tweets only went past 30 days (previously developers
generally had access to 7 days worth of data).

With DataSift, companies can now tap into both real-time and over two
years of historical Tweets to filter and extract insights and trends
that relate to brands, businesses, financial markets, news and public

Basically, DataSift aggregates this public social data, enriches it,
filters and delivers this data social media monitoring, business
intelligence and CRM applications. DataSift enriches every Tweet with
details including sentiment, topics, web-links, location and social
media influence so that companies can extract meaning and create

As CEO Rob Bailey explains to us, one example of a use case of the
historical data could be a mobile handset company that may want to see
how a previous product launch unfolded, who Tweeted about it, from what
geographic regions, and which customers were positive vs. negative.
Financial organizations can analyze popular sentiment, trends and
indicators relating to businesses and economic events and news and
research organizations can surface new trends around historic and
popular culture events. According to Bailey, demand for Historics is
high. Almost a thousand companies – including 100 of the Fortune 500
companies – have already joined the waitlist for Historics.

Founder & CTO Nick Halstead tells us that DataSift’s platform is so
powerful that it can process a whole month of Twitter data in an hour.
He says that currently DataSift us running over half a petabyte of

In terms of availability, Historics is available today as a limited
release to existing customers and is currently scheduled to be
generally available in April 2012. The service itself is available as
either as a corporate subscription or pay-as-you-go service.

Bailey tells us that 2012 is going to be a huge year for the recently
launched company, and will be increasing the number of social data
sources by a factor of ten. We’ll also see a number of big customer
announcements, as well as making the UI of parsing through this social
data as easy to use as possible with features such as drag and drop

  Pablo Garaizar Sagarminaga
  Universidad de Deusto
  Avda. de las Universidades 24
  48007 Bilbao - Spain

  Phone:       +34-94-4139000 Ext 2512
  Fax:                  +34-94-4139101

More information about the Air-L mailing list