The ethics of geotagging
Posted by
Martin Stabe
on 6 December 2007 at 12:50
Tags: Archant, Ethics, Press Complaints Commission, geotagging, thelondonpaper
In a letter published in this week’s
You report Archant bosses as saying that their ground-breaking plan to introduce postcode-targeted “personalised” websites will “require a change of thinking from our reporting staff”. This will involve them procuring the postcodes (or GPS co-ordinates) of parties who feature in their stories (how many people know their GPS co-ordinates?) and then meta-tagging them into the texts uploaded to the sites.
For Archant reporters - and no doubt others if other groups follow the lead - there will be an addition to the familiar “who, what, where …?” components of their stories. To the five Ws and the H there must be added a P.
What happens if a citizen firmly declines to give it, apprehensive about junk mail or even frauds that may be perpetrated using the information? Will it be a matter for discipline for the reporter who fails to procure it? It would not just be a missing element from the story, of course, but a blow to the company’s commercial strategy.
And how long before Archant reporters will be heard calling out: “Anyone here been raped and got an NR postcode?”
Some of these fears are a bit far-fetched. In fact, cooperation from sources will rarely be needed to obtain the necessary data. Even before it is widely implemented by news organisations, new gadgets and software tools are increasingly automating the process of gathering geographical data.
But that merely raises new issues, which are worth thinking about now as more news organisations have started talking about geocoding their stories.
Some photojournalists already collect coordinates every time they release the shutter, because modern GPS-enabled cameras embed geographic data in each image file. Some mobile phones, like the Nokia N95 used in Reuters’ new Mobile Journalism Toolkit, has GPS capabilities. The same phone was used earlier this year in the Geo-Stories experiment by BBC Innovation and the University of Brighton.
For desk-bound reporters reporting from the newsroom, online mapping tools already make it simple to cross-reference known locations with their approximate coordinates.
Both approaches are already widely used by consumers. The photo-sharing site Flickr lists more than 1,000 photographs taken within a stone’s throw of my current location in Underwood Street, London (which my computer knows as 51.529910, -0.091104). Globally, the site has millions of geotagged images.
Where all of this fails, reader input or text-mining software can compensate. The German regional news portal Der Westen relies on its readers to help add geographical metadata to stories (like this one) that reporters have not geotagged. A simple online mapping tool lets readers identify the location of untagged stories. Other users can then vote on the accuracy of the user-submitted location.
Sites like YourStreet, meanwhile, use software that can extract and identify places named in copy.
As Steve Yelvington argued recently, such automated methods will be crucial to widespread adoption of geotagging:
The real barriers to geotagging news are … the practical problems associated with workflow and manpower implications.
So long as content creators (professional or amateur) have to think and act in order to geocode information, it’s not going to happen consistently.
So I think the most valuable contributions are going to be in tools that are transparent to the user.
But this increasingly automated, transparent, and potentially non-consensual, geotagging raises the ethical issues Gopsill alludes to.
As geotagging becomes incorporated into newsgathering, journalists will have a duty to consider its privacy implications and whether it is appropriate to publish the geographical data they have automatically generated.
In effect, geotagging is much like naming the street address of people and places mentioned in a story. If you would be comfortable providing a street address in a story, it is difficult to see how embedding machine-readable geographical coordinates is significantly different.
The Press Complaints Commission has never ruled that addresses are intrinsically private under clause 3 of the Code of Practice, and has taken complaints about newspapers publishing addresses on their merits. The PCC generally considers whether people living at the address are vulnerable to stalkers, or if publication is likely affect their private home life. In a 2003 case involving singer Ms Dynamite, for example, the PCC ruled that the Islington Gazette had acted improperly by providing her full address. Similarly, the PCC decided that the Mirror should not have published details about JK Rowling’s home, even though they were already in the public domain. Similar considerations will probably be used to determine when journalists decide that content should not be geotagged on privacy grounds.
War reporters will obviously want to avoid publishing the precise location of the units they are embedded with by not filing geographical coordinates with photographs.
And journalists will also need to be vigilant about not accidentally revealing geographical information that could compromise the identity of a confidential source.
In February 2006, a Washington Post reporter and photographer interviewed a young hacker about his computer crimes on the understanding that his identity would be protected. But Slashdot users claimed to have identified where the interview had taken place after discovering that photographs published with the story on WashingtonPost.com included automatically-embedded GPS coordinates.
Their sleuthing revealed that the location was a small rural town, where the identity of a young computer expert would probably not have been difficult to guess. The newspaper would not comment, but, even the Slashdot account is untrue, the incident highlights the new considerations that geotagging will impose on journalists.
