Google’s Dylan Casey introduces real time search and timeline discovery via the Google Blog.  This is great for folks looking to harvest what Twitter was failing to do which is go back in time to see what people have been saying.

In other words Google is allowing users to search and explore the public archive of tweets.

My Angle

What Google has done is created a “corpus” or  index of all tweets going back to March 2006.  After some data cleaning they are applying the “Google secret search sauce” to give users a search and discovery experience from the “old real time web” or as I call it “yesterday’s web”.

This is where Google shines applying classic search in a new way.  One thing about Twitter is the speed of user updates.  It’s fast and the notion of relevance changes.  Another issue is content definition type and how to treat certain tweets.  Google is fixing this with the archive version of real time –  Yesterdays Real Time Web.

Google (and Microsoft) have a great opportunity  to use the real time data to “link to” other relevent deeper data sources to generate a better user experience.  In other words Google has more diverse content sources and can “front end” them with real time data and this new old realtime data as a filtering mechanism to other content types like blog posts, adwords, media, etc.

This is very powerful and something Twitter can’t do and makes the case that Twitter will be bought by Google after Twitter fumbles the ball with Paid Tweets.

Here is what Google is introducing.

Google introduced real-time search last December, they’ve added content from MySpace, Facebook and Buzz, expanded to 40 languages and added a top links feature to help you find the most relevant content shared on updates services like Twitter.

With the advent of blogs and micro-blogs, there’s a constant online conversation about breaking news, people and places — some famous and some local. Tweets and other short-form updates create a history of commentary that can provide valuable insights into what’s happened and how people have reacted.

Google wants to give users a way to search across this information and make it useful.

With Google timeline search and discovery a user  can zoom to any point in time and “replay” what people were saying publicly about a topic on Twitter.

Here’s How To Get Started

To try it out, click “Show options” on the search results page, then select “Updates.”

The first page will show you the familiar latest and greatest short-form updates from a comprehensive set of sources, but now there’s a new chart at the top.

In that chart, you can select the year, month or day, or click any point to view the tweets from that specific time period.  In this example the user searched for [golden gate park] and browsed to see March, 2010:

The chart shows the relative volume of activity on Twitter about the topic. As you can see, there are daily spikes in the afternoon (when parks are the most fun) and an unusually high spike on March 27. Clicking on the 27th, you’ll discover it was a sunny Saturday, which may explain the increased traffic on Twitter.

People were tweeting about disc golf and tennis, biking, riding a party bus, craving chips and salsa…the kind of local, time-specific information that up until now would be almost impossible to find online.

By replaying tweets, you can explore any topic that people have discussed on Twitter. Want to know how the news broke about health care legislation in Congress, what people were saying about Justice Paul Stevens’ retirement or what people were tweeting during your own marathon run? These are the kinds of things you can explore with the new updates mode.

Availability

The replay feature is rolling out now and will be available globally in English within the next couple days (if you want to try it now, try out this special link).

For Google’s initial release, you can explore tweets going back to February 11, 2010, and soon you’ll be able to go back as far as the very first tweet on March 21, 2006.

All of us are just beginning to understand the many ways real-time information and short-form web content will be useful in the future, and we think being able to make use of historical information is an important part of that.