Matt Andrews

Imagine journalism in ten years' time: notes from my talk

08 Mar 2013

A few months ago I was approached by John Mair of the University of Northampton who'd attended my class at the Guardian's Digital Journalism masterclass. He asked me if I'd be up for speaking at a mini-conference for journalism students titled "Imagine journalism in ten years' time". I went along and presented my thoughts, along with a brilliant set of speakers: my former colleague Judith Townend, now finishing a PhD at City; Andy Dickinson, lecturer in journalism at UCLAN; Teodora Beleaga, Insights Analyst at KBM Group EU, and Jay Rosen, professor of journalism at NYU, who joined us via Skype. The event was chaired by former BBC editor Kevin Marsh.

This blogpost is a writeup of the ideas I rattled off in my fifteen minute talk. Judith and Andy have also shared their presentations, both of which were extremely interesting and well-presented -- go and check theirs out, too!

---

A caveat to begin with: this kind of thing is difficult. Trying to anticipate the future is almost always an exercise in futility. Let's start by looking backwards: what world did we live in ten years ago?

It's 2003: Facebook and Twitter are just twinkles in the eyes of Silicon Valley. Google's dominance of the entire information sector is mostly limited to search. No iPhone, Android or any other true smartphone. News websites are growing but not eclipsing print yet. Burgers were still made out of beef.

Today we live in a world where news stories break in seconds and the first people to report on them often aren't journalists. Young people no longer buy newspapers but get their information from their social networks. The most profitable technology companies are the ones with the most information stored and capacity to use it. A huge proportion of society uses mobile computing as their primary or secondary device.

Trying to anticipate this all over again is challenging, but not impossible. It's tempting to imagine that a decade from now we'll all be using some revolutionary new device or app, living in virtual reality, that kind of thing. The reality is perhaps a bit less "futuristic": the things that will dominate the year 2023 are probably already with us now, or at least, the seeds of their conception.

The journalism industry is in a strange place at the moment. An optimist might describe it like a phoenix about to burn to ashes, shortly to be reborn in a blaze of glory. A pessimist might describe it as just being burnt to ashes.

What's certain is that change -- dramatic change -- is a constant. The brilliant thing about the web is that it's levelled communication and collaboration -- and it doesn't wait to ask permission from those it's making obsolete.

That doesn't mean to say that in 2023, journalism will be obsolete. I'd argue that it will be even more relevant, perhaps more important than it has ever been. Right now, we're in the infancy of what I'm creatively calling the "data age". A decade from now, society will probably look back and laugh at us for the way we willingly surrendered our personal data to Facebook, Google and co, without really worrying. We don't really appreciate the scale and the scope of what we're giving away.

Prediction #1: Google and co will still dominate, but the news industry can add value via curation and verification of information.

The dominance of Google, Twitter and Facebook will continue -- they've laid the groundwork for storing and retrieving all kinds of information, to the point where it's not financially viable to even consider competing with them. Their monopolisation of the information space means their value lies in being able to do things with that data.

How is this relevant to journalism? Well: isn't that all journalism really is: doing things with data? You don't have to be writing about government statistics or financial scandals to be working alongside information: especially the "social" information Facebook specialises in.

This means that by 2023, large news companies need to offer something that Google and co can't or won't: human curation. Google pride themselves on the power of their algorithms: computational functions which decide the importance of various bits of data. Introducing human judgement into the process would be an admission of failure in their eyes; a weak link in the chain.

You can't automate all journalism, though (but that won't stop people trying). In a world where things are reported almost in real time, the problem of 2023 will be information overload. Actually, I'll correct that: the problem of 2013 is information overload. By 2023 journalism should have fixed that problem. I can't possibly keep up with everything that's posted on Twitter, nor do I want to. Good journalism can find the story there, work out what's relevant to me, and show me the parts I want to see.

This means, then, that the real value of future journalism is in verification, curation, and access. Even today when news is broken by "citizen journalists", people openly speculate about the truth of the story until it's "confirmed" by the mainstream press. A gas plant exploded up the road from my flat a few years ago and Twitter was the first thing I turned to for news: the papers didn't cover it for almost 24 hours but in that period, Twitter "witnesses" had variously labelled it a tube train accident, an electricity failure, a gas explosion and a terrorist attack. The papers, when they got around to reporting it, had access to the right people and institutions and were able to get to the source of what had actually happened. In a future full of armchair experts, the journalism of 2023 can cut through the information density to show us what's truthful.

Prediction #2: we'll see sharper, more focused journalism.

One thing we'll see in 2023 is sharper, more focused journalism. Gone are the days when newspapers were all things to all people. In a world where the news was hand-delivered to your door every morning, shipped to your town or village from the busy printing press every night, it made sense that your newspaper told you the news, but also the weather, the TV listings, the best new films and culture, and the football results. These days it's getting harder to argue for that monopoly: in 2023 it'll be gone. It's already becoming unlikely that a single publication can provide the majority of content its readers consume. People are used to shopping around for the things they want to read, especially when much of it is free online. Larger publications might have to scale back to be leaner and more specific to the most loyal or profitable parts of their audience.

Similarly, because the concept of a "physical" newspaper is already in decline, by 2023, a general-purpose print newspaper will be a kind of luxury or "boutique" product, a bit like bands today still putting out their records on vinyl or even cassette: aimed at specialist, even hobbyist followers with niche interests. The upshot of this is that the "narrative" of a daily paper is completely removed: nobody will read the entire day's news and features.

Again, even today this is already the case: every weekday the Guardian publishes enough words to fill the average Tolstoy novel. Newspaper websites can no longer rely on all their visitors being "Guardian readers" or "Daily Mail readers": traffic will largely come from referrals via search and social networks where readers are less interested in the specific brand of the newspaper they're reading as the content of that news itself. This is great news if you're a journalist but perhaps less great if you're a newspaper company.

Prediction #3: (some) journalists and their networks become more valuable and powerful than newspaper organisations.

This means that the real value for newspapers of the future are journalists with a strong network who know how to promote their own content. In the past, less popular editorial content was buoyed up by virtue of its inclusion in the larger body of a popular publication. In the future, individual journalists will become more empowered to own their content from production through to publication and promotion: you don't need a printing press any more to be a journalist, but a decade from now, you might not even need the newspaper company.

Prediction #4: hardware becomes even more fragmented; networks will be always-on.

Hardware is another interesting area: while we probably won't be strapping on headsets and goggles to view augmented reality eight-dimensional newspapers, we will certainly be using handheld, portable computers as our main entrypoints to the web. It'll be a given to consumers of the future that the content they read follows them from device to device, seamlessly. Leave the house and read half an article on the bus, then get to work and carry on from the same paragraph on a desktop computer. Similarly, we'll always be connected: the days when we had to stand, arm outstretched, in a field in Glastonbury just to send a text will be long gone. Fast 4G (or better) signal will be a given, even underground or in the sky.

In relation to this web of devices, the role of digital editors will evolve to become much more data-focused: specifically metadata about content itself. No more will journalists battle with CMSs to arrange pixels on screens to tell the day's news. Instead they'll focus on providing context and guidance to algorithms and functions which can in turn lay out pages for a variety of screens, devices and other contexts -- automatically. In a world where the concept of a single template for a page of content is almost completely impossible to define, we'll focus our energies on automating what we can't keep up with manually.

The always-on connectedness of the future means that breaking news will become a kind of arms race: some publications will choose to end hostilities altogether. News companies will have to choose between chasing breaking news and competing against less scrupulous sources who don't wait to verify or confirm stories, or accepting they'll never beat the social networks and instead focusing on in-depth analysis, verification and comment. Newspapers were originally created to tell people what had been happening: the future won't need them for that any more. This gives journalism a unique opportunity to act not only as the voice of record, but the voice of response.

Prediction #5: readers will replace editors (in some capacities)

Will the journalism of the future need editors any more? The great democratisation of the web means that users are becoming commissioner, interviewer and reporter. Sites like reddit offer an "Ask Me Anything" section where members of the public, celebrities, and, occasionally, sitting presidents of the USA turn up to ask questions submitted by reddit users. This community is self-organising: reddit's staff didn't create it and they don't police it. Users suggest interesting topics or people to talk about; users submit questions and vote for the ones they like best; users step up to answer questions where they've had interesting jobs or experiences. Users do the investigative work when a dubious source appears, often tracking down internet history to prove or disprove a particularly bold claim. They're doing the jobs of journalists without knowing it. Reddit gets over four billion pageviews every month.

In this kind of environment it becomes harder to justify the inner circle of privileged people who control the day's new agenda. Power is transferring away from institutions and directly onto content. The internet today is becoming a tapestry of connected links, with traffic crossing and diverging from every thread. Individual nodes on that graph attract readers and revenue, but those nodes are pieces of content, not entire publications.

Prediction #6: models of content become much more web-first and driven by information

The model of that content is also likely to be dramatically different in the future. For an industry claiming to be "digital first", much of the production work is still rooted in print. Newspaper websites use clunky content management systems to dump lengthy blocks of text into boxes, add a few tags and then press "Publish". In the future this will be exposed as dumb and unhelpful. Even today, forward-thinking news organisations are looking at how to re-model news to show its narrative: the articles that cover a specific event; the events that happen as part of a larger story; the importance and tone of these stories against other things currently occurring.

In ten years' time there'll be a much reduced concept of the "atomic unit of news". Articles will no longer have a single, one-size-fits-all form. We'll tailor content automatically to different audiences: the reader who has five minutes before their train arrives; the reader with half an hour on their lunch break; the reader who's an expert on this topic and doesn't need the context; the reader who's never heard of Oscar Pistorius and needs background. These newly-modelled articles can show you what's new since you last visited; show you the key events in the sequence; cut out the paragraphs of "explainers" you already read in the four previous articles.

Prediction #7: digital advertising will never replace print advertising in terms of revenue.

Digital advertising as a revenue source will perhaps never replace print advertising in terms of scale. The future of ad-supported journalism is therefore at stake: it seems unlikely that it can continue a decade from now. This could mean that all large publications eventually put up some form of paywall, but I think the trend in free digital products on the web will continue and the news industry will eventually conclude that it can never reproduce the model it had in the print days and find alternate revenue streams. Advertising will be much more focused and quality-over-quantity than it is today: even in the age of Facebook advertising where brands can aim their ads at ever-more-specific niche audiences, we still see newspaper website lumbering with dozens of unrelated and irrelevant ads. In the future there might only be one ad on the page, but it'll be the one most likely to engage you.

Prediction #8: user comments “below the line” become tamed (or neutered)

Reader interaction with papers could transform hugely: most newspapers today offer comment sections for readers to respond, and these are currently becoming fairly polarising among journalists, many of whom have called for comments to be disabled on their articles. We've also seen the rise of Twitter users being charged for contempt of court, with a gradual understanding coming into place that the "free speech" of the web isn't always as free as we think. Newspapers of the future may decide to revert back to a one-way form of communication, moderating all responses via the equivalent of the "letters page", rather than opening things up as a free-for-all.

Prediction #9: we'll learn how to “scale” journalism.

There's a term in software development called "scalability": it refers to the ability of the software in question to expand outwards to satisfy increasing demands. In software terms it means writing efficient code which doesn't put unnecessary strain on servers, and also the ability to deploy code into "the cloud", meaning that in times of high traffic, we can quickly add more server capacity to cope with the extra load.

By this definition, journalism as we know it today doesn't scale. To put it in software terms, it's a "legacy product". Once, it was feasible to have journalists around the world in various bureaux, each reporting on their patch and filing copy over the wires. The day's news was assembled in one location and sent off to the printers, who then shipped it around the country while the nation slept.

Today, though, this approach is no longer scalable. More and more of the world is becoming "connected": where once an Egyptian revolution would've been covered by foreign correspondents, we're now seeing local citizens tweet minute-by-minute accounts. The notion of a newspaper reporting events from 12 hours or more ago seems laughable in this information age: it's old news. The physical labour and financial burden of shipping newsprint around the world seems archaic and quaint. The old mechanisms can no longer scale to meet the challenges of modern journalism.

Luckily for us, though, the means of upscaling the news are already here -- we don't have to wait for the future. Where once an investigative reporter had to sift through six thousand documents to find the three important ones, now they can work with data scientists to find the stuff that really matters: or even open it up to their audience. This doesn't devalue journalists: in fact, it gives them more value because they have the ultimate say in what goes and what doesn't. They have the specialist knowledge of the subject, whereas the crowd of assistants might simply recognise something juicy and pass it on.

Prediction #10: there will still be journalism in ten years' time.

This period of change that we're in now is a perilous time for those old structures: it's hard to see how we get to this utopian future of curation, automation and, well, profitability. Nobody feels confident enough to make the first move and every mis-step is branded a failure. Journalism is too important to be tied down to nostalgia and traditionalism

I'm definitely a believer in the phoenix-from-the-ashes analogy, though: the internet may herald the death of many "offline" traditions and methodologies, but it also ushers in a thousand new ones, each with massive possibilities for change and engagement, in ways impossible even a decade ago.

Our choice as journalists is simple: we can lament the "death of journalism" as the things we once knew become a legacy, or we can feel reinvigorated by the possibility of what's just around the corner.

---

Endnotes: I'm much indebted to the work of C.W. Anderson, Emily Bell and Clay Shirky for their fantastic Post-Industrial Journalism, which I read midway through preparing these ideas and finished feeling like I'd have to throw everything out and start again. Also thanks to Theresa Malone at the Guardian for letting me bounce some of these ideas off her, as well as offering me feedback and even reading the damned thing of her own volition (unlike my long-suffering partner Maddy who was forced to listen to me read it out).