News media APIs: more on mashups
Posted by
Martin Stabe
on 28 August 2008 at 09:40
Tags: Journalism
The idea of a website that is not merely a destination, but a “platform” or a set of tools on which users can build additional services, is nothing new — it’s a key feature of many of the thinking underlying many of the current “2.0″ startups, and has been a fundamental part of the success of Internet giants Google and Yahoo.
What is new is that news organsiations are beginning to adopt this approach and are beginning to release application programming interfaces - more commonly abbreviated as APIs - to allow outside web developers easy access to their content for use in non-commercial online projects.
Publishers see it as a way of expanding their reach online by gaining access to niche platforms such as new mobile phone handsets and games consoles, new niche reader communities they do not have the resources to serve directly, or simply discovering new ways of using their content by tapping into the creativity of their online users.
Some also hope it will be a pipeline toward more traditional commercial syndication deals. It’s a trend I looked at in the first issue of the monthly Press Gazette, which is out today.
The BBC has been a leader in the field, with its BBC Backstage programme, which has been making available APIs and fostering a community of online developers since 2004.
Releasing data to the public is the easy bit; building a community of developers to actually use it as part of a platform strategy can be more difficult for some publishers. To get around this, some are co-operating with established networks. An API of listings data from the Press Association is expected to be released this week through BBC Backstage.
Ian Forrester who heads Backstage, says fostering the online development community in the UK is part of the public service remit of the BBC project.
Projects built by the BBC Backstage have been spun off into commercial projects. The website Trafficeye, for example, started out as a Backstage prototype using traffic information supplied by Trafficlink. Exceptional Backstage projects may be integrated into the main BBC website. The BBC home page archive is an example.
Thomson Reuters has several projects accessible to outside web developers. It has opened an advanced semantic tagging tool used to organise its internal archives to the public in a limited form called Open Calais. It has quickly build a community of 2,500 developers actively experimenting with it, and a Reuters spokeswoman says it has already generated requests for commercial licensing of the more powerful underlying software from Reuters-owned Clear Forest.
In May the Reuters Labs research and development unit released Spotlight, and API providing access to the content of Reuters.com in various languages. One impressive prototype combines the two Reuters services to create Gist, a self-organising news website.
In July, the US National Public Radio network unveiled an API providing access to 250,000 articles from its website NPR.org, dating back to 1995. It has already been used to create a widget that (somewhat inevitably) places NPR news stories on a world map and to make NPR material available on the Apple iPhone.
Some projects have caused a stir online before even being released. The New York Times revealed in May that it was planning to be release a number of APIs.
Oren Michaels, chief executive of Mashery, a company that is working with both Reuters and the New York Times on their API projects, says the announcement has spurred interest from other news organisations.
“Everyone saw the New York Times announcement and it’s all starting to buzz around,” Michaels said.
“We’re getting a lot of preliminary feelers from people. That often happens in a lot of the areas we work in: As soon as a market leader makes an announcement like this all he others think ‘we need to do something like this too’.”
Marc Frons, the chief technology officer for digital at the New York Times said the first beta tests of a New York Times API should be expected in early September. They have come about as part of a wider newsroom IT project called the “data universe”, which is setting up new ways for journalists to store various types of information, such as events listings, restaurant reviews, BMDs and political party funding information, in a more computer-friendly structured format — an approach that some tech-savvy journalists have been advocating for some time. Frons explained the data universe project in a recent web chat on NYTimes.com.
Closer to home, both the Guardian and Telegraph have been hinting at developing a similar approach to their websites. The Telegraph has made outreach to the web developer community one of the tasks of its Telegraph Labs setup and has already hosted one weekend open house to encourage programmers to come up with new ways of using Telegraph.co.uk content. The Guardian, meanwhile, hired Matt McAlister, who previously headed Yahoo’s developer network, to launch a Guardian Developer Network.
More links about this topic can be found on my delicious.com page.
