Reflecting on the information market

Synaps sees itself as part of an “information market” and “knowledge industry” in crisis and in flux. We view this as an opportunity to innovate, through our slow-read publications, our experiments with data, the systematic unpacking of our own methodology and, generally speaking, a highly reflexive approach to everything we do and are part of. We are keen to discuss our trials, hear about other models, and engage with organisations willing to think through best and worst practices in what is, all told, an economic “sector” in need of reinvention. This series collects preliminary initiatives and findings to that effect.

Visualising trends in the information market: some data

The widely-felt sentiment of being swamped by a groundswell of information and analysis is easily visualized in data. Google Books’ Ngram Viewer tool, which displays the occurrence over time of words or phrases in a corpus of digitized books, offers an apt illustration. As one searches the corpus “English one million,” which contains a yearly random sampling of 6,000 English language books, covering the period from 1800 to 2008, the frequency of select keywords reveals interesting inflections: in recent years, “data” and “information” have risen quickly, surpassing such phrases as analysis, knowledge, understanding, story, news or insight. The latter terms all tend to designate relatively sophisticated forms of communication--in which meaning is essentially narrated--while the former connote raw or semi-processed material.

Indeed, “data” in itself is not immediately intelligible, unless it is structured and explained. The data that features in the above graph, for example, is nothing more than an endless list of digits separated by cryptic symbols, until a human operator chooses to display it in ways that tell the reader something. It bears noting, moreover, that data can typically be represented in infinite different fashions; most of these are more confusing than illuminating, thus increasing the role played by human determination. “Information,” meanwhile, usually signifies data that has been sufficiently organized to convey a message to its recipients, but the expression also suggests something straightforward or even crude--devoid of context, depth and interpretation. “Analysis,” “knowledge,” “understanding,” and the like, evoke higher levels of processing and added-meaning--not to say subjectivity, which is inevitably present across the board.

What the Ngram tracks is the respective prevalence of such concepts in our linguistic and cultural environment. The takeoff of comparatively unrefined content correlates with a perception of overwhelming information that doesn’t necessarily add much sense.

Data hyperinflation

An obvious suspect behind the growth in unrefined content is technology, and above all the internet. Between 1990 and 2008, the total number of websites went from one to 162 million; it is now well past 1 billion.

Another Ngram points to the jarring effect of digitization, leading the concept of “website” to suddenly overtake, in our cultural landscape, historical reference sources such as the “dictionary” and the “encyclopaedia.” Again, this graph starts in 1800 and ends in 2008, therefore attenuating the skyrocketing effect that would only have increased in the past decade.

Overall figures put this trend into much sharper contrast. The total annual production of data of all kinds, worldwide, is currently estimated to double every year; in other words, 2017 generated as much data as did the rest of human history, and each subsequent year will again multiply that by two. The processing power of computers doubles also, but only every other year. We therefore produce infinitely more data than our best technology can hope to process. Naturally, such computing speed in itself by far outpaces our human capacity to assimilate data, through cultural products, iterative scientific research and educational systems. Our own “learning curve” is anything but hyperbolic: although it cannot be credibly measured and visualized, an instinctive representation would posit at best a mostly linear uptick, at worst a chaotic squiggle. 

The knowledge onslaught

The problem is not that humankind has stopped or slowed its production of analysis, stories and, generally speaking, knowledge--quite the opposite. The annual number of books published worldwide, which oscillated around 100,000 throughout the first half of the twentieth century, rose steadily after the Second World War to reach 700,000 by the year 2000. More recent estimates place the number above 2,000,000, a figure that excludes the explosion of self-published books, which in 2015 had reached 700,000 in the U.S. alone. In 2010, Google Books calculated that the total number of books issued in history approached 130 million, while recognizing a serious dearth of reliable metrics. Individual libraries face a serious challenge as catalogues grow exponentially. 

Technology has obviously played a key role by lowering the entry costs to publishing, making it immeasurably easier for authors to draft, edit, format and circulate their work. New U.S. fiction titles, which hovered between 5,000 and 10,000 from 1940 to 1998, were soaring over 50,000 by the end of the 2000s. The rise of the internet has gone hand-in-hand with that of the computer, whose spectacular cultural ascendance--surpassing in prominence not just its direct ancestor, the typewriter, but also the transformative technologies that were the telegraph, the radio and the television--clues us into the democratization of content production.

Alongside books, a similar inflation affects periodicals, websites, institutional paperwork and, of course, emails, social media and messaging. It is striking, therefore, to note how little publicly available data exists on the expansion of this textual universe. (A rare break-down, in dire need of an update providing a comparative vantage point, can be found here.) Cisco Systems has aggregated data showing the multiplication of internet traffic by a factor of 100 million between 1990 and 2016, yet one is hard-pressed to find, for example, figures tracking the recent growth of Gmail exchanges, WhatsApp chatter, or Facebook and Twitter material.

Instead, users are mostly left to experience the swell intuitively, compounding their sense of overflow. Indeed, individuals both generate and consume growing amounts of information, but remain extremely, if not increasingly, ill-equipped to process it. “Aggregators,” “feeds,” “filters” and the like tend to funnel more content toward our bloated interfaces, but do little by way of sifting for quality and relevance. Meanwhile, people retain virtually no control, paradoxically, over the rising amounts of “personal data” they generate and surrender to digital monopolies such as Google, Facebook and so many others. A collective sense of disempowerment is the natural consequence of what could be described as an asymmetric relationship to data.

Digesting the glut

Our inability to absorb the volumes we produce, even when it comes to the most refined forms of writing, is discernible in what statistics do exist. The explosion of books published, for instance, is offset by stagnant or declining sales, which also means that books are ever-less read on average--often remaining captive to the author’s most natural audience or immediate entourage. A majority of academic articles, which now total approximately 2 million per annum, are said to be read by virtually no one aside from their authors, peer reviewers and editors. Ample literature in any scientific field also implies that incoming scientists spend evermore time reviewing it, before adding their stone to the edifice; knowledge production, which requires producers to first assimilate what already exists, thus tends to follow a law of diminishing returns that slows progress more than it catalyses it.

Another interesting Ngram narrates the rise and fall of knowledge-production entities in the decades of intense economic boom and geopolitical competition that followed the Second World War in the Western sphere. The curves indicate a change not in the actual numbers of such institutions, but in their centrality to an English-language cultural landscape--as books gave “research” unprecedented prominence, reflecting a moment of intense collective faith in scientific progress that seems to wane with the collapse of the Soviet Union.

The concept of “think tank” offers a somewhat deceptive counterpoint, rising continuously throughout the period under consideration. A notion first introduced by the great American capitalists cum philanthropists of the late 19th century, it developed during the Cold War and further flourished in its aftermath; more detailed and recent data, however, would show 2008 as a peak, after which declining funding (in part related to the concomitant financial crisis) caused a slump that most such institutions have experienced. 

All of these trends create the sense that we are only beginning to grapple with the evolution in our relationship to data, information, analysis, knowledge and meaning. Most likely, we are also at the beginning of that evolution itself. Much of the data presented above suggests that the past several decades have been significantly more disruptive than previous phases in a highly dynamic, centuries-old process of reinventing human communication and learning. The latest craze, known as “big data,” postulates that computing power will somehow make sense of it all. But data remains a raw material, incapable of speaking for itself; nor does it contain any intrinsic truth that automated processes can help reveal. Data is what human beings make of it. And that question remains largely, and luckily, open.

3 December 2017

The plight and promise of niche publishing: a brainstorm

Lebanon provides a fertile environment for creative forms of civic engagement, not least in the shape of publishing platforms addressing a significant gap in the broader information market. Synaps, which launched in Beirut in June 2016, is keen to draw on the wealth of inspiring local initiatives; although publishing is not a primary function of Synaps, issues relating to knowledge production, circulation and consumption are particularly relevant to it. We are grateful, therefore, to the following organizations for contributing their input during an informal brainstorm, held on 24 November 2017, on the challenges raised by “niche publishing”:

  • Daraj is a current affairs platform set up in November 2017 by mid-career Arab journalists seeking to produce and disseminate quality, independent reporting on the Arab world (represented by Alia Ibrahim);
  • Labne&Facts is an upstart social media-based platform (on tumblr., Facebook and Instagram) exploring the many facets of Lebanese identity “one post at a time” (represented by Marie-Jose Daoud);
  • Lebanon Support is a research center founded in 2006 and devoted to enhancing local civil society through original knowledge production, skills trainings, mentoring, and connecting job opportunities to candidates (represented by Lea Yammine);
  • Mashallah News (currently on hold) was launched in 2010 to report on the Middle East in ways that would challenge the “ruins, rockets, religious fanaticism” paradigm and resonate positively with a regional audience (represented by Micheline Tobia);
  • Raseef 22 is a fast-expanding platform inspired by the 2011 Arab uprisings, covering all 22 Arab countries with homegrown Arabic content “bringing personal perspectives to politics, lifestyle and culture” (represented by Zahraa Mortada);
  • Refugees Deeply, a branch of News Deeply, provides sustained, in-depth, comprehensive information and analysis on forced migration, opposing dogged focus to other media’s flailing attention (represented by Preethi Nallu, who gracefully hosted the event).
  • Also in attendance, while speaking in his own capacity, was Mansour Aziz, an editorial designer, online editor, and publisher  with a number of publications, to include the Arabic language intellectual and cultural platforms Bidayat Magazine and

The event’s original agenda

Lebanon is home to a number of individuals and initiatives that have attempted, each in their own way, to address the public’s desire for better publications. In other words: better substance, less noise. Is this ambition hubristic? Futile? Salutary? Most importantly, what practical lessons can we learn from our own endeavors, as we struggle to adjust to a shifting, rich, but increasingly overloaded information market? How do we successfully separate ourselves from the confusion?

  • Providing an alternative: In an over-saturated media market, how do we definitively separate ourselves from, and push through, the noise? What qualities and distribution strategies aid this process? And at what editorial and financial costs?
  • The monetization riddle: In a buyer’s market where free content rules supreme, how do we maintain the production of distinct and valuable content? How do smaller publications find a balance between funding and editorial independence? How can we compete with larger, better-funded teams? How do we manage the growing prominence of video?
  • Social media and the echo chamber: By nature, online advertising channels reward everything that is not “niche”, but they are a necessary evil. What does a successful social media strategy look like?
  • Successful engagement: Amid tweets, posts and newsletters, how much engagement is too much? What type of engagement proves most productive? What is ineffective? How do we best nurture and build a community with a vested interest in our publications?

This rendition of the discussion reflects the takeaways of Dylan Collins, who kindly organized the brainstorm, and Peter Harling, representing Synaps.

Perhaps most striking in the exchange was the level of professionalism and dedication displayed by participants—best captured by the quality of the platforms they have been running amid highly challenging circumstances. All can attest both to the existence of distortions and gaps in the information landscape, and to the reality of an audience eager to engage with specialized, substantive coverage. Zahraa Mortada mentioned that Raseef 22 had crossed the four million page views per month threshold. Preethi Nallu, along with Micheline Tobia, noted that certain analytics indeed point to readers’ willingness to invest time in quality long-reads like those offered by Refugees Deeply and Mashallah News. Both Alia Ibrahim and Marie-Jose Daoud testified to a strong public response to Daraj and Labne&Facts in their start-up phase.

‘Cracking the code’

Equally notable, however, was the consistency with which participants described economic precariousness and sustainability challenges. Each platform presents a specific configuration, but all rely to some degree on a mix of philanthropy and volunteer work, rather than product monetization via paywall and subscriptions. (Synaps, incidentally, is no exception, as we produce the bulk of our publications out of pocket.)

All in all, everyone agreed with Nallu that so far “no one has cracked the code” of a sustainable business model. She added that, for Refugees Deeply, a particular set of challenges relates to finding the right source of financial support, which “doesn’t compromise your editorial policy.” The Middle East presents an especially difficult environment in that respect, with an audience wary of politically-affiliated media and limited opportunities for funding respectful of editorial independence.

Bidayat’s Mansur Aziz remarked that even large, successful platforms in the West usually rely on an array of revenue streams, ranging from sales and advertising to endowments, direct or indirect subsidies, and various derivative products and services. “Things are messy in the region because you have none of the support mechanisms that underwrite intellectual output in the West,” he added, alluding to various streams of public and private funding that tend to underwrite professional media in the United States and European countries.

Participants stressed, at the same time, that good publishing is inevitably resource-intensive: “Signature content is expensive,” said Ibrahim. “You can’t create an amazing video without good solid reporting to base it on.” Crafting high quality templates for video and other content was in itself a major investment for Daraj, on top of the more obvious costs of journalistic work. She further noted the difficulty of tapping into advertising revenues, given the flightiness of the digital era: “advertising is fast-changing, making it too early to tell how exactly we can harness it in the future.”

Sustainability challenges are always constraining, and sometimes outright disruptive. Mashallah News recently paused its activities when a 2014 grant ran out, and is thinking through its future model. Daraj has at least temporarily suspended early efforts to raise venture capital, pending its own proof of concept. Labne&Facts, as a purely digital platform, has received some seed-funding originating from the “start-up ecosystem,” but continues to probe for a more long-term model.

‘Algorithms are overrated’

Alongside the puzzle of financial sustainability, participants noted the persistent challenges of reaching their desired audience in an era of ever-increasing media saturation. Most converged around the view that paid promotion offers little in the way of solutions, tending to create noise more than constructive engagement. Ibrahim thus noted that “boosting Daraj’s work as we kicked off grew our audience quickly, but I distrust the results. ‘Reach’ doesn’t convert into clicks.” Indeed, she added, creating buzz on a platform like Facebook can spawn disruptive, abusive “engagement” that detracts from substance rather than promoting it.

Similarly, Mortada has come to conclude that, in Raseef 22’s case, paid posts are mostly “bullshit.” Tobia felt that algorithms punished platforms that promoted content only irregularly, creating an incentive not to do so at all. Yammine noted that Lebanon Support continues to experiment with sponsored posts, but remains skeptical of the payoff.

Aziz, of Bidayat Magazine, elaborated on the idiosyncrasies of digital promotion, noting that Facebook’s algorithm will show or hide posts on a given individual’s timeline based on counterintuitive criteria: “A page I follow can have 100,000 likes and I never get to see its content because it all depends on how exactly people interact with it.” As such, he concluded, “social media algorithms are overrated. More than they assume, it’s the people on social media who decide.”

‘Getting to understand our audience’

Indeed, participants coalesced around the notion that high-impact publishing hinges on a dynamic, interactive relationship with the public. Each organization, in its own way, conveyed a high level of deference to its target audience: Mortada of Raseef 22 insisted on the imperative of “getting to understand our audience, through trial and error, and paying attention to people’s interests rather than our own vision only.”

Mortada described this process of balancing Raseef 22’s priorities with those of its public as a “constant struggle,” but one that is nonetheless critical: “We address issues that bring the public to us, and approach ‘ambassadors’ of our work. Resorting to every possible medium is indispensable, as they all add up into a critical mass. Now we are debating the idea of creating an app.”

Oftentimes, this sort of engagement appears to hinge on consistency of output. Labne&Facts thus seeks to leverage very short content—less than 150 words per post—with minimal trappings, as a way of fishing for readers willing to invest in more substantive material. Daraj’s short videos, paired to investigative articles, serve a similar purpose. Mortada explained that Raseef 22 makes strategic use of teasing, tailored headlines and visuals, often requiring real-time adjustments. “The randomness of social media attention is the rule, and we must abide by it,” she added.

In a similar vein, Aziz emphasized that he only came across the New York Times’ long-reads because daily news drove him to the website, again restating the imperative of dynamic coverage even when the real value is elsewhere.

Ultimately, however, virtually all platforms converged around the notion that a hard inner core of supporters played an essential role in building a base. Ibrahim expressed gratefulness at the responsiveness of other journalists in the region to Daraj’s work; Tobia highlighted that Mashallah’s base had grown in part thanks to its approximately one hundred correspondents who, in turn, brought their own circle of connections on board.

Relatedly, multiple participants stressed the value of augmenting digital content with other forms of engagement. Daoud mentioned that, for Labne&Facts “the nature of our own, personal networks has made for unusually high engagement rates for social media. Interaction in real life, through offline events, the Lebanese Salon du Livre and so on, truly pays off over time.” Echoing this approach, Lea Yammine outlined how Lebanon Support’s other activities—such as events surrounding publication launches—helped “create a community around our content.”

A final, optimistic set of observations related to the notion that high quality, specialized output retains its ability to leave a mark on its readership—despite all the challenges inherent to an overcrowded digital sphere, and regardless of various “metrics” that increasingly shape the rules of the game. Nallu, of Refugees Deeply, thus remarked that “sometimes, you need just that one key person to read your piece and be changed by it.” Aziz echoed the point: “Today’s discussion often tends to veer toward technique—how to hack this or that technology to get through to more people. To me, what counts most is the formative experience one gets from a long read, and which will stick for years to come.”

Synaps’ own learning curve

In September 2016, Synaps launched its very first publication. Our published content has since evolved, encompassing a growing set of topics and team members across multiple disciplines. Our overarching ambition was to fill an obvious gap, by combining three things: a sensitive perspective on a socially relevant issue, informed by a lived-experience of it; sufficient depth of analysis to help produce meaning where confusion too often holds sway; and an agreeable read, through intensive editing and an appealing layout.

Publishing is not, by far, Synaps’ core activity. It has been, nonetheless, an extraordinarily enriching experience; here are a few of the lessons we have gleaned in the process.

  • We now know it takes up to ten months to produce a high-quality piece, assuming a rookie author starts off from scratch with no prior experience of research, writing and the topic itself. That is both enormous and surprisingly short, given just how much must be acquired in the process.
  • Publishing is an extremely effective component of our mentoring cycle: it structures the workflow, sets standards, imposes frequent discussions and teamwork, and provides the author with real-albeit delayed-rewards. Authors don’t just learn: they build their own brand and take ownership of various forms of post-publication engagement.
  • Our interactions with our audience have convinced us that there is a huge appetite for such work. Our biggest challenge in addressing it comes from the fact that the intermediary between a publication and its public is social media. Here we are also learning critical lessons. First, we have found that the fate of a new release is determined within 48 hours: either it catches on and spreads or quickly fades out, depending on the early response of social media users. Second, “shares” are the decisive factor in that respect, as “likes” leave you within the boundaries of your preexisting audience. Third, different social media platforms provide very different connections to the audience. Facebook offers the most meaningful, stable engagement with readers. Superficial as it is, Twitter’s forte is to branch out more randomly, reaching people outside of a given circle. LinkedIn seems to be growing steadily as an interface where professionals access content at a somewhat more leisurely pace.
  • Having tried out various communication strategies, we are taking a principled position against outreach strategies that involve paying to promote our own content—which we already publish for free. This low-noise marketing will mean, inevitably, slow growth. We will continue to explore ways of building a rich relationship with our public, by discussing not just our work, but our internal processes where relevant.
  • Finally, we feel that all of us are caught up in a complex, confused, transitional phase in our relationship to information and insight. Through our publishing experience, we are privileged to look at this from both ends: as keen producers and avid consumers too. The market of “meaning” is on the move, and we’re thrilled to contribute, even on a tiny scale, to a general shaking up.

5 December 2017

Dylan Collins is a former managing editor of Syria Deeply, currently polishing off his Arabic skills. Peter Harling is the founder and director of Synaps.

Illustration credits: John William Waterhouse Echo and Narcissus by Wikipedia / licensed by CC; Kiipsaare Leaning Lighthouse by Wikipedia / licensed by CC; fractal tree by Wikipedia / licensed by CC.