Plug.Direct: Optimising news content for independent platform publishers

Journalist using Plug.Direct

What is Plug.Direct, what is it for and how does it work? A brief introduction to the ideas behind the new app and how it uses the platform publishing model – to amplify the voice of independent media abroad, expand the reach of small publishers at home, and promote good practice & media ethics everywhere.

As you know – if you see a news article you like on Facebook or one you find on Google, to read it you click and follow a link to the article publisher’s own site. But Facebook & Google want you to stay, not leave. So they offer news organisations the option to place their stories on their platform in full so there’s need to go direct to the original source website for the news.

In exchange it shares some of the advertising income it gets from the stories with their original creators. It’s called Platform Publishing and Facebook’s version is called Facebook Instants. Google’s version is called Google AMP.

Big publishers blow hot and cold on platform publishing. The biggest global & national players don’t like it because it buries their branding and reduces visitor numbers on their own sites. Data suggests that a reader ‘lost’ to search & social almost never ‘returns’ to the publisher’s own website. Mid-sized local publishers envy the audience data search & social collects, but they can’t use to reach targeted community readerships as well as they’d like.

But stories published as Facebook Instants & Google AMP stories load up to ten times more quickly than regular stories on regular websites – an attractive offer to the average mobile web user –  and they get more readers.  Google, and in particular Facebook’s algorithms are designed to find people who will be interested in your news article, based on who they are, where they are, what they like, and when, how and why they like it. And Google & Facebook’s behemoth branding and audience retention powers pose less of a threat to small publishers more concerned with quality before quantity, and put accessibility & reach before brand promotion & bulk marketing.

For people with our interests – supporting independent journalists in difficult environments for a free press – there’s another advantage.

Hack Big Social’s user model, not its code

Plug.Direct tracks the general performance of the user's plugged stories. Users can add extra plugs for a fresh boost as they like.
Plug.Direct tracks the general performance of the user’s plugged stories. Users can add extra plugs for a fresh boost as they like.

If you get your news on smartphone, like half the world already, then platform published stories are vastly faster and cheaper in pay-as-you-go MB download. And  news and you are in one of a growing list of 62 developing nations, you can get access to Facebook and its news feed free on your mobile data plan, sponsored by the Facebook Basics system. The ethical issues raised by Facebook Basics are numerous, and its impact on Net Neutrality freedoms is controversial.

But many of these countries have deeply repressive governments and strict media controls, but also massively expanding smartphone usage rates and huge, free access to news on Facebook Basic’s news feeds. We want to help more journalists and media-active community groups to get on the platform in countries on the Facebook Basics list, especially where independent journalism is threatened.

If however you are an independent journalist, freelance or member of a small news organisation in Europe, Plug.Direct – which balances access to Google AMP & Facebook Instants in a single publishing package  – is still just as easy to use, and just as potentially advantageous. Fiction & non-fiction writers self-publishing on digital platforms like Amazon are already significant players in the e-Book market. Local businesses like bars, dog groomers, sports clubs and bakers thrive on the localised markets Google & Facebook foster: Why can’t self-publishing local journalists and small community media use platforms to similar advantage?

A simpler solution in mind

In theory Facebook Instants and Google AMP make it much simpler to get your stories to the people you want to reach and want to read your work. In practice it’s a substantial commitment in tech & admin time. And you have to pay, even if you make some of the money back from the advertising attached to a popular story. Plug.Direct is a simpler, ‘pay-as-you-go’ solution for small media and independent journalists.

Screen showing the number of €5 Plugs available to promote the story on social media. Links to top up page.
Screen showing the number of €5 Plugs available to promote the story on social media. Top up if you need to.

The plugs are the payment. You buy a plug at €5 each and you use it to promote your story across Facebook & Google . One plug will get your story to scores of specially chosen readers, two plugs to twice as many, three plugs to three times, and so on. Think of plugs as vouchers.

Here, I can use up to four plugs or €20 worth to promote the story again to another story or the same again. Or I can go ‘all in’ and use all 14 available. The idea is to use the Facebook & Google algorithms to find more people with a potential interest in your story, so they can get the chance to read it.

The app’s algorithm learns from the analytics, so if you plug the same story again, it will use the results to send it to more receptive audiences. The trick is to pick your moment when you judge the moment right, then check how your story is doing and plug it some more while attention is rising.

Ride the algorithms. Catch a wave of public interest in your work by pushing it out to new audiences, as Plug.Direct detects them emerging.

The app also attaches advertising to the story for you. You get a cut from the money earned – not much for most users, but a bit. The money that you get back is converted into plugs and the plugs added to your account balance. Supporters of good journalism can contribute plugs to news organisations whose work they respect, especially in countries where distribution of conventional news is difficult or obstructed.

Screen showing the impact the Plug has had on the plugged story today.
Screen showing the impact the Plug has had on the plugged story today.

The screen tells you that the stories that you plugged earned €1.21 from 234 interactions with the page that included 21 ad clicks. That earned money is added to your account. When the total reaches €5.00 you get another plug added to your account.

Here in this example, as you can see, I have 12 plugs, which at €5 each comes to €60. If you are falling short, just click the top up button and buy some more plugs. As many as you like, up to a maximum of 20 a day, or €100 worth.

The plugs will be used on Facebook Instants, or Google AMP.  It’s very much not a Facebook or a Google product, despite the donation Google has made towards Plug.Direct’s concept development from its Digital News Initiative Innovation Fund. Both corporations are of such size, they come with an organisational unwieldiness than can make them unreliable, inconsistent business partners. Plug.Direct is designed to adapt iteratively to changes in corporate practice, second by second, by responding to changes in reach and performance.

Independent journalists working within their elaborate information ecosystems must learn to hack Big Search & Social’s user models, not their code. The independents that succeed will be the ones that can ride the system, not depend on it. We aim for a capacity to adapt quickly to the kind of rapid and unexpected changes in Facebook & Google’s terms, technical permissions and trading practises that routinely trip up the pair’s business partners.

What are the GP&P Scores, and why does Plug.Direct need them?

You’ve heard of Fake News on Facebook, right? The Plug.Direct GP&P (Good Practice & Provenance) Scores are a partial answer to this problem and a quality benchmark for the app’s users. It’s designed to make it difficult for Fake News writers to use the app, by blocking the anonymous users (or bots) and fly-by-night spoofers that post most Fake News stories.

The app’s spider crawls the app user’s website for evidence that the writer is a known person – that he or she is who they say they are. It looks for Twitter accounts, website registration details, links to other sites, and 50 other specifics that measure indicators which on the balance of probability means that the writer is legitimate – their actual story aside.

Screen showing the provenance score and general performance of the user's plugged stories.
Screen showing the GP&P score and general performance of the user’s plugged stories.

So the app also tests the author’s story as well,  by looking for Good Practice – evidence that the story has been professionally presented and sourced. Here the app ticks off a further list of 50 specifics evidencing good journalistic practice – fair copyright declarations, use of appropriate credits and captions on attached media, consistent metadata, working “contact the author links” – collective evidence that, on the balance of probability, suggests that the author is a professional. (A third phase of development proposes a machine learning enhanced test of the story’s use of sources, level of attribution, and another layer of provenance placing it among stories on the same subject from other news outlets.)

Added up – that’s the Plug.Direct GP&P score, X out of 100. As a journalist you can improve your score by adopting good journalistic practices – we’ll show you ways. The app recognises professionalism and raises (or cuts) the score accordingly. If you score less than 33, the app won’t publish your story. It is competitive, but the way for a journalist to join the Plug.Direct community and top the list is to adopt more ethical practices and write better-sourced, more accountable and up-to-date stories, judged by transparently applied criteria.

Sign up for an Autumn 2017 beta test account at the Plug.Direct website.

Plug.Direct’s initial concept development is supported by a grant from the Google Digital News Initiative’s Innovation Fund.

An Emotional Design plug-in for data sets lacking in emotion


Should data scientists & artists ‘design in’ engineered capacity to induce human emotional responses to otherwise impenetrable data sets?

In his book Emotional Design: Why We Love (Or Hate) Everyday Things, the former Apple exec and design critic Don Norman discusses three levels of emotional responses to product design — “visceral, behavioural and reflective”.

As Wired Magazine’s review noted, for years Norman had done nothing but rail against hard-to-use products. Yet in his 2004 book, he admitted that emotional attachments often trump practicality. Creative design can trigger a personalised emotional response to even the most hard-to-use, superficially impractical objects.

As a means of communicating information, bulk data sets are probably as hard-to-use and superficially impractical as it gets, even data on a subject as essential as climate change. In fact you don’t need to be a rocket or a data scientist – and Donald Trump is neither – to recognise the limitations of climate change data as a means of generating engagement with the issue.

An alternative approach: Climate Symphony turns hard data on climate change into a symphony, telling the story of what climate change means through sound. It uses data sonification, a technique designed to make statistical analysis more accessible to the mind via sound and the ear, just as traditional data visualisations like graphs make data accessible through imagery and the eye.

Disobedient & Forma Arts: Climate Symphony Performance #2degreesfestival. 7.30pm, Saturday 17 June, at Toynbee Studios, 28 Commercial Street, London E1 6AB.

“Research shows that sound touches us in inexplicable ways,” Leah Borromeo and Katharine Round of Disobedient Film Company, co-creators of the work with composer Jamie Perera. “By using music,” they told Anita Makri of SciDevNet, “the hope is to create an emotional response to something that for many might look meaningless on a page.”

Norman, of course, was talking about emotional response in the context of product design. How we relate to a visual element, and how to create something that is more appealing, effective or well received via design, by design. But his perspective, coming from a former Apple exec, is instructive in a digital media era when the case for challenging climate change is suddenly less appealing, effective and well received in some crucial quarters.

I’m writing here, gathering thoughts ahead of participating in an all-day workshop on data sonification and its uses for creative advocacy. But as a journalist and media technologist, looking at the premise of Disobedient’s unchallengeably attractive and principled work set me thinking about the whole idea of generating emotional responses in a creative environment, for the purposes of creative advocacy in the public interest.

Politics, media, communications and advocacy for change have been transformed by the spread of media techniques led by emotional design – the capacity to create an emotional response to a packaged information-based argument. Shockingly, we’ve discovered that the technique works even better – and even more profitably – where the package is light on information and argument to start with.

Whether manifested as politically motivated disinformation, ‘fake news for-fun-and-profit’, or deep state subversive ‘information operations’ – the results are proving as catastrophic for the global information environment as human intervention and climate change has for our planetary space.

The “visceral level” is the first impression, a purely instinctive emotional response to information. Emotional designers set their ‘information delivery mechanisms’ to trigger a visceral emotional response – to leave users feeling something. But what kind of emotional response is the emotional designer seeking? Anger? Fear? Grief? Compassion? Empathy?

We need to ask which one, and furthermore, why that one in particular? How does the emotional design of the delivery mechanism secure the desired response at the behavioural level? A successful visceral behavioural response leaves you wanting to interact with the information. And a successful information delivery system should leave you wanting more of the same to act on again and again. Should we use these techniques more cautiously?

Major charities and media organisations have aggressively monetized the generation of visceral behavioural response. Now they must constantly ramp up their trigger mechanisms to sustain repeat business with ever more desperate images of starving African children and ever more improbable exposes of paedophile rings in DC pizza parlours.

What does this say about using cultural practice, such as the creation of symphonies from data sets, to trigger emotional behavioural response? Norman’s emotional design is especially relevant to information tech at behavioural level when it comes to “the UX”, the ‘user experience’ – what users physically need to do to successfully interact with information. And latency. How obstructive to this successful interaction is the information delivery system function’s intrinsic design? Should we ask what is good UX and acceptable latency in a symphony?

This kind of behavioural level context is key to effective emotional design. At a basic level, regardless of motive, in creating something that is more appealing, effective or well received. “At the simplest level, the design of a business card implies a behaviour,” says Norman. “To take it and file the information away for later.” What information can tonight’s audience “take away and file for later”? How will its UX perform? How will its latency stand up to critical review?

This is also a test of Norman’s third level – the “Reflective” – in his view, the “highest” level of the emotional-visual thought process in visual design. It’s the interpretation and understanding of creatively expressed information combined with feelings about it. At this stage of thought, a person determines and creates a lasting impression of something. Is it memorable? Does it leave a lasting impression? Will he or she reference it later?

Memorable, yes. Tonight’s performance in London includes live illustrated performances by leading sound artists Kate Carr and Lee Patterson, who use location recordings captured in the field in the composition of their work.

The performance includes an audio-visual exploration of Ólafsfjörður, a small fishing town in northern Iceland. The set explores the way weather patterns are enmeshed in the experience of home and place. It combines field recordings of iced fences, snow, ice melting, wind, boats and blizzards with recordings of traditional Icelandic sea songs performed by the town’s choir.

Referrable? We’ll see. The idea behind ‘Climate Symphony’ is to translate hard data on climate change into a musical composition that engages the public — encouraging people to question their feelings and the stories behind the data, and create a conversation. “In a world where we’re saturated with hearing the same messages,” say Borromeo and Round, “any way to engage people with a subject [as] important [as] climate change is worthwhile.”

Disobedient & Forma Arts: Climate Symphony Performance #2degreesfestival. 7.30pm, Saturday 17 June, at Toynbee Studios, 28 Commercial Street, London E1 6AB. Supported by Arts Council England and Sculpt the Future Foundation. Book here.

What is the sound of a dying planet?

Climate Symphony:What is the sound of a dying planet?

A weekend creative workshop in London aims to turn scientific data on climate change into sounds, and then into music.

What is the sound of a dying planet? Might it sound a little like a musical composition, parsed from the billions of bytes of data that can be mined from decades of change climate change research? An eclectic group of climate scientists, data analysts, journalists and sound artists will be gathering this Saturday in East London’s Toynbee Hall to find out in a collaborative process open to all, with a live performance to finish.

The experts, plus artists Freya Berkhout, Danny Keig and Jamie Perera, will help members of the public select and turn climate data sets into sound, creating digital instruments for a personal Climate Symphony. They will use data sonification, a technique designed to make statistical analysis more accessible to the mind via sound and the ear, just as traditional data visualisations like graphs make data accessible through imagery and the eye.

‘Sonified’  datasets – data changed into musical notes, timings and phrases – are raw material mined and refined into musical scores performed by the people, places and things reflected in that data. Turning depersonalised, abstracted data representations of looming catastrophe into a musical soundscape, the sound of a dying planet, it’s a data extractive industry event of a different sort, if you like, featuring human intervention of a more constructive kind.

Working at the intersection of technology, data journalism and art, Climate Symphony is a fully scaleable project with a vision to use sound as a journalistic tool – social engagement through sound. Devised by filmmaker Katherine Round and Leah Borromeo of Disobedient Films, in collaboration with composer Jamie Perera, and co-produced by Forma Arts and Disobedient, the workshop runs 11:00-6:00pm this Saturday 17 June. The £10 ticket includes admission to a 7:30 performance. It is nearly fully booked, but check the Facebook event page or go direct to the ArtsAdmin booking page for updates on availability.

Tickets are also separately available for the evening event at 7:30pm. Curated by Forma, the evening event will also feature live performances by sound artists Kate Carr and Lee Patterson who use field recordings in the composition of their work, often from environments in which climate change is taking effect. Find out more here on the Facebook events page.

Disobedient & Forma Arts: Climate Symphony Performance #2degreesfestival at Toynbee Studios, 28 Commercial Street, London E1 6AB. Supported by Arts Council England and Sculpt the Future Foundation.

Audience data strip miners are failing to deliver local advertising markets

Arthur Greenberg, EPA

Small ‘hyper-local’ news publishers and community journalists are no better placed than major corporate media when it comes to accessing hyper-local audience market data granular enough to make their news businesses sustainable. For that they have to go to Google & Facebook. The question is, how?

It’s a sad truth: Advertising technology designed to find people for advertisers to sell stuff to, is easier and cheaper to produce than content that attracts people who want to buy stuff from advertisers. Cheap ad tech thrives on cheap scale – trawling millions and millions of pages that cost almost nothing to produce and can be strip-mined of data and re-targeted with ever more high-volume, low-value adverts.

It’s why US ISPs persuaded Congress and US President Donald Trump to let them to sell their customers’ browsing data without their permission. By pumping out cheap content in massive quantities, the US ISPs can finally collect user data in bulk, and develop new ad tech to trawl it for what’s trending, not what’s good, in the way that market behemoths like Google and Facebook already do.

The worst offender, AOL CEO Tim Armstrong, headed Google ad sales before taking over AOL in 2009 to bulk buy audience targeting ad tech start-ups. “None of this was a secret when Verizon bought AOL — the Ad Age headline was straight-up ‘Ad-Tech, not Content, Is King in the Verizon-AOL Deal’,” wrote The Verge’s Nilay Patel and Ben Popper this month. “Can’t be any clearer than that.”

Patel & Popper argued that Trump’s action will free Verizon and other ISPs to take data generated by the tracking super-cookies they stick on its customers as they use their broadband network, and mash it up with AOL’s ad stack. This “hyper-targeted marketing information” will be effectively unstoppable, because Verizon will own both the pipes and much of the content flowing through it.

Cheap scale digital advertising like this is a blight on the internet. A new industry study estimates that $12.48 billion of ad spending in 2016 was fraudulently picked up by bots rather than real customers or was not properly loaded on a page where a human could view them. This made almost 20% of the $66 billion spent worldwide on digital ads last year fraudulent or misreported, and valueless.

Even so, the Interactive Advertising Bureau (IAB) has reported that the third quarter of 2016 saw a total of $17.6 billion spent on digital ads, a 20% increase from the same period a year earlier. This was a reflection of growth in mobile and video and “marketers’ continuing trust in the internet’s power to connect with today’s audiences”.

There is less trust in ad-tech’s ability to facilitate that connection. Some 29% of automated ‘programmatic’ ads miss a target, as opposed to 12% of ads sold directly by humans, yet eMarketer predicts that by 2018, over 80% of digital display ads will be programmatic, up 37% from 2014. The problem is exacerbated by a lack of accurate metrics and viable regulation.

Advertisers need data on user preferences and behaviour. Even if the ad tech misreads the data,  Google and Facebook has more of it to interpret. Facebook tracks what its 1.5 billion users see and click on in ways that other corporations like Verizon can’t yet match, even if by buying up ad-tech start-ups and accumulating its own mega-stores of customer data, Verizon hopes that it can.

The best advertisers can hope for is better analysis, Dan Jaffe, of the Association of National Advertisers, told Axios: “Because digital and mobile are growing faster than any other industry, the regulatory and foundational steps that you see in more mature media — like good metrics reporting — didn’t necessarily happen immediately. Now we’re making major efforts to catch up.”

What happens if they fail? “I expect even more ad blocking and less trust of digital advertising, which ends up harming everyone that matters, including marketers, publishers, and importantly consumers,” Jason Kint, of Digital Content Next, a trade group seeking better online advertising, told The Verge. “They (Verizon/AOL) know what they’re doing. I just think it’s bad for the marketplace overall.”

Digital Content Next found that Google and Facebook are currently  impossibly unbeatable in the digital advertising market, scooping up 99% of $2.9 billion in new advertising growth in the third quarter of last year, with Google making up about 54% of the total and Facebook about 45%, leaving just 1% for everyone else.

If there’s to be a future for digital-first advertising, it lies in developing ad tech that delivers better quality ads – that is ads that are more directly reflective of the interests, locations and needs of their target audiences – and a regulatory and foundational environment that advertisers & audiences can trust.

For this, we may be in for a long wait. In the meantime, how do we respond? Cheap scale programmatic ads will be cheap and profitable enough to sustain short-term growth for businesses with no connection with their audiences, but it is precisely because of this that programmatic has proven unsuitable for small or locally-owned enterprises.

Local businesses cannot afford a 20 percent failure rate on their ads, so in the short term they are keeping local print media advertising viable for a while. “Hyperlocal” news organisations can still thrive in areas where high-speed Internet access is often limited, but the wider rollout of 4G mobile services will eventually drive these audiences to fetch their news from smartphones.

The Adirondack Daily Enterprise in up-state New York was founded in 1894 and counts a print circulation in the few thousands. It continues to draw advertisers, says publisher Catherine Moore, thanks to reporters with a “deep commitment, sense of purpose and engagement in the community in which they live”. Could that attitude reap audiences and advertising income on Facebook as well?

Google & Facebook will inevitably find better ways to trawl their huge customer data sets for actionable intelligence they can sell to local advertisers targeting local audiences. The reality is that cheap scale advertising is too blunt a tool for “hyper-local” news outlets, but like ISPs like Verizon, the hyper-locals lack the data sets and the ad-tech to target local audiences precisely enough to make digital advertising worthwhile.

That privilege currently remains exclusively with Facebook and Google. Tapping into that data and making innovative use of it is the challenge.

Fiction & non-fiction writers that self-publish on digital platforms like Amazon are already significant players in the e-Book market. Local bars, dog groomers, sports clubs and bakers thrive on the localised networks Facebook’s digital platform creates: Why shouldn’t self-publishing independent journalists or small media organisations use platforms to similar advantage?

Plug.Direct is a content optimisation app for journalists, specialist writers and ‘small media’ moving into professional platform publishing on search & social. It offers freelance journalists, expert writers, activists and community media entry-level access to the Social Network’s computing power, underwritten by a quality score – a test of provenance and professionalism to help distinguish (not define) good reporting. Its development is supported by the Google Digital News Initiative’s Innovation Fund.

Empower the critical voice of quality on social media

Solving social media’s crisis of confidence in its role as a global news player requires greater access and renumeration opportunities for independent journalists of proven experience.

Strategies of empowerment and a critical approach to imbalances in power relations should come naturally to the principled, socially aware journalist. The media institutions that employ them take a different view.  Empowerment keeps them on the upside of that power imbalance, at the very least it sustains their struggling business models a while longer.

Yet those who analyse, report and seek to remedy challenges to freedom of expression rights, the focus is on raising the standards of media institutions, not empowering media workers. And by setting those standards against legal or quasi-legal frameworks that are nominally or cosmetically easily met, the institutions can ignore the kind of imbalances of power that lead to wider human rights violations – even if their journalists do not. Cannot.

The answer to social media’s crisis of confidence in its role as a global news player could be in providing greater access and remuneration opportunities to more independent journalists and news producers of proven experience.

We need a shift of focus from mainstream media institutions to these kinds of journalists. We need to ask if personalised journalistic relationships with information networks are a more sustainable route to an effective response to imbalances of power.

This suggests convergence as a peaceful, but subversive accommodation with corporate telecom & social media networks.  Though governed by quasi-legal frameworks of their own, in far less transparent and consistent fashions, social media platforms nevertheless raise the prospect of easier access to audiences and potential income streams on mobile news. Mobile is already the dominant means of news exchange in many countries. Mainstream media may regard publishing on mobile social platforms with suspicion, but smaller publishers have little choice but to take the jump.

We need to look at tools that facilitate the ‘exploit’ – disruption from within – hacking the corporate user model, not its code, using tools and techniques that can disable or subvert the negative applications of corporate media network publishing protocols, and make the most of the user model’s positive functions.

One of those potential tools is Plug.Direct, now under development by the digital media R&D shop Vivarta, with the support of the Google Digital News Initiative Innovation Fund.  Plug.Direct aims to empower and sustain professional working journalists and media actors outside institutional frameworks, but inside corporate publishing platforms.

Platform publishing’s USP lies in its promise of bigger audiences and speedier delivery combined, presenting major opportunities and challenges for small media publishers. It should offer independent journalists and news producers access to expanding audiences increasingly reliant on mobile technology for up to date news.

At a time when the capacity and willingness of the social media giants to manage quality of information across their networks themselves is in doubt, independent journalists and news producers of experience and proven reliability need that access. Pages with rich content that work alongside smart ads that load nearly instantaneously – and faster than regular mobile responsive websites. A means to reduce the cost in time and money that mobile users must bear before downloading, digesting and sharing news.

A network is a vital working space for a journalist, media-savvy expert or specialist activist, acclimatising, mapping and collecting information across a network topology that he or she has made her own ‘beat’. It needs to be made both sustainable and accessible.

Discretion is the better part of valour in the war on extremist narratives

John Kerry at the White House Summit on Countering Violent Extremism

There are no good options for journalists co-opted by political agencies into joining the ‘war’ on violent extremism and propaganda narratives online – only bad or ineffective ones. Why not just walk away?

It’s one of those opinions that as soon as heard, just sound true. That against their better judgement the media development sector feels pressure from their donors and the authorities to participate in government programmes for “countering violent extremism” (CVE) and propaganda.

The  Global Forum for Media Development (GFMD), the sector’s de facto ‘trade body’, believes it is a threat. It fears its members may be pressed to contribute to CVE programmes, “much against their will and in contradiction to their own ethical standards”. And to help address the problem, it has organised a session at next week’s RightsCon in Brussels, to come up with a set of ‘Do-No-Harm Principles & Best Practices’ in response.

In fact a well-researched 2016 report by Courtney Radsch for the Center for International Media Development (CIMA) found little conclusive evidence of a significant trend in donor funding favouring CVE. There was increased sensitivity, notably in Lebanon and in the wake of major European terror attacks, but it was hard to track a wider impact. Increased funding for such programmes appeared to be coming from new sources and budget lines.

The media itself is facing a score of existential crises; the incredible growth of social media, its own declining credibility, ‘fake news’, political interference, worse and worse, and more. No surprise then that media development – international aid in support of free expression, media diversity and the independent media’s contribution to democratic discourse – should share in the collective soul-searching by the media as a whole.

The RightsCon session aims to set Do No-Harm Principles & Best Practices for the media development sector in addressing the ‘Preventing Violent Extremism/ Countering Violent Extremism/ Counter-Propaganda (PVE/CVE/CP)’ agenda. In truth the CVE agenda is losing political traction in the US and UK, and appears to be failing in Europe, where dozens of lives have been lost to terrorists driven by motivations far more complex than online chatter.

The idea that online propaganda is radicalising a generation of isolated, angry killers is a pervasive one. But as University of East Anglia researcher Kate Ferguson notes, there’s little evidence to prove a direct link between propagandised radical ideas and terrorist intent. Even the FBI is cautious about using evidence of interaction with radical ideas online as a clear predictor of terrorist acts, according to internal papers just published by the US Brennan Center for Justice.

Ferguson’s study, published last November, led her to question both the perceived problem and the proposed solutions. “The hypothesis,” she writes, “that VE (violent extremist) narratives or the real life threat of VE can be countered by an alternative set of communications remains unproven.” UK CVE literature and policy strongly focuses on the language of counter-narratives, she says, yet “a common understanding of this relatively new lexicon has yet to emerge”.

A greater fear is that CVE strategies are a cover used by the authorities for collecting actionable intelligence in the war on terror. The Brennan Center cites another internal FBI document that refers to CVE as a way to “strengthen our investigative, intelligence gathering, and collaborative abilities to be proactive in countering violent extremism”.

The authorities in the US – and the UK, where the government’s CVE strategy Prevent has been similarly accused – maintain that “investigations or intelligence collection” are “not the goal of CVE efforts,” as the Obama administration said in October 2016 (though it’s clear that Donald Trump has a different agenda.)

Whatever, most media rights and free expression advocates are suspicious of the whole argument. Academics Matti Pohjonen & Rima Ahmed call on people to put “critical distance” between themselves and both “utopian and/or dystopian conceptualisations of digital technology and conflict”. Writing in the journal Security & Peace they described a new ‘dispositif’ of risk; pre-emptive action targeting the imagined dangers of digital technology.

At this point it may appear to many that the simplest thing would be to just walk away from the CVE & counter propaganda agenda. But the GFMD still calls for “clear and proportionate principles” to guide the media development community when contributing to CVE programmes. Media development expert Joan Barata Mir outlined the key objectives in a paper circulated to GFMD members ahead of the session. It aims for a stable discourse between media institutions, donors, political institutions and authorities – one that protects digital rights and media freedoms.

The CVE & counter-propaganda agenda, Barata Mir correctly says, is generally seen from the perspective of counter-terrorism operations, or the promotion of tolerance and non-discrimination. “(Its) impact on digital rights, activism, investigative journalism, as well as media development has not been properly considered yet on such (a) multi-stakeholder basis.” Barata Mir also rightly cites concerns about the role of digital intermediaries, the internet service providers, regulators and search & social corporate giants. They are under pressure to act more directly to cut off violent extremist content.

This raises concerns that wider free expression rights may be put at risk. He poses the question: “In what ways can the media development community, media freedom and freedom of expression activists, media regulators, and digital rights groups collaborate further in strengthening the ethical approach towards (CVE & counter propaganda)?”

This is a difficult one to answer for media rights defenders, many who will feel that the only truly ‘ethical’ approach to CVE & counter-propaganda is to have nothing at all to do with it. But this is ‘media development’, regarded here as a sub-set of international development assistance, not exclusively human rights or free expression rights and not specifically media rights either.

“The current consensus-based development system is dependent on reaching broad agreement among highly diverse political cultures,” James Deane of BBC Media Action has argued. “Such a system does not provide an effective platform from which to devise meaningful strategic action on an issue as politically charged, and apparently divisive, as integrating support for free media into development strategies.”

It is difficult to use theories of aid & development practice as a means to rationalise the competing demands of responsibility, consensus, representative democracy and freedom of speech, in considering journalism’s duty to all four. Barata Mir has a go anyway, co-opting academic Mary B Anderson’s “Do No Harm” concept. Her thinking recognises that international assistance in crisis situations, no matter how well-meaning, always have unevenly shared and unintended consequences, both good and bad. Thus the impact of aid is not neutral in its effect on whether conflict worsens or abates.

Under Anderson’s lights, supporting CVE & counter propaganda in a conflict setting “can reinforce, exacerbate, and prolong the conflict. It can also help to reduce tensions and strengthen people’s capacities to disengage from fighting and find peaceful options for solving problems.” Often it will do some of both. In some ways it worsens the conflict, and in others it supports disengagement. “But it all cases,” she argues, “aid given during conflict cannot remain separate from the conflict”.

So to borrow from Anderson, aid workers and media development NGOs alike should:

  • Try to identify local actors and local networks that can promote peace.
  • Design their aid programmes to support local peacemakers and reinforce positive networks.
  • Be aware that judgments identifying local peace actors may constitute dangerous and inappropriate social engineering.

None of this can be said to be bad advice, even if you cite critiques of “peace journalism,” developed by the Norwegian media theorist Johan Galtung, which raised some of the ethical and political conundrums borne by the debate around media development and CVE today. But none of it sits well with CVE & counter-propaganda strategies forged in complex political situations either. All are almost exclusively focused on immediate threat before context, and evidence of Islamist extremist activity.

Ferguson was struck by the way this overwhelming focus on current jihadi violence framed the CVE agenda across academic and non-academic research. Recent research into far-right extremism, or lessons from Northern Ireland, the former Yugoslavia, Great Lakes, and other identity-based crises are rarely cited. “Instead of asking what role new media plays in facilitating violent radicalisation,” write Pohjonen & Ahmed, “we need to examine the context in which it has been imagined as something in the first place and with what consequences.”

In the end the most beneficial outcome of the GFMD’s meeting will be if it encourages a broader and more rigorous approach to gathering evidence about the true nature of the problem it poses. And whether the media development sector is really best qualified to solve it.

Universal Do-No-Harm Principles and Best Practices for Preventing and Countering Violent Extremism Counter-Propaganda – How Do We Get There? Organised by the Global Forum for Media Development (GFMD) for Thursday  30 March, 13:15-14:30, in the Arabesque Room, Ground Floor at RightsCon, The Crowne Plaza Hotel Le Palace, Brussels.


Quality journalism, ‘Fake News’ and the Zuckerberg Manifesto

Where does the system of Instant Articles, Facebook’s priority news source of choice, fit in with founder Mark Zuckerberg’s manifesto of ambitions for a more ‘social’ social network? 

Facebook should be supporting more journalists and better journalism, helping reporters to connect with communities online and off, by targeting audiences, personalising communication and offering proven accountability and in the process, challenge the ‘fake news’ phenomena on its network.

“Giving everyone a voice has historically been a very positive force for public discourse because it increases the diversity of ideas shared, said Facebook founder Mark Zuckerberg in his 5,000 word plus ‘manifesto’ released last Friday. “But the past year has also shown it may fragment our shared sense of reality.”

Zuckerberg’s post is worth reading in full, as Backchannel’s Steven Levy fairly noted, “both as a powerful and thoughtful individual’s attempt to grapple with a global ‘Winter is Coming’ moment and as a corporate strategy meant to link a company’s ambitions to a wider social movement that it portrays as a global benefit for all humanity.”

Other observers like Kara Swisher, interviewing “a very intense” Zuckerberg about his letter, thought his tone was “of someone who knows that the solid earth he has been standing on has drastically shifted recently”. And she suspected that Zuckerberg was trying to distance Facebook from responsibility for fakery.

“If I could wave a magic wand and get rid of all misinformation, I would,” he told her. “But people would still use some sets of facts, the true facts, in order to fit whatever bias they have.” In fact Zuckerberg picked out media ‘sensationalism’ as a bigger problem than ‘fake news’ per se, something that will give some pause. Facebook’s revenue system depends solely on page views, writes Frédéric Filloux, an articulate critic of the network and its ‘walled wonderland’. Facebook doesn’t care about news in the journalism sense, he warns. “News represents about 10% of the average user newsfeed and news can be cut overnight if circumstances dictate with no significant impact for the platform.”

But as long as Facebook does accommodate news, it will have to address the fake kind too, and using the algorithmic tools that surface it so readily. Elsewhere Dean Pomerleau’s Fake News Challenge focuses on discrepancies between misleading headlines and the actual texts as a means to identifying ‘fake news’. Though the Challenge’s likely efficacy is disputed, Zuckerberg cites data suggesting that a lot of ‘fake news’ is shared over Facebook on the basis of the headline alone; that is without actually reading the article itself. His data suggests that articles that are read and then shared should be treated more respectfully by the Facebook news feed algorithm.

The idea that algorithms might provide a means of distinguishing truth from fiction on a story-by-story basis has been widely challenged. But it’s possible that despite the failures of the past, the Facebook algorithm could quite quickly tag industrial-scale fakery by tracking its sources and pre-emptively reducing their spread across the network. Twitter did something similar to restrain pro-ISIS tweets on its own network.

Plug.Direct’s provenance scores test author and article against 125 measures of good journalistic practice and accountability. It leaves digital ‘breadcrumb’ trails for algorithm-driven distributors like Facebook and Google’s own AMP platform to follow, and hopefully favour the works with higher rankings in news feeds.

Steps to reduce a particular item’s prominence in Facebook and Google news feeds, rather than block it outright, would be a qualified answer to charges of censorship. But it’s hard to imagine Facebook over-regulating a service that thrives financially on sharing. And few things share faster or more profitably on Facebook than outrage, especially the made-to-measure kind.

So it’s hard to imagine Facebook living long on an exclusive diet of faux-outrage, no matter how profitable it was. Its corporate scale already makes it a favourite target for regulators and political agenda-setters. If Zuckerberg’s manifesto has a recurring theme, it is the desire to make Facebook lovable again.

In time, writes Amy Webb for Mother Jones, “algorithms will analyse your online habits, personal data and communication preferences. They will then merge your information with current news, package it in the way most likely to grab your attention, and deliver it in a format optimised for you.”

Will an algorithm pick readers up and take them where they want to go? “We’ve already seen that a lot of news consumers aren’t interested in being safely and reliably delivered to accurate news, says commentator Anil Dash in a well-argued analysis of Facebook’s business model.

Fake news exists in a interdependent media ecosystem that uses Facebook to feed virality and programmatic ads to raise income that is hard to beat. The New York Times tracked down an ex-student in Maryland who faked a story about Hillary Clinton on a fake site called Christian Times Newspaper, promoted it with half a dozen fake Facebook pages, and saw it shared six million times online, earning him $5,000 from programmatic ads in a few days. A BuzzFeed analysis found that pro-Trump fake stories before the 2016 vote received more shares, reactions, and comments than the top election stories from 19 major news outlets combined.

There’s plenty to be said about the shortcomings of Facebook’s approach to the challenges it swiftly creates but only slowly addresses. The Register was particularly tart about Zuckerberg’s manifesto and his “Technology Jesus moment”. “Extreme viewpoints will feel like the norm, not the outliers they actually are, warns Amy Webb. “And by the time we finally accept that our democratic internet is fatally hobbled by clever code, Big Data, and our own fallible instincts, it may be too late.”

Understanding the ties that bind global media development

The first results from an analysis in-process of the Global Forum for Media Development were released at its recent meeting in Jakarta. They make for interesting reading.

A superficial skim of Adam Saffer’s initial network study of the Global Forum for Media Development appears worrying. A handful of ‘cliques’ in the west holding the ties that bind its 50 members, in a network dedicated, it seems, to managing their smaller partners’ roles in their big ticket funding applications.

Presenting his results to the September 20-22 GFMD meeting in Jakarta, Saffer, an Assistant Professor at the University of North Carolina, takes a more positive view of the GFMD from his data. He sees his kind of network research as moving towards a “more relational, contextual and systematic understanding 1” of our social world, and the work of the GFMD itself. It’s an understanding that can only benefit the GFMD – indeed any media rights or media development alliance – and deserves a measured, patient response.

Saffer’s focus is on the relationships that exist among members of GFMD, and the opportunities and constraints that the network prescribes. He began his study of the GFMD with an assessment of its capacity to organise in support of its commitment to monitor and help deliver the UN’s Sustainable Development Goal on access to information (SDG 16.10). Survey participants were asked to identify those members whom they had worked with on GFMD activities in the past year. Thirty-one members, or 78%, replied.

The survey, Mapping the Global Forum for Media Development: A Network Analysis of Members, found that the GFMD network had a limited outreach. Only 2.6% of all the different possible relationships between the network members had been taken up. “Members are selective with their relationships to other members,” Saffer found. “There is a high degree of centralisation that indicates a majority of the relationships are coming from a few select members.”


In fairness the GFMD is not – not yet anyway – organised around relationships based on joint communications and advocacy in the way that the similarly structured freedom of expression rights network IFEX is. Media rights, especially journalist safety issues remain important, but the GFMD has evolved as a kind of trade organisation, where the ‘trade’ involves managing its members’ share of the estimated $625m provided by the world’s governments and foundations for media development support yearly. The objective is to do this more efficiently, with better results, and, naturally enough, to generate more business.

GFMD coordination goes a long way towards mitigating the core problems of the wider non-profit sector, critically described by social entrepreneur Jake Hayman for Forbes this month, as an “economy that rewards individual performance instead of collective impact, and short-term activities over long-term learning”. Yet inevitably it is funding partnerships that are the ties that bind GFMD members, not joint advocacy, or the various kinds of innovation research or situational analysis that it could do.

The number of larger partnership-based projects has increased in the last ten years, requiring members with the capacity to manage these programmes to international audit standards to take the lead. It’s these organisations that predominate amongst Saffer’s ‘cliques’. Not in the pejorative lay meaning, but ‘cliques’ in network terms, based on the number of common connections between network members.

His analysis revealed 109 ‘cliques’ in the network, primarily formed by six northern NGOs with the most connections and thus the most influential in the network. (GFMD Coordinator/Secretariat (n = 106), Free Press Unlimited (n = 101), Deutsche Welle Akademie (n = 58), ARTICLE 19 (n = 34), International Media Support (n = 34) and Ethical Journalism Network (n = 22). ) “It is just a matter of identifying who is connecting with whom to see the groups,” Saffer clarified later by e-mail. “In theory, these members of this group are critical for disseminating information or organising other members.”

While the amount cliques is not uncommon for a network of this size, as Saffer himself wrote later, it is not sustainable for six members to be members in a majority of the cliques. “An effort needs to be made to bring other members to the center of the network to form more relationships and reduce the load of these cliques.”

The key members of the GFMD, mapped.
Adam Saffer’s map marking the key members of the GFMD’s communications network.


From my perspective, a network as diverse as the GFMD ought to have a larger number of critical members sharing and organising communications, not least because all of them are NGOs in the communications sector. “Through engagement and taking on strategic responsibilities,” Saffer wrote, “members can become positioned more centrally and join existing cliques to begin distributing the overlapping relationships (i.e. cliques) more equitably across the network.” And as Tom King ‏of the Organized Crime & Corruption Reporting Project tweeted: “Whether or not you agree some orgs in @GFMD are over-mighty, it’s surely still beneficial for other orgs to step up and do more #mediadev16”

This is only the first round of studies. Saffer has not yet collected the data on shared donors, strategic partners and the balance of network ties between media development NGOs in the north and the global south. This will be quite a challenge, but one worth taking on. The network study doesn’t yet allow for the distorting effect that the demands and practices of the donors have on the topology of the GFMD’s network, and the formation of sub-networks.

“Charities are pitted against each other in competitive processes with their success judged by annual organisational growth and delivery numbers,” writes Hayman about the non-profit sector as a whole, “not long-term value, learning or dissemination.”

The donors’ preference for consortia to handle major investments in non-profit media development is welcome in many ways, as it facilitates bigger ambitions, but it does create sub-networks of its own. One example: The EU is granting 4.55m to a major three-year initiative to promote freedom of expression online & off, and access to information, in 11 target countries from next year. Yet the EU’s on-going bid process divides at least five GFMD members with shared objectives into different consortia (and doesn’t seem to include any of the several GFMD members already based in the list of target nations).

Interestingly the only ‘real’ funder tracked by the survey, the Open Society Foundation, which presumably relates its network ties to its grant portfolio, participates in fewer cliques (13) than most. That arguably illustrates focus rather than elitism. To its further credit OSF rated much higher than most when judged as a communication partner.

Competition for shrinking funds among the membership might qualify other members’ willingness to advertise existing relationships. But the ‘big six’ tend to have more demanding public disclosure requirements in their home countries, and a closer study of their networking habits might give a clue to problems and solutions. Meanwhile, out on the far edges of the GFMD network map, the smaller, mainly southern media development NGOs remain largely dependent on the bigger players to stake them in the game.

Ayman Mhanna
GFMD director Ayman Mhanna

And while the bigger northern NGOs rely on the smaller southern groups to give their project funding bids both credibility and viability, the majority of the relationships in GFMD mainly link the global north’s membership. This tends to be a one-way affair: Only 20.1% of the traced network relationships were reported to be reciprocated.

Nevertheless, of these relationships, Saffer found that their ‘social capital’ or value, was seen as fairly strong. “On three measures of social capital, members indicated others as being ‘somewhat’ trustworthy, cooperative and worth exchanging information,” he noted. “However, a few select members were identified as being significantly important to communicate with.”

More needs to be done before recommendations can be finalised. But the results so far suggest that the GFMD membership’s inter-relationships are based on perceptions of dependency rather than expectations of shared growth, whether in capacity, impact or relevance. That, if true, would be a problem. As said, a second round of surveys and a narrowing of focus is planned. In the meantime, presenting his opening assessments to the GFMD Global Forum in Jakarta last month, Saffer came up with three initial recommendations:

  • Continue meeting around the world: Members are able to establish new relationships and foster on-going relationships with other members at these meetings.
  • Create opportunities for engagement: A low amount of reciprocity was found, which suggests that many of the well-connected members are not reciprocating relationships to those less involved.
  • Need for regional leads: A closer analysis found that many members do not have relationships with other members from the same region, and in some cases from the same country.

But ultimately the true value of the study will be in further identifying the social capital generated by the network: The personal relationships; the emotional, material, practical, financial, intellectual or professional resources it enhances; the engagement and participation it fosters; and finally the trust it engenders, measured by standards of cooperation, reciprocity and non-discrimination.

(1) Borgatti, S.P. and Foster, P.C., 2003. The network paradigm in organisational research: A review and typology. Journal of Management, 29(6), pp.991-1013.

This article was updated on 10 October to include additional data and comments from Adam Saffer.

The media: Agenda makers or agenda changers?

Should the media development sector  prioritise journalism’s contribution to ‘good governance’ – ahead of more contentious challenges to unjust power structures that lead to violations of human rights?

Should the media development sector push agendas that prioritise journalism’s contribution to ‘good governance’ – ahead of more contentious direct challenges to the kind of unjust power structures that lead to wider violations of human rights?

A truly independent media is not instrumentalised easily or well, for any purpose. Even effective and sustainable media institutions may not be willing or able to deliver outcomes designed to monitor and mitigate the effect of imbalanced power relations rather than challenge and change them.

Should the media act more directly to expand opportunities and the power to make choices, achieved, as Rosie McGee & Duncan Edwards write, “through a collective, rather than individualised notion of empowerment that focuses on addressing structural inequality and inequitable power relations”.

Are media institutions are the right means to this end? Alternatively, can individualised empowerment have a collective effect on unequal power relations at a societal level? If so, are personalised, individual relationships with media networks a better way to that end, rather than through the collective representations of media institutions?

For as James Deane of BBC Media Action notes: “some of the greatest media and communication changes shaping governance outcomes are being played out at the societal rather than institutional level.”

The institutional response is illustrated by the priorities of the new UN Sustainable Development Goals, specifically SDG 16.10, which requires states to: “ensure public access to information and protect fundamental freedoms, in accordance with national legislation and international agreements.” Bill Orme, writing for the Global Forum for Media Development, SDG 16.10’s firmest champion, argues that with its adoption, “the onus is being placed on governments to proactively disclose information or to explain why certain information is not being made public”.

Alternatively, the focus on SDG 16.10’s priorities, arguably sidelines more subjective and politically fraught demands for freedom of expression rights at the societal level in favour of a more objective – and less politicised – offer of greater access to governmental and institutional information and data.

This value of this may be reflected in the response to the media development sector’s “supply driven strategies”.  These include access to information projects that Deane dryly observes, “are not necessarily being complemented by increased citizen demand for such information.” To borrow from Mike Gurstein, advocate of ‘community informatics’, this is not to argue against ‘open data’. But in the absence of a strategy to provide the means to effectively use it, only those with the means already will benefit, or as Gurstein put it, merely “empowering the empowered”.

Another problem is the shortcomings of the technology providing the “supply” of information feeding transparency and accountability (T&A) initiatives. Indra de Lanerolle & Christopher Wilson highlight the “failure of uptake” of this technology, digital tools declined by the very people the tools were intended to assist, demand-led initiatives without a demand.

The tools are often deployed based on limited understanding of their intended users in their intended contexts. Values are brought into question, says Deane, asking if media development strategies are inherently associated with normative, democratic ‘Western’ frameworks. “Media support initiatives are particularly vulnerable to the charge that they start with a set of assumptions of how they think things ought to work rather than how power, politics and government is in fact organised and how change can be best achieved.”

Whether normatively ‘western’ or not, the current international development sector’s view of democracy is always consensus-based, dependent on somehow reaching broad agreement among highly diverse political cultures. Regardless of values or motive, says Deane, “the search for consensus does not provide an effective platform from which to devise meaningful strategic action on an issue as politically charged, and apparently divisive, as integrating support for free media into development strategies.”

From a talk by Charlie Beckett of Polis, the journalism & society think tank.

Since 1991, the year of the seminal Windhoek Declaration, to paraphrase UNESCO, the media development sector “has understood press freedom as designating the conditions of media freedom, pluralism and independence, as well as the safety of journalists.”

How is this served by testing media development against its contribution to good governance? How does an expectation that the media build consensus as that contribution, help “make(s) sense of the evolution of media actors, news media institutions and journalistic roles over time,” to borrow from UNESCO again?

If the aim is empowerment – implying a focus on addressing structural inequality and inequitable power relations – can media institutions be effectively instrumentalised to deliver this outcome? I would argue no. A strategy of advancing empowerment through personalised, secure and direct relationships with information networks is a better means of achieving this end.

A shift from a focus on institutions to a focus on journalists might reflect the debate in the non-profit, state and international aid funded media development and media rights sector, historically the strongest supporters of independent journalists, activists and advocates.

We need to look at tools and techniques that facilitate “the exploit” – disruption from within, hacks that can disable, subvert or repurpose the functions of corporate media network protocols.

The likes of reality TV stars and terrorists are already rewriting hacking’ corporate social media’s user models user manuals in ways their creators never imagined. Similarly, journalists, activists and advocates can hack user models in ways that respect and preserve sustainability, ethical principles and visible social benefit.

Alexander Galloway & Eugene Thacker’s concepts of network protocols consider their capacity to “modulate” – to modify the organisation of nodes, edges, protocols and connectivity – and shape “different topologies of organisation and control”. Networks may also accommodate coexistent alternative topologies at any one time, even incompatible ones, they say.

“A whole new topology of resistance must be invented, that is as asymmetrical in relationship to networks as the network was in relation to power centres.” This could be the first step in realising “an ethics and a politics of networks,” capable of “identifying and critiquing protocol modulation, while fostering its transformative capacity”. This is, I believe, also a useful working space for a journalist, acclimatising, mapping and collecting information in a space he or she has made her own ‘beat’.

‘Beat’ reporters are specialists in a thematic or geographic area. Once central to mainstream media production, their numbers have dramatically reduced as revenues have fallen. Their contributions are enhanced by their engagement with their networks; their networks are enhanced in turn by their contributions. Media institutions fail to make this transaction sustainable. The challenge is to make this so, outside the institutions, but inside the networks.

Understanding the nature of network topologies underpins any strategy to help networked journalists reclaim journalism from institutionalised media networks. Based around supporting individual journalists, social activists and cultural advocates rather than media institutions, such a strategy would better fit a sector overwhelmingly going mobile, in need of more accountability, diversity, independence and sustainability.