March 7, 2010
News Without the Narrative Needed to Make Sense of the News: What I Will Say at South by SouthwestThese are my notes. You can help advance the discussion by reading them over and commenting.Suppose your laptop continually received updates to software that was never installed on your laptop. If you can imagine a situation that absurd, then you are ready to partake in the Future of Context panel that I’ll be part of at the South by Southwest festival in Austin next week. Here are some of my ideas, questions and puzzlers in advance of that event. I am posting them today in hopes of generating a discussion I can use to improve my performance in Austin. (It’s already happening, see the comments.) 1. Why are we serving people the news without the background narrative necessary to make sense of the news? I first became interested in this problem after listening to The Giant Pool of Money, the awesomely effective one-hour This American Life episode that finally explained to me what the mortgage banking crisis was, how it happened and why it implicated… well, just about everyone. I was grateful, because up to that moment I had absorbed many hundreds of reports about “subprime lenders in trouble” but had not understood a single one of them. It wasn’t that these reports were uninformative. Rather, I was not informable because I lacked the necessary background knowledge to grasp what was being sent to me as news. On the other hand there was no easy way for me to get that background and make myself informable because the way our news system works, it’s like the updates to the program arrive whether you have the program installed or not! Which is rather messed up. But what do we do about it? The first thing I did is write my 2008 post, National Explainer: A Job for Journalists on the Demand Side of News. So if you want to help me out, start there. 2. Another way of putting the problem, though I admit this is kind of abstract: why are Wikipedia (which specializes in background knowledge) and nytimes.com (which specializes in newsy updates) separate services? Why aren’t they the same service, so that the movie still makes sense, even if you come in during the middle of it, as most of us do? The news industry’s current answer to that question is topic pages, like this one on global warming, which gets linked to in a news story such as this one, “Lawmakers From Coal States Seek to Delay Emission Limits.” Not terrible. (Also see Google’s Living Stories experiment.) But is that the best we can do? 3. The Giant Pool of Money is far and away the most downloaded program in the history of This American Life. It won a slew of awards. Clearly, there was lurking demand for explanation that was going unnoticed and unmet prior to the program airing. This is one of the reasons I created explainthis.org, which is up in beta form. The idea of the site is partly explained in the subtitle: “What’s your question? Journalists are standing by.” (But also see this post, and this report on the idea.) The goal is to surface the hidden demand for explanation and create a kind of user-driven assignment desk for the explainer genre, which is itself under-developed in pro journalism. Are there other ways to surface this kind of demand? 4. My next move is to match up explainthis.org with a news site that has journalists who are skilled at explanation “standing by.” For a little preview of how that might work, see the special page we made for The Nation magazine’s explainer podcast, The Breakdown, with Washington correspondent Chris Hayes. I can’t say its clicked so far, but we will keep trying. What ideas do you have? 5. Why is explanation under-emphasized in the modern newsroom? A number of factors, I think.
What factors should be added to these? 6. Alex Bloomberg, one of the reporters on The Giant Pool of Money, was interviewed about the program’s success and mentioned something that I consider pure gold: People were saying things like, “I didn’t really understand this. It was in the news all the time but I didn’t know what they were talking about until I heard that episode.” It was very gratifying because that’s exactly what my intent was. Because that was me; I didn’t understand it either. This suggests that if journalists could put themselves in the shoes of ordinary users more effectively they would realize all the places where It was in the news all the time but I didn’t know what they were talking about applies. When you make the journey from “I didn’t know what they were talking about…” on over to “…now I see why this is news,” you’re more qualified to assist others in the system as they travel from cluelessness to informable. (Note that journalism and journey share a common root.) And so part of the puzzle here is: how do we put the news producers into the shoes of users who are getting the updates to programs that were never installed on their news and current affairs hard drives in the first place? I will keep adding to this post, so please comment and leave links. Thanks! After Matter: Notes, reactions & links… The SXSW discussion has its own site: Futureofcontext.com, with ways to add posts of your own, comment on key concepts and get up to speed. UPDATE, March 17, 2010. The event at South by Southwest went well. We had a full room and good buzz. Steve Myers of Poynter live blogged it, showing considerable skill in that form. Elise Hu wrote a very good summary at her blog. Also see my follow-up post, How the Backchannel Has Changed the Game for Conference Panelists. You can listen to the event—panel plus Q & A—here. Comments from Twitter helped me realize a key point I’m making: In order for the news to be informative people have to be informable, and simply delivering a steam of newsy updates won’t get us there. I mentioned the Giant Pool of Money, but as two readers pointed out in the comments http://www.crisisofcredit.com/ is also very effective in explaining the mortgage crisis. It’s an 11-minute video by Jonathan Jarvis. This Giant Pool of Money spawned its own unit at NPR: Planet Money. In order to explain what a “toxic asset” is, two of their reporters decided to buy one, for real. Man, it makes for great radio and it’s a brilliant example of providing the requisite background knowledge to make sense of toxic asset news. The South By Southwest panel was put together by Matt Thompson, now of NPR. See his post, parallel to mine: The case for context: my opening statement for SXSW. Journalists spend a ton of time trying to acquire the systemic knowledge we need to report an issue, yet we dribble it out in stingy bits between lots and lots of worthless, episodic updates… Matt has been writing effectively about this subject at Newsless.org. Especially valuable are: The three key parts of news stories you usually don’t get and his essay in Nieman Reports, Antidote for Web Overload. Joining us on the panel will be Tristan Harris, the 25 year-old CEO and co-founder of Apture, which makes it simple for users to intuitively get more information without leaving the page. Tristan’s preview post is up: Context: The Future of the Web. “What news needs is object-oriented journalism in which context is a basic building block upon which to create articles.” (This of course is a reference to object-oriented programming.) Moderating and leading will be Staci Kramer, who is co-editor and Executive VP of ContentNext Media, publisher of paidcontent.org In The market for explainables Doc Searls builds on this post and points out that when we don’t have stories that explain an issue to us, we fall back on default narratives like, “who’s winning the politics of health care.” So the easy thing is to go back to covering the compromise bill’s chances in Congress, and the politics surrounding it. That at least makes some kind of sense. We have all our story elements in place. It’s all politics from here on. Bring in the sports and war metaphors and let automated processes carry the rest. Don’t dig, just dine. The sausage-machine rocks on. Loyal PressThink reader Andrew Tyndall of the Tydnall Report writes about the CBS Evening News trying to develop the explainer genre with mixed success. Howard Weaver, former VP for News at McClatchy newspapers, in the comments: “Incrementalism in reporting — one or two new facts atop four paragraphs of old B-matter — don’t keep readers well informed. In fact, they may hurt; after while, we give up on chasing every incremental update. When we do tune in, what we get is a brief glimpse of parts of stories.” More Weaver at the futureofcontext site. While narrative prose will always play a central role in human communication, the future of public service journalism does not reside with “the story.” Serving news audiences today demands the ability to deliver information that is, as Matt Thompson says, “both timelier and more timeless.” Chuck Peters, CEO of The Gazette Company in Cedar Rapids, Iowa, a forward-leaning newspaper and broadcasting company, says in the comments that “context is critical.” However, I can’t see providing that context without changing how we create information in the first instance. Any factual element (photo, incident, quote, data, etc.) can be relevant to numerous contextual narratives. So each of those elements needs to both “stand on its own” and be tagged with as many potential relationships as possible…. We usually create information today in locked-down packaged articles, which block the easy flow of the elements between and among narratives. Read the rest. Getting disciplined and strategic about tagging may be one way professional journalism separates itself from the flood of cheap content online. Graphing narrow and broad against shallow and deep to create a matrix, Josh Young complicates our picture in this 2009 think piece that is well worth re-visiting. In the comments, J-professor Donica Mensing brings up a critical issue: “how to organize explanations about issues that are highly contested.” As Mensing notes, “Explainers aren’t neutral. Actors and motives have to be identified and shaped; arguments over those can be endless.” Click here to return to the top of After Matter. To see what I am up to on Twitter go here. Posted by Jay Rosen at March 7, 2010 5:00 PM Print Comments
Can the "Living Story" project of Google (which they did with The New York Times and The Washington Post I think) be a solution? Posted by: Roland Legrand at March 7, 2010 6:21 PM | Permalink I didn't add that in yet, but definitely something we will discuss in Austin, Roland. The "explainer" as a genre it a great topic. I wish there had been better explainers about the military, Afghanistan, and Iraq. It has been especially disappointing, and frustrating, that over the decade, the "News Without the Narrative" about the military has continued without improvement. Perhaps, the news about the military already has entrenched Master Narratives that permit no explanation? The problem with Google's Living Stories is that it still sucks at explanation. The background provided is just a list of articles previous written about the topic. That represents far too much effort to put into understanding a topic. Seems to me that adding a wikipedia like entry in place of previous articles to living stories gets you much closer to the format you’d want. Add in the playback feature from Google Wave so you can see how the topic has evolved over time, and I think you’re well on the way to a new, not-the-article, format. Posted by: Joey Baker at March 7, 2010 8:44 PM | Permalink Google's collaboration with a couple of newspapers on "Living Stories" seemed so promising when announced. It hasn't amounted to much, though. Tell me, what's the difference between Living Stories and a well-done "topics page?" It's still a collection of old stories on a single page arranged by subject. That's not what we need. What we need is stories that really are alive -- that evolve and change and expand organically as the news moves. I want to be able to check a living story that will introduce me to the news the first time I visit and then easily alert me to changes every time I stop by. Couldn't' this start with a simple internal-wiki treatment? With a little tweaking, it could, perhaps, display the most recent changes in red. It could link to a chronological list of all the changes since it was created. It could make it simple for readers to say "I don't get this" or "Have you checked on this?" and thus help shape the evolution. Incrementalism in reporting -- one or two new facts atop four paragraphs of old B-matter -- don't keep readers well informed. In fact, they may hurt; after while, we give up on chasing every incremental update. When we do tune in, what we get is a brief glimpse of parts of stories. This seems so plain to me that I predict it will be implemented soon. It will be very popular. Posted by: Howard Weaver at March 7, 2010 10:15 PM | Permalink I recommend an exploration of the semantic web - a subject I'm still wrapping my head around - and its implications for journalism. I googled Semantic Web and Journalism and didn't find a ton. But I think this might be due to its relative novelty and people's difficulty understanding what in God's name it is. Ironically if it existed in full form, understanding it supposedly wouldn't be a problem! :-) Here are a couple of resources I dug up. Not sure how helpful they'll be, but I think the emergence of the semantic web is HUGE. From what I've read so far, it's going to change just about everything: Semantic search – an Interview with Brooke Aker Semantic technology gains publishing foothold Oh, just remembered this great piece from Read Write Web: "Sir Tim Berners-Lee, the inventor of the Web, gave a must-view talk at the TED Conference earlier this year, evangelizing Linked Data. He said that Linked Data was a sea change akin to the invention of the WWW itself."
Posted by: Emily Cunningham at March 7, 2010 10:20 PM | Permalink Jay - I am glad you, Matt and the rest of your group are going to focus on this. I agree that context is king. To get there we need better raw materials and better tools. As we look to develop our communities, context is critical. However, I can't see providing that context without changing how we create information in the first instance. Any factual element (photo, incident, quote, data, etc.) can be relevant to numerous contextual narratives. So each of those elements needs to both "stand on its own" and be tagged with as many potential relationships as possible. As others work with that element, they need to be able to add relevant relationships that may have been missed originally. For example, one quote from the mayor might be relevant to his personal narrative, his political narrative, the city narrative and the particular issue narrative. Any news organization might put that element into only one of those narratives. In contrast to these flexible elements, we usually create information today in locked-down packaged articles, which block the easy flow of the elements between and among narratives. As a companion to elements of information, we need easy-to-use local wikis, so that many people can create the narratives to create context for the elements. We are working diligently on creating information in the first instance as tagged elements, so that we can power whatever applications will be developed for context. Our hope is that such a system will make individuals stronger by enabling them to get relevant information, in context, whenever, wherever and on whatever device. Stronger individuals will enable stronger communities. Thanks for keeping this conversation going. Chuck Posted by: Chuck Peters at March 8, 2010 8:24 AM | Permalink What we need is stories that really are alive -- that evolve and change and expand organically as the news moves. I want to be able to check a living story that will introduce me to the news the first time I visit and then easily alert me to changes every time I stop by.-- Howard Weaver The Living Stories project is a good step in the right direction. I think that the sensibility of Rich Gordon's NewsMixer is another component. Having people comment on parts of a story begins the growth of a "vine" around a topic. Some way to trigger that the comments have been vetted and then injecting those comments into a living narrative would be the next step. I encourage you to keep an eye on the Public Insight Journalism effort at American Public Media for a project that I am working on. It will use conversation and research in a collaborative way with topics of interest that I hope gets closer to the ideal we seek. Matt Thompson has been key in pushing us away from the spasmodic lurching of old news models. Posted by: Michael Caputo at March 8, 2010 9:44 AM | Permalink There's just one problem with your Question 5: A century of data on news content shows the elements of explanation — the how and why of events— and the emphasis on interpretation increasing, especially in newspapers and even for political reports in their online editions in this century. The increases in broadcasting tend to be more opinion than background and context. But journalists across all U.S. media have been doing more explaining for decades. You can see the studies at http://www.uic.edu/~kgbcomm/longnews Posted by: Kevin Barnhurst at March 8, 2010 12:10 PM | Permalink Reading this post and the comments thus far, make me think of some really great explainers. I also personally feel so strongly that context is so important. From the world of large public meetings on controversial and important issues comes these non partisan (or trans partisan, if you prefer) issue guides: I also think Lee and Sachi LeFever at CommonCraft warrant a mention here. Their "in plain english" videos are only 3 minutes long, are extremely well received, and at the very least act as a jumping off point for people to research further: Thanks Jay. Posted by: Arjun Singh at March 8, 2010 12:35 PM | Permalink This is a great subject Jay and I suggest that we could take it one step further to say that some stories could be explained in advance. Take your example of the Giant Pool of Money. There was considerable evidence that this was a train wreck in the making before it ever became a train wreck. In fact there were many people trying to get the story out there in an effort to warn everyone. We have alot of people watching reality TV -- which in many cases like Survivor -- is manufactured reality. Why can't we take the interest in real things and focus it on real things? All good stories revolve around a good conflict and we have plenty of conflicts to track. I agree that not all real conflicts are predicatable, and many expected conflicts fizzle, so we would have to find a way through that. Wars and polital races are logical items for this. Highly competitive marketplaces with big personalities (Like Microsoft vs Apple vs Google) are other easy ones. I would envision the "background" being a timeline that would be like a Wikipedia page for the conflict, and each new episode would add a dot to the timeline. This is a fantastic topic. I too felt rescued by TAL's The Giant Pool of Money; so much so that I sent the link to everyone I knew. My comment/suggestion is a bit wiggly -- is there a way for explainthis.org to make use of user-generated explanation? I suppose this would take a great deal (and time) of moderation, but it seems like the old-school ideal of the web -- strangers teaching strangers -- a community of users... Posted by: C Kehoe at March 8, 2010 5:42 PM | Permalink I think C Kehoe's wiggly comment is very good! Citizen education. Posted by: Arjun Singh at March 8, 2010 10:15 PM | Permalink Who is going to pay for this context, in journalism? (I say "in journalism" because when I want context I read a book on the subject--a book often written by a journalist.) Isn't that the issue? Supporting the creation of long-form journalism? Also: context is dependent upon a thorough understanding and digestion of material. Digestion takes time. What we commonly think of as "news" travels a shorter arc. Further also: as stories unfold, limning them is difficult. For many reasons, which we all can enumerate. Both my daughters are finishing their Ph.D.s at Princeton. And when I look at their fields, Russian history and art history, what I see is that even in the empyrean where they are working there are very few people with the gift of providing elucidation. Not that many journalists have the ability either. Another hurdle for context. What I hope will be a possible solution is the iPad--or rather a new way of telling a story that the iPad will introduce. To take an example from yesterday's NYT: a brief account of religion-motivated slaughter in Nigeria. For its iPad version, the NYT could post the brief account. Any reader who wanted context could run a finger over "Nigeria" and summon up a piece on African colonialism, which the NYT might provide from its own archives or from a recent book via some kind of profit-sharing arrangement. Running a finger over "Christian" and "Muslim" would link to similar historical articles. And maps. Video clips. A window into an in-house discussion (or pad-cast) on why the NYT chose not to say which religious group had been victims, which slayers. With further links to articles from the NYT's nineteenth-century archives. And enriching the context in breadth rather than depth, links to current stories about religious conflict between Christians and Muslims in Iraq, Egypt and Malaysia. And comity in Palestine. This new form will require the link-link and the cooperative, multiple-sourced journalism that you advocate, Jay. And then the question becomes, it seems to me, whether inquirers will pay for an Africa book by a top journalist and Africa expert like George Packer when it will be so much more engaging and intriguing to assemble the context for ourselves. Posted by: Mayhill Fowler at March 9, 2010 4:26 AM | Permalink The Apture tool looks promising. I still think there's a need for another layer: syndicated, embeddable backstories that have been vetted http://friendfeed.com/clique-with-claque/84fe266f/it-seems-like-there-would-be-market-for One question I have is how to organize explanations about issues that are highly contested. Explainers aren't neutral. Actors and motives have to be identified and shaped; arguments over those can be endless. Wikipedia's technology deals well with this on less controversial topics, but for hot topics even it can break down. How might a journalist work with a public to build an explanation that is mostly accepted by the affected community? One thought is to start with a simple list of facts: what facts can we all agree on? Have the list be open for debate; use a post like this to come to some judgment about the most agreed upon facts. From that the community can assign meanings to facts, hypothesize tradeoffs, identify consequences, etc. This could help the journalist build a shared understanding of the context of a story. Using all of that collective intelligence, a concise and credible explainer could then be created to help those less connected get the context they need to follow the story further. I'm sure there are many ways to do this. I look forward to hearing what the panel comes up with. Posted by: Donica Mensing at March 9, 2010 4:14 PM | Permalink A very important set of questions, Donica. Thanks. I think a crucial question is: how much explanation and background are enough? Obviously there is too little today, but when is too much given? Posted by: Michael Walters at March 9, 2010 10:12 PM | Permalink How does a journalist determine the following for an explainer: 1. Appropriate length/depth (3 min. Common Craft video, 1 hr Giant Pool of Money, 1250 pp. On Point & On Point II professional history texts) There was one post after "On Point II" that caught my attention: If you read one Sunday op-ed... how long did producers work on "giant pool of money?" Probably several weeks - minimum. Today fewer and fewer journalists are afforded the time to make sense of the news they are covering and the time it takes to produce high quality journalism like "giant pool of money." give talented journalists time and good things can happen. give them daily deadlines - an echo chamber of wire copy is the result. Posted by: jason Samuels at March 9, 2010 10:35 PM | Permalink Here's another bullet point: Most news companies don't have the web resources - staff, organization, even basic website design - to make it worthwhile. If you're going to do an explainer, make sure it lives forever somewhere so people can get it if they missed that first day's paper. But companies still lack the internal organization to effectively think both online and print on some major stories. Plus, they lack the staff to work on the site to make it presentable. Posted by: Lisa Fleisher at March 9, 2010 11:02 PM | Permalink It seems tautological, but I'm pretty sure it is important and true: Providing context requires understanding Context as ingredient and container… Good one, Lisa. I'm adding that in to the bulleted list. I have had an idea for a project for a while, which i've been unable to execute due to time and resources which is complementary to your project. The flip side to providing context for the content presented to media consumers is knowing who is delivering the content, and particularly, whether an interviewee has an agenda or ulterior motive in providing the information they are offering. There are many examples of sources both of news and policy advice used in the media and politics which are, at best, naively misrepresented by journalists and politicians to bolster the points they wish to make. The nice thing about claims of expertise and authority in the news media however, is that most of the claims about a person (who they purportedly are, and what they represent) leaves a public trail which can be tracked. What i would like to propose is a system that scrapes information from the web (such as Rachel Maddow's guest lists, which are posted weekly, or NPR's archives via their API) to put together a record of appearances for interview guests, AND their particular affinity to specific venues. The first goal being the ability to represent a map/node network to see how guests and shows cluster. Is it the case that David Brooks and Juan Williams are bipartisan in their choice of venue? (i'd hazard a guess that the answer is yes) Is it the case that Liz Cheney only ever shows up in conservative (and presumably friendly) venues? Is that the case for DNC Chair Tim Kaine? The data exists to be able to build such a map, and provide links to previous appearances or other material these people have produced. From there one could presumably be able to build data driven infographics that one could use as a short hand for what kind of interviewee they are. Everyone talks about how we've entered the Facebook era, where history is forever, and your actions will stick with you for the rest of your life, and yet, none of that seems to apply to the news media. Where's the accountability? Where's our news media equivalent of reddit karma? (Okay maybe reddit karma is a bad example, since at a casual glance most people don't consider which redditor posted a link when clicking on it) It was a travesty that Betsy McCoy was allowed to completely mischaracterize the 1994 health care proposals, but once discredited, it is unequivocally irresponsible for the news media to give her yet another 15 minutes of fame in an attempt to destroy the 2009 health care proposals in exactly the same way. If we can't trust the news media to either vet, or fairly represent their guests, then why don't we do it for them? In the end, why should the Daily Show be the only ones capable of data mining news history for hilarious incongruities and hypocrisy? A parting note: One of the epically frustrating things things about the news media, especially for online print media, is the dearth of links provided. If a claim is made, IMO, it should be backed up with citations, and not doing so is in essence a dereliction of one's journalistic duty, and akin to creative and unfair editing of TV clips. This is where wikipedia's citation standards are in essence the correct position to standardize around, but FSM only knows how to compel the media to change their norms, when the only benefit is accountability to their consumers at the cost of time/money. Posted by: knowtheory at March 10, 2010 12:47 AM | Permalink Fantastic piece, Jay. Congrats. Jay, I'd like to take you back to the brilliant essay by Christopher Lasch, The Lost Art of Political Argument published by Harper's in 1990. Lasch deconstructs the professional press in such a way that we see that the decline in participation in the political process in America is tied to the rise of professionalism in the press. So I don't think there's a real demand for the details (and, conversely, a demand for the bizarre, breaking and surface), because the American public generally doesn't believe they need those details. We may believe they do, but by their behavior, they prove otherwise. But rather than blame the viewers or the readers, we should be looking in the mirror, for as Lasch argues, we have brought this on ourselves by delivering "the news" in "just the facts" packages. Argument, he concludes, is what's missing - that explanation, whether blaring or otherwise, of WHY. Posted by: Terry Heaton at March 10, 2010 9:48 AM | Permalink I agree with your idea that a background narrative is needed to accompany the news, but what happens when the narrative itself creates controversy. For example, how do you give an explanation involving the Gaza Strip or the theory of evolution, without creating a heated debate over the explanation itself. My feeling is that without a system to resolve the debate, the narrative will lose credibility, and as a result, we are back to where we started (not knowing the underlying facts of the news article). Posted by: Steve Abramson at March 10, 2010 3:25 PM | Permalink One key motivator for context is this: Like ''The Great Pool of Money,'' explainers don't have to be boring. They don't have to go on forever. They can come in baby steps, even be part of the way there -- 12 keys toward understanding the banking crisis; 10 things to do right away if you get a recall notice on your Toyota; 14 reasons why the earth shifted/quake hit Haiti, or Chile. Episodic info with teensy bits of context stuck in, as Matt says, is a starvation diet. I second Emily's excellent links in her comments and the TED speech by Tim Berniers-Lee on Linked Data. Posted by: David Beard at March 10, 2010 11:59 PM | Permalink Jay, I tend to find your ideas profound. You're saying things I don't hear anywhere else, especially not in my law school classes. I first saw you on Economist.org and it looks like your putting your "Pro-Am" concept into practice here. As to your prompt: I think explainers, if they were to exist, could be most effective if they were highly visual. Something like a series of pictures/diagrams so a quick look would help someone understand the main actors in a story. Personally, when I heard about the banking crisis it was just plain difficult for me to understand how money was flowing from where and to whom? The most helpful stuff I found was 5 minute youtube video where the person had diagrammed out all the main actors of the crisis and then explained how each one interacted with the other. For someone not looking to become an expert, and as a highly visual learner, it was the first time things started to click in my head. Thus, traditional journalists might not possess the skills to come up with visually effective and appealing ways to tell a story, let alone explain the background of story. I guess what I'm wondering is what if background stories always started with a few really creative diagrams, and then went into a more traditional text explanation? Posted by: Drew at March 11, 2010 3:24 AM | Permalink Tyndall Report watches CBS Evening News dipping its toe into the explainer waters with its start-the-decade Where America Stands series. Posted by: Andrew Tyndall at March 11, 2010 12:07 PM | Permalink Drew, I remember that video: The Crisis of Credit Visualized I agree, that was great! I tend to think about my blog as a way to episodically capture and explain things to myself. That's probably not a very journalistic approach, but it seems to fit in this context. I also think this quote fits in the "who is the audience" of an explainer seeking context? "One is an expert on whatever subject we are writing about, someone who will read this story no matter what, but who will be highly judgmental. ... The other is your basically curious person, but without a lot of time, who is, in my mind, the real challenge. He or she might read the story. But it has to hook them. The game in my head is: Okay, how do we write this so that it is accurate and has weight, but is still fun to read for someone who really doesn't care much about say, college dorms or tutoring?" Jay, I think your ongoing investigation of the “explainer” is terrific. I have been thinking about three different modes of journalistic and filmic storytelling for another project I am working on and it occurred to me that they may provide a useful way to think about explainers. The three modes are The proclamatory, The way this might work with explainers is something like: Proclamatory: An op-ed from an economic columnist/business editor saying this is the way the economy works because I know I’m an expert The evocative: A lead feature in the Sunday Mag from a literary journalist on Wall Street traders which explains both the way they work and their motivations through a detailed set of examples from their daily lives and work practices. I suspect that Giant Pool of Money fits here although I haven’t had a chance to listen to it yet. The Reflective: may come from a publication’s blogs or public editor, it not only tries to explain the background of the story but the background of the storytelling. How well have we been doing with our coverage of the GFC, what were our missed opportunities, where do we need to go from here to understand this story better? This type of piece explains the omissions and identifies questions which still need explanations and is up front about what we can’t satisfactorily explain yet. It seems to me that the mode of storytelling adopted by journalists in doing an explainer/backgrounder has big implications for audience trust and involvement. Traditionally explanatory journalism fits the old fill-em up model of journalistic knowledge transfer so what modes invite readers into a participatory process of understanding? As you say in your original post about Giant Pool of Money it put you in the market for more news, but in what other ways can explainers invite participation? The evocative mode seems to hold the most potential on a day to day basis but at times the proclamatory mode may be useful – for example as a way of quickly adding explanatory information to some new aspect of the narrative as it emerges. The reflective mode is also critical in keeping the whole explanatory process transparent and highlights the fact that it is ongoing not singular. I also think Chuck Peter’s comments about new tools, new modular forms of journalism are important. In my examples above for example how do we move to a new modular – but just as evocative – form of the long Sunday Mag feature which allows its partial reuse in a range of explanatory contexts? How do we work with interactive modules that link effectively across these and other modes of storytelling to explain the background narratives of news? Best of luck with the panel. Posted by: Marcus O'Donnell at March 12, 2010 7:18 PM | Permalink I don't have a comment to make on the topic itself, except I admire the freshness of your approach. What I also admire is that you should put online at your blog a "talk" you are about to give at a major conference. It would make for a richer conference if all speakers did the same. Would make for a more interactive talk. More would come out. More people would contribute. And then record it all on video and post online and invite the rest of the world into the experience. Posted by: Paramendra Bhagat at March 13, 2010 3:01 AM | Permalink Thanks, Paramendra Bhagat. That's the idea. We also created a site for the panel event itself: http://www.futureofcontext.com/ Marcus: thank you for those thoughts. I think you're right that the "mode" of explanation is an important variable. I would add to your list the "dialogic," the back-and-forth explainer. A common example is the FAQ. Better example is this brilliant Planet Money podcast in which two journalists buy a toxic asset and take turns explaining it, alternating with a source. I'm quite taken with the abstract but pointed question, "Why are wikipedia.org and nytimes.com separate services?" But I don't think it's quite right. We agree that what Howard calls the "incrementalism" of contemporary journalism is narrow in scope. I think Matt would agree that "episodic information" is narrow in scope. The problem, as I wrote the last time I really thought about this context stuff, is that wikipedia focuses on topics narrow in scope as well. In part, in a culture obsesses with NPOV, this is a strategy for confining and limiting controversy--or cutting down what's "highly contested" into a greater number of smaller pieces. My point is that you probably don't get broad, synthetic, "Giant Pool of Money"-type journalism out of incrementalism+wikipedia. You just get deep updates, which may not make citizens very informable on the big issues of the day. Posted by: Josh Young at March 13, 2010 1:10 PM | Permalink "My point is that you probably don't get broad, synthetic, "Giant Pool of Money"-type journalism out of incrementalism+wikipedia. You just get deep updates, which may not make citizens very informable on the big issues of the day." Astute point, Josh. Wikipedia - great at depth, poor at distillation. You might also like my post at FutureOfContext.com: "The trouble with wikis." Jay, We've spent three years building a system around the idea of context. It will launch this year. The core concept: Everything should be nestled in a related material that helps explain. In this scenario, every Web page is a complete capsule, not just a story (or blog post, or photo, or whatever). Every page is a topics page. One problem: the labor to make it happen. As mentioned above, the semantic Web could create tremendous possibilities. That's why the publishing platform to be launched is built for it -- for Web 3.0. You can't make every page a capsule without help. Another problem: How to pull out the essence of explanation buried deep within stories. As we were designing the system, it occurred to us that we needed something Wikipedia-esque on a local scale. We're using reporting and citizen journalism efforts to gather 500 local items as a start. They will be critical journalism units to surround those letters and comments and stories. I wrote some short essays for students as we prepare to launch. They may be of use (or not): Tom Warhover Posted by: Tom Warhover at March 14, 2010 12:48 PM | Permalink Jay, I think you're onto something here, and as a journalist I'd actually much rather explain (and learn) how the international banking system works than be just one more reporter droning on about Iceland's demise. But I wonder how the global warming example fits in. There's no shortage of good explanations available on the science of global warming--by both scientists and journalists--and yet I'd bet a good portion of the American public doesn't understand it at all. That's why people continue to confuse it with the hole in the ozone layer and various other environmental problems that are unrelated--and why, at least in some cases, people continue to deny that it's happening, despite the evidence. (Another reason is political, of course--but leaving that aside, I think it's safe to say that lack of understanding is a factor.) I'm curious what you think about this. Does global warming fit into your "explainer" model? Or is it the exception to the rule? Wish I was at SXSW--sounds like it'll be a great session. Posted by: Hillary Rosner at March 14, 2010 1:43 PM | Permalink The point, high up, that people have to be "informable" is very true. In order to even get people to that point, you have to "sell" them on wanting to be informed. This means a big part of the "news" is in the execution and process of its delivery. You mentioned Planet Money. One of the things that makes this a GREAT show is the PRESENTATION. The hosts often sound like a couple of good friends, sitting on couches in a living room just casually talking away. The audience is "eavesdropping" without feeling guilty about it because they are made to feel like they ARE a part of conversation, it's just that they haven't chosen to chime in. Now, compare that to column inches of jusified type in the Great Grey Lady and it's no contest who wins. Posted by: Brock Meeks at March 15, 2010 10:05 AM | Permalink The medical community has studied informability and competence for many years as it pertains to consent. - The capacity to receive information It is an interesting analogy. It can be applied to #6 in the post where Alex Bloomberg didn’t understand it either and had to be informable and become competent to explain it to others. It also brings into question how Bloomberg "overcame" the informability obstacles of his audience. One of the primary problems with the media today is explained nicely in the following quote: The Internet (Facebook, Twitter, YouTube and the blogosphere) offer so many contradictory “facts,” “truths” and “informed opinions” that people everywhere can essentially select and interpret facts in a way that accords with their own personal, idiosyncratic and often flat-wrong versions of reality. The greater the flow and amount of information, the more likely it will degrade toward noise or sterile uniformity. People deluged by a flood of meaningless variety quickly reach a saturation point where, as a means of self-defense, they develop the capacity to tune most everything out and become extremely selective, jaded, blasé and callous. And people bombarded by redundant information come to view life as banal, colorless, insipid, boring and characterless. As an aspiring journalist Jay, I am wondering what are some day to day actions I can take to push this movement along. The context movement will no doubt eliminate or lessen this ennui. Posted by: Gabriel at March 17, 2010 2:32 PM | Permalink > "But I wonder how the global warming example fits in. There's no shortage of good explanations available on the science of global warming--by both scientists and journalists--and yet I'd bet a good portion of the American public doesn't understand it at all." Hilary, if the public's been convinced by the press (and by the PR folks that pull their strings) that climate change is too confusing to understand, they'll turn their brains off when they encounter it. "Don't confuse me with facts" is a rational response, if you can't tell the real facts from the impostors. Here's what journalists have done. If there is a hell... (beyond the one we're creating on earth, I mean)
Posted by: Anna Haynes at March 19, 2010 7:58 PM | Permalink (i.e., a belated ditto to Gabriel above) Posted by: Anna Haynes at March 19, 2010 8:01 PM | Permalink A very important set of questions, Donica. Great read. Spin, science and climate change Politics, like journalism, tends to simplify and exaggerate.Simplify and exaggerate in episodic updates without the background narrative necessary to make sense of the news. Sounds very descriptive to this news consumer. |
|