Tue, 10/25/2016 - 04:22

Posted by Tamara Baluja on March 19, 2013

By Romayne Smith Fullerton

The future of journalism lies in “robot reporting” and “cyber stringers.” 

That was the cheery message Ben Welsh, a database producer with the Los Angeles Times, and Steven Levy, senior writer for Wired, delivered on CBC’s The Current on Tuesday.

The discussion began with a clip from President Barack Obama, and then Welsh asked, “Was the President telling the truth?” Using a computer program called “Truth Teller” that converts audio to text and then uses an algorithm to compare claims in the text to a database of known facts, it’s easy and fast to answer that question (in this instance, yes) and many others.

Welsh also noted that there are myriad ways computers can outperform their human counterparts, and he offered the example of when an earthquake struck California in early February, the Los Angeles Times covered the story. In fact, a post appeared on the Web only eight minutes after the quake occurred. And it appeared under reporter Ken Schwenke’s byline (who was sleeping when the quake hit) because, after all, Schwenke had written the computer code that ‘wrote’ the story.

Amazing. Yes. Audible sigh.

In the last year, the journalism industry in Canada has lost more than 1,000 jobs – at least 650 at CBC and 500 at Sunmedia -- because resources for newsrooms are shrinking.

Surely now is not the time for a naive love-in about how technology can do it better, or to laugh off the very serious ramifications for those reporters facing replacement by computers that can certainly do it cheaper.

To be clear, I absolutely agree that there are many ways that computers can assist reporters to better cover a variety of stories (see some examples here, but I want to emphasize the word ‘assist’ because it seems to me that behind all this sexy talk about technology is the assumption that questioning its place or its proper role makes me backward.

The much quoted phrase that ‘the only constant is change’ often leads to the charge of ‘luddite’ if one seems to ask questions that challenge change. I like to think that there’s a difference between mindless acceptance of change and a thoughtful assessment of it.

But when I read Steven Levy’s piece on robotic reporting about how he hoped he didn’t get scooped by a Macbook Air, or how wonderful it is that parents can use an iPhone app to ‘write’ the story of their children’s little league games to the web, it sounded like we’re using humour to rationalize something we don’t even know we might miss.

When my son played minor league hockey in our small town, we parents loved taking turns writing the sports stories for the local paper. We’d all chip in bits and pieces and observations. It was a great opportunity to exchange a laugh or an idea, help each other out with players’ names, and so on. It was a kind of intimacy that leads to building friendships and communities, and highlighted the inherent value of communication that’s not mediated by, or through, a device.

Last year, the Canadian Association of Journalists considered the question, What is Journalism? And they came up with three requisites, all of which had to be present, to fit the definition: (1) a disinterested purpose; (2) an act of creation; (3) a particular set of methods. While computers could excel at 1 and 3, surely they can’t even approach the second?

Story-telling is a creative act and at its heart is an artist. The best journalists use traditional forms and story shapes to help see the world in a new and different way. It’s similar to the way in which a poet might choose to express his or her view using a sonnet or a ballad. But it’s still a creative act that requires more than the sum of its constituent parts: intellect, research, emotion, or even skill.

In Das Kapital, Karl Marx writes that bees construct better cells than architects. But, he argues, “what distinguishes the worst architect from the best of bees is this, that the architect raises his structure in imagination before he erects it in reality.”

It’s imagination that will be sacrificed if we let technology push humans away from the centre of journalism and story-telling. We need people to assign values to what’s reported, and how it’s reported, and why we need to feel something about the stories we see and hear and read. And if I can’t convince you with the positives, I’ll settle for sounding the alarm.

Humans invariably become victims of their own creations; think Mary Shelley’s Frankenstein. In part, the lesson is we do things because we can; but we don’t always ask why we do them or ought we to do them in the first place.

Canadian communications scholar Harold Innis suggests that every communication technology —the phonetic alphabet, the printing press, a magazine, a computer, and so forth —creates, shapes and (re)produces the attitudes, values and character of its culture(s) or public(s).

If robotic reporters become not only the communication tools with which we use to think, but they also become the technology that does our thinking and knowing for us, what then?

Today we are all under increasing pressure to digitize, mediate, streamline and up- or download. People know they’re not really in meaningful relationships unless they’re ‘friends’ on Facebook. It’s as if we doubt our own existence if we can’t find an external reflection, some sort of mediated representation to confirm us to ourselves.

How pathetic.

The pressure to conform to or be seduced by technology is everywhere; but I don’t think resistance is futile. In the end, I would still prefer to live in a world led by ideas than one driven by technology.





















I'm glad you're covering this issue, but I disagree strongly with your take. My hunch is that you've never written any computer code. Underneath every algorithm is a human mind or multiple human minds. Those that write code know that their interests and ideas are indelibly embedded. Technology shapes those ideas, no question. And technology shapes the consumption of those ideas as well. But I think you set up a false opposition between human values and technologies.

It's human values and human ideas that drive algorithms. Same in statistics. Same in science. Same in Python, Ruby, PHP or Javascript. 

Put another way, it's not the technology that worries me. The automated data gathering and analysis, and publication, is awesome. What worries me is the values that are making these algorithms. And it also worries me that these processes will probably be proprietary and closed. News outlets are too often commercial enterprises first, and "journalistic" enterprises second. In scientific endevours the methods and the statistical analysis have to be published and peer reviewed. I doubt news outlets will open up their data sets, or their methods. 

In my view, this technological development is only a new face to a long time challenge for journalism: what passes as journalism, often isn't.

J-Source and ProjetJ are publications of the Canadian Journalism Project, a venture among post-secondary journalism schools and programs across Canada, led by Ryerson University, Université Laval and Carleton University and supported by a group of donors.