M T Fahey

Archive for the ‘Internet and Technology’ Category

Some Thoughts on Internet Publication and Filtering

In General, Internet and Technology, Journalism on June 17, 2011 at 1:59 am

Post-Gutenberg Economics

In his two books and many lectures, popular internet commentator Clay Shirky of NYU’s Interactive Telecommunications Program has described the end of the predominance of Gutenberg Economics in publishing. The traditional publishing house setup came to be soon after the creation of the printing press, a response to the financial risk of upfront publishing costs that put the burden of quality-control on the publisher. This economic pattern has applied to most media industries (music, television, books, radio, newspapers) for the last 500 years:

“If a printer produced copies of a new book and no one wanted to read it, he’d lose the resources that went into creating it. If he did that enough times, he’d be out of business. Printers reproducing Bibles or the works of Aristotle never had to worry that people might not want their wares, but anyone who wanted to produce a novel book faced this risk. How did printers manage that risk?

Their answer was to make the people who bore the risk – the printers – responsible for the quality of the books as well. There’s no obvious reason why people who are good at running a printing press should also be good at deciding which books are worth printing. But a printing press is expensive, requiring a professional staff to keep it running, and because the material has to be produced in advance of demand for it, the economics of the printing press put the risk at the site of production. Indeed, shouldering the possibility that a book might be unpopular marks the transition from printers (who made copies of hallowed works) to publishers (who took on the risk of novelty).”

The main point is that the economic connection between physically producing and distributing media and selecting which media are worth publishing is a historical contingency. It no longer holds in the modern world of low-cost internet self-publishing. We are now, says Shirky, in the world of Post-Gutenberg Economics. There are no longer significant financial barriers to becoming a publisher. I do not need a printing press, a network, or a radio tower with FCC-allotted frequencies to make my content available to the world.

The financial risk has been reduced to a negligible level, more or less just the time it takes to create the content itself. Volume has increased and quality assessments have become unnecessary. In the recent past, Shirky argues in Cognitive Surplus: Creativity and Generosity in a Connected Age, we dedicated the collective free time of this most affluent nation to one-directional consumption (mostly television programs). Now, in the low-cost world of Post-Gutenberg Economics, any ex-consumer can choose to be a producer of content. Clay Shirky looked upon our Wikipedia entries and lolcats and saw that they were good.

Lehrer’s Objection: Consuming properly and creating poorly

In Cognitive Surplus, Shirky points to lolcats as a good candidate for the “stupidest possible creative act.” Still, he maintains that “The real gap is between doing nothing and doing something, and someone making lolcats has bridged that gap.” In a brief review, science writer Jonah Lehrer objected to the suggestion that any act of creation is superior to an act of consumption:

“There are two things to say about this. The first is that the consumption of culture is not always worthless. Is it really better to produce yet another lolcat than watch The Wire? And what about the consumption of literature? By Shirky’s standard, reading a complex novel is no different than imbibing High School Musical, and both are less worthwhile than creating something stupid online. While Shirky repeatedly downplays the importance of quality in creative production–he argues that mediocrity is a necessary side effect of increases in supply–I’d rather consume greatness than create yet another unfunny caption for a cat picture.”

This is a legitimate objection, although Shirky may well respond that what he was celebrating in the book was the rebalancing of the overall production/consumption equation, not necessarily an increase in production for its own sake. Although it is not explicitly stated in the book, I suspect that Shirky’s preferred producer, even of lolcatz, is also still an avid consumer of similar products, even if only of lolcatz. If asked directly, I doubt the Shirky would prefer a world in which everyone writes and no one reads, and I would be surprised if Shirky and Lehrer wouldn’t both rally around the value of true mental engagment, in terms of people becoming part of a network of inputs and outputs.

Looking at only the creative side of the equation, is today’s glut better than the old narrow stream of media production? Shirky acknowledges that the trade-off between quality and quantity is undoubtedly part the bargain: “Increasing freedom to publish does diminish average quality – how could it not?”

To Shirky, the increase in experimentation and the new diversity of contributors (previously locked out by economic of scale in the topics and people that would be included) makes the decrease in average quality worthwhile: “In comparison with a previous age’s scarcity, abundance brings a rapid fall in average quality, but over time the experimentation pays off, diversity expands the range of the possible, and the best work becomes better than what went before.” Even if it is buried beneath a vast excess of junk.

Identifying the problem: Filtration

This is the part of the story I find most compelling. As Shirky said in his 2008 talk at the Web 2.0 Expo in New York (but didn’t really mention in his 2010 book), the problem we are facing today isn’t properly framed as one of information overload (as it so often is), but one of filter failure. If we are to have exponentially more content, some of it better than before in some ways but much of it of stunningly low quality or simply irrelevant, how will we filter our content to find what we want?

We want to spend out limited free time consuming worthwhile media, not wading through worthless creations in search of the gems. This need represents a new, production-decoupled, front end (and therefore potentially lucrative) niche in the new ecosystem. It is as a symptom of Post-Gutenberg economics that the new quality-assurance role was separated from content production and publishing, which were bundled together by traditional media organizations. This functional fragmentation has some interesting consequences.

Take, as an example of a successful new news organization, the Huffington Post. In the HuffPost business model, traditional news organizations and independent writers produce content and publish it to the internet, and the site functions as the quality-assurance agent, aggregating content that has been chosen to satisfy HuffPost’s specific reader demographic and presenting it with links and well within fair use copyright law. On the internet it is content access which lends itself to advertising, and therefore it is the point of access that makes money, especially if the expensive task of content production has been avoided (allowing Arriana Huffington to sell the company for $315 million).

The success of news aggregation services seems to be an indication that these organizations have fallen into the ideal slot left in the new economy. In a Pew study of 199 leading news sites, 47 were classified as primarily being based on aggregation or commentary (although only three of the top ten were in this category).

Another interesting consequence of this new economic system is in news journalism, where content filtration and quality-assurance can be far from synonymous, and where the disruption in revenue flow has been especially disastrous. The reputation of a source like the New York Times is based around vigorous fact-checking and strict journalistic standards. The HuffPost can present NYT info to its readers without any regard to accuracy. In this case, the news producer retains the responsibility for the expensive quality of accuracy while the aggregator only has responsibility for presenting interesting and recent stories. This diffusion of responsibility can lead to dangerous situations in modern journalism (see my post on churnalism), such as the recent Texas mass grave hoax and the Gay Girl in Damascus hoax. Although Shirky’s observation that “the filter for quality is now way downstream of the side of production” seems to generally be true, it is interesting that it depends on what “quality” is being discussed.

Different Filters for Different Folks

If I were to make a sweeping prediction, it would be that filtration will necessarily be the main role of any successful new media organization. Excluding my internet email service, the sites I visit most frequently all serve this role: Google.com, Reddit.com, thebrowser.com, instapaper.com.

Google’s search engine is a clear example of a super-successful algorithmic filter. The Huffinton Post and Newser use human curation. A different sort of human curation (one closer to the sort of free and collective work that Shirky glorifies) can be found in websites like Reddit and Digg, which use members’ votes to move stories to the top of a long list. Reddit is a favorite of mine, as my selection of smaller communities (subreddits) gives me a uniquely personalized list of content that swings between user-created lolcat-like content and breaking international news.

Editorial curation of long pieces of written content takes place at sites such as thebroswer, instapaper (which also provides a service for saving content from other sites to read later), longreads.com, and longform.org. These sites simply sort through older publications (eg the New Yorker, the Guardian, or the Atlantic), high-quality blogs, and occasionally user submissions and judge pieces based on their individual value. These sites are pleasantly egalitarian (A blogger and Noam Chomsky can be equals if they produce equal work), while still driving traffic to the old high-quality publishers.

Not only can anyone produce content, anyone can be a curator of content. Anyone who consumes a fair amount of media content can pick the best and make a list for others, such that they act as a content filter to save others like them the trouble of sorting through the ever-increasing flood of info. I do just this on my own blog’s Reading List. Twitter also works as a sort of community-driven filter, as people retweet information they propagate that content throughout the network and make it more likely to be seen by others with similar interests.

Eli Pariser recently released a book The Filter Bubble: what the internet is hiding from you, about the level to which many of the most popular internet filters are highly personalized. It is fair enough to argue that a good filter should be personalized. Although I have yet to read the book, I gather from an interview of Pariser (NPR) that his main objection is that few of us realize how the information being offered to us by these services is being filtered. If we’re filtering for relevance and quality, what relevance and what qualities. Who goes Google think you are, and what information has the algorithm decided you don’t want to see?

Google, Facebook, and Twitter already change their results to cater to the individual user. Given the same search parameters, each of us is returned a different selection when we search or look at our various information feeds. News organizations (e.g. the Washington Post) are also considering customizing their homepages to the interests of the individual. It doesn’t take too much imagination to see how this could have an isolating effect for individuals and a polarizing effect on groups, an “echo chamber” situation. I imagine that the future of information filtration will have to include a system to introduce novelty into content that is presented (Google’s algorithm already does something similar, giving an outlier link a few results down the list). I doubt that the lack of transparency in the details of content filtration will last very long.

Addendum: How I consume media, the future looks bleak.

When I picked up Shirky’s book several days ago, I did so at our local library. I haven’t bought a CD, musical single, book (with the exception of used textbooks), magazine, newspaper, movie, or access to a television show in the last four years.

And yet, I have seen the last several seasons of a couple television shows, I have read the New York Times regularly for years, and enjoy access to the Atlantic, New York magazine, and dozens of blogs and news sites. There may be advertisements, but as far as I know I’ve never clicked on one, let alone purchased a good or service. I have almost one Terabyte of music and movies on my computer, and I’m working my way through the several hundred books that reside on my Amazon Kindle (the only tangible object I bought on the internet in the last year).

Yes, some fraction of the onus for the decline of the media, part of the breakdown in the economics that served these industries for so many years is on my opportunistic, pirating shoulders. As an unemployed 20-something, I use media products without any qualms, with the rationalization that were I somehow put in a situation where payment was necessary, I would out of necessity simply stop reading, listening, and watching.

I have no idea how any of these media are going to survive if there are many people with the same outlook as me, and I suspect there are many. It is not surprising that banner advertising on websites is not an adequate source of income (84% of US internet users don’t click on a single ad in a month, and of clicked ads Gawker’s Nick Denton wrote: “clickthroughs are an indicator of the blindness, senility or idiocy of readers rather than the effectiveness of the ads.”). How could ads be adequate? I don’t know anyone who makes purchases on the internet in the blithe way that those ads seem to presuppose.

Excluding music consumption, I think I spend close to thirty hours a week reading books and internet articles and watching television shows on the internet. If I had any sort of disposable income, I would be willing to pay one lump sum for each of the services I use (that is to say, I would pay one news organization and one internet television service like Netflix). This sum would be nothing close to the amount the NYT wants users to pay to get behind their easily-circumvented paywall ($455/year). It would be like the $30/year I voluntarily give to NPR, despite the fact that I clearly consume more than $30 of NPR programming.

To be clear, it’s not that I don’t think that the NYT deserves $445 for the services they provide, it’s that there’s no way a debt-laden unemployed ex-student such as myself could ever pay for such a thing, and I see no reason to willingly make myself any more ignorant or culturally stunted than I need to be by choosing not to use easily-accessible information and services just because I can’t afford them. I have heard many things about my generation, including that we act more entitled than previous generations, and in this case I wonder if that is true.

If one of the aggregating services I use daily took a donation and spread it in a proportional manner amongst their news contributors, I would give the small amount I could. If I could go into a music store and pay 99 cents for an entire album so I could go home and give it a listen, I would do so. If I knew of an organization that ranked news producers by the quality of their output and then distributed donated money to those groups (and maybe the individual authors and reports), I could get behind that financially. It’s about opportunity and convenience, not poorly-enforced commercial contracts. Could we raise money for internet media like the Grobanites Shirky describes raising charity money in his book? Could it work at a grassroots level (using new internet organizing resources to solve internet media problems)?

I suppose my question is what will be the organization that will provide the following service: one that acts as a single trustworthy filter that gives me the content I want, the news I think is important, and gives me the opportunity to pay for the content from all those disparate sources in one lump sum, to my ability, such that that money makes it back to the original producers so that they may continue to exist and produce. Is such a thing possible?