At some point or another, almost all of us have experienced breaking news which reaches us via Twitter. Tweets spread like wildfire, and give a 140 character word of mouth synopsis that often makes news outlets jump to attention, and start the fact finding process to bring us more on a story. If your job role revolves around the delivery of news content, it couldn’t be a better social network to be a part of.
There are however a number of tricks and tips that news hounds can use to provide them with the extra information and potential detail they need to run an article.
Obviously good journalism relies on gathering the facts as quickly as possible, and the web’s inherent urgency magnifies that. From a web marketing perspective, the outlet who breaks the news is the one which gets the traffic.
So, here’s a couple of tips to find the information you need before your competition does, which should apply to all publishers but particular those that rely on breaking news.
Finding Sources – How to find the first tweet
We’ve all been there. Some emergency hits Twitter, and everyone scrambles to find sources for information. The first person to naturally ask details from would be the person who posted the first tweet. But what if its a retweet though which has proliferated miles away from the original source of the news? The person who shared the information may have no idea where it started.
Enter the magical world of Twitter search. Right as I begin to write this post, Tweetdeck comes alive with reports that an Earthquake has hit Chile, a perfect example to illustrate the information within this post.
First thing I did? Change the date filters within Twitter’s Advanced search.
This is going to get rid of any information which isn’t relevant to todays news. To find the first Tweet, I simply typed earthquake into the search box, and paged through the results. With the massive amount of pages to get through, I simply did what I call a manual binary search algorithm.
Non programming geeks. Stay with me. It’s simple.
Due to the fact that Twitter page their results via a querystring in the URL, i.e. page=30 denotes go to page 30 of the results, we can exploit this to our advantage.
To find the last page, I simply change ’30’ to ‘100’ and see if any results are returned. If there is results, I try ‘page=200’ if there are no results, then I get the half way point and try ‘150’, if there are still no results, I try ’75’, if there are results, I try the half way point again, and try ”113′. Eventually you will find the point where results are returned, and where the paging finishes. You could of course start at page one and work your way up until you find no more pages, but hey where’s the geekiness in that?
This places you pretty much at the start of when Tweets occurred.
For the live example which is happening RIGHT NOW. Here’s what I find.
The NewEarthquake account appears to be where things kicked off via an RSS feed at, 8.16pm, 11th of February. It links through to the official Earthquakes Hazard program where the news hit the web at 8.05pm. Not bad. A delay of 11 minutes before the news hits Twitter.
How to find the exact location of breaking news
So lets say that this still hasn’t provided you with any decent information on the event. You want to find people who are near the event and potentially affected by it. With an earthquake, finding this information is a piece of cake as the epicentre geo-coordinates are provided on government websites as can be seen on the above screenshots. Other news it is often harder to get an exact pinpoint for this data.
Twitter to the rescue again.
So you’ve found your first tweet. If it isn’t immediately obvious where the news has happened, how do you go about extracting co-ordinates from it? The API of course provides full meta data on Tweets, you just need to know how to go about querying it.
Querying the Twitter API for Geo information
Twitters public API contains many useful snippets, however to find the geo information for a Tweet, follow these simple steps.
1) Find the Tweet ID
You can find the ID for a tweet very simply in the web interface. Hovering over the time a tweet was posted at will provide the URL (and Tweet ID) in the status bar of most decent browsers. Clicking on it will also reveal the ID, but its basically the big long number which represents each status update.
2) Use the API to work out where that Tweet was made.
http://api.twitter.com/1/statuses/show/[INSERT-TWEETID].xml so for this particular example, this would be 36156575169118208 and the web page you need to visit would be: http://api.twitter.com/1/statuses/show/36156575169118208.xml. You can see from the source code of that page that geoinformation is contained within, giving you the co-ordinates you need to do the detective work.
Querying Google Maps for Geo information
Unless you are rain man,working out the latitude and longitude for a particular location is a bit of a pain. Thankfully, Google maps makes this information freely available for ANY location in the world. For the recent tragic events which unfolded in Cork this week, I knew that Cork airport was going to provide the most interesting and relevant tweets for that news story, and thus needed to find its geo-coordinates.
Armed with Google Maps, it became a simple process to work out what the latitude and longitude co-ordinates of the airport, and the same was true tonight with the Chilean Earthquake. Here’s how to extract latitude and longitudes with a Google labs plugin that makes things a breeze.
5. Hold down shift over the place you want to see co-ords for.
Using Advanced Twitter Search to find tweets in a location.
No we can return to twitter search to find people in the vicinity. Click on Advanced Search. Where it asks ‘Near this place’ you can use co-ordinates. The very same co-ordinates that we managed to data mine earlier. For tonight’s Chilean Earthquake – I used the following query.
This was amazing to watch. Some of the tweets coming were right as people felt tremors and shocks. Throwing these directly into Google translate helps if your Spanish isn’t great. Google Chrome automatically translated many of these for me on the fly as I browsed the Tweets.
Photographs were also easy to find by simply search for ‘links’ and retweets. For example, here’s a recent photograph uploaded onto Flickr showing some of the damage caused by the earthquake. This YFrog image expressed concern being placed on infrastructure at the petrol pumps, as people left the area. (Notice the RT and share counts for this mundane image).
Obviously this could be used in any situation to provide realtime information on what is happening, but when news breaks it is particularly effective at gathering both media resources and contacts on the ground at the time and place of events. For the rest of us, it makes for fascinating real time insight into how social media is currently being used in disaster scenarios to share and spread useful information.