Tumblelog by Soup.io
Newer posts are loading.
You are at the newest post.
Click here to check if anything new just came in.

June 30 2011

19:00

A year after its big redesign, how Google News is thinking about the best ways to present news stories

It’s been a year since Google News launched its big redesign, the first major update of the Google News interface since it launched in 2002. The revamp put a new emphasis on customized news content, focusing in particular on the social elements of news: personalization and, then, sharing.

The design you’ll see on the site today isn’t too far off from what was introduced last year. It still strikes a balance between personalization and serendipity, with a design that is dominated by a Top Stories stream, and filled out by customized stories, locally relevant articles, Spotlight-ed items, most-shared pieces, and other content.

But there have been tweaks, too, many of them aimed at writing into the service a happy medium within the polar aspects of news consumption: something between total personalization and total universality; between breadth and depth; between pre-existing interests and discovery; between want to know and need to know; between expectation and serendipity.

I recently spoke with Andre Rohe, the product’s lead engineer, to learn a bit more about the way he and his colleagues are thinking about news presentation as they continue to build out Google News.

News content: breadth and depth

One thing Rohe highlighted: Google News’ desire to help users indulge their curiosity about particular news events. Last month, Google News tweaked its interface to make it easier for users both to scan for stories that might interest them and to dig deeper once they’ve found them. The Top Story on Google News at any given moment is now expanded — which is to say, visually contextualized, with clustered links and multimedia offerings — by default.

The expansion-default UI emphasizes the diversity of coverage surrounding a given news event, Rohe notes. Links included in an expanded cluster (just like a regular cluster) might include opinion pieces, local and international news, in-depth articles, satire pieces, and, intriguingly, highly cited pieces. (Relevant Wikipedia articles are also included — since, Rohe notes, Wikipedia can often offer great context for news stories, and occasionally even offer news coverage itself.) Multimedia — videos, images, etc. — makes it into the mix through a slider mechanism at the bottom of the expanded entry.

The updated UI lets people “get into the breadth of the story,” Rohe notes — and breadth, in this case, can actually equate to its own kind of depth. Some big stories will generate something in the neighborhood of 20,000 news articles, Rohe notes. “The idea is, how can we get a good summary of all the aspects that are inside of these 20,000?” The even broader goal is to make it easier for users to dig into the stories that interest them, and to benefit, in the process, from the diversity of news coverage that has been Google News’ driving goal since Krishna Bharat founded it. (And Google News’ quite-logical-but-also-significant default expansion to the top Top Story, I’d add — a story that is ostensibly, if not always, one of some kind of civic import — provides users with a nice nudge of encouragement to explore the stories that are not only curiosity-inducing, but also just important to know about.)

Personalization: explicit and implicit

Google News is also doing a lot of thinking about the best ways to personalize news content for its users. The product currently makes use of two main types of customization, Rohe notes: the explicit and the implicit. Explicit personalization is the kind Google News emphasized in last year’s redesign, the kind that asks users to tell Google their interests so their news results can be appropriately tailored.

But you don’t always know what you like. So, starting this April, signed-in Google News users in the U.S. began seeing stories in their “News for You” feeds that were based not on their stated preferences, but on their behavior: their news-related web histories. (For example, within Google search and its other services, if you click on a lot of articles about Lady Gaga, Google News will serve you up breaking news about Lady Gaga.) “We found in testing that more users clicked on more stories when we added this automatic personalization,” software engineer Lucian Cionca explained, “sending more traffic to publishers.”

Google News is also experimenting with ways to combine explicit and implicit personalization — through an article recommendation process that surfaces stories users have exhibited an implicit interest in, and giving them the option to convert that into stated interest. (Have your web searches revealed a hidden passion for all news Gaga-related? Google will give you the option to make it official on your Google News feed — or to, you know, not.) Ultimately, Rohe notes, “it’s about making it easy for the users to say what they have an interest in.”

Mobile: updates beyond design

Google News, like so many other products at Google, is increasingly thinking in terms of mobile. In last year’s redesign, Google News launched an auto-local section, “News Near You,” that serves up geo-targeted news items. And this April, Google released a mobile version of Google News designed especially for low-end phones. The implications of news consumption’s shift to mobile platforms are, to repeat the obvious, huge; it’ll be interesting to see how Rohe and his colleagues address them — and how consumers adjust to the changes.

May 13 2011

14:00

This Week in Review: New business models and traffic drivers in online news, and wrangling over app ads

Every Friday, Mark Coddington sums up the week’s top stories about the future of news.

Leaving the old ad model behind: Much of the commentary about digital news this week was generated by two big reports, one on the business of digital journalism and the other on its consumption. We’ll start on the business side, with the Columbia j-school’s study on what we know so far about the viability of various digital journalism business models. As Poynter’s Bill Mitchell suggested, the best entry point into the 146-page report might be the nine recommendations that form its conclusion.

Mitchell summed the report up in three themes: The audience for journalism is growing, though translating that into revenue is a challenge; the old model of banner ads isn’t cutting it, and news orgs need to look for new forms of advertising; and news orgs need to play better with aggregators and sharpen their own aggregation skills. In his response to the study, Reuters’ Felix Salmon focused on the advertising angle, arguing that journalism and advertising have too long been linked by mere adjacency and that “when you move away from the ad-adjacency model, however, things get a lot more interesting and exciting.”

The New York Times’ story on the report centered on advertising, too, particularly the growing need for journalists to learn about the business side of their products. (That was media consultant Mark Potts’ main takeaway, too.) Emily Bell, a scholar at the center that released the study, said that while journalists need to understand the business of their industry, integrating news and sales staffs isn’t necessarily the way to go.

The J-Lab’s Jan Schaffer recommended that news orgs respond to their business problems by learning from smaller startups and incorporating them more thoroughly into the journalism ecosystem. And paidContent’s Staci Kramer advised news orgs to focus on regular audiences rather than fly-by visitors: “Outwardly we like to complain about content farms; in reality, a lot of what news outlets are doing to the side of those front-page stories isn’t very different.”

Facebook’s growth as news driver: The other major report was released by the Pew Research Center’s Project for Excellence in Journalism and looked at how people access news on the web. This study, too, found that despite a small core of frequent users, news sites are dependent on casual users who visit sites infrequently and don’t stay long when they’re there. Poynter’s Rick Edmonds conveniently distilled the study into five big takeaways.

The study also found that while Google is still the top referrer to major news sites, Facebook is quickly emerging as a significant news driver, too. University of British Columbia j-prof Alfred Hermida said this lines up with recent research he’s done among Canadians, and GigaOM’s Mathew Ingram said it showed that while Google is a dominant source for online news now, Facebook is primed to succeed it.

Meanwhile, the study also found that surprisingly little traffic to news sites is driven by Twitter. Lauren Dugan of All Twitter said this finding casts some doubt on the idea that Twitter is “a huge link-sharing playground,” though the Wall Street Journal’s Zach Seward said the study misses that Twitter referrals are undercounted.

The Twitter undercounting was one of several problems that TBD’s Steve Buttry had about the study, including inconsistent language to characterize findings and a bias toward large news organizations. “This study probably has some helpful data. But it has too many huge holes and indications of bias to have much value,” Buttry wrote.

Pricing ads and subscriptions on tablets: Condé Nast became the third major magazine publisher to reach an agreement with Apple on app subscriptions, and one of the first to offer an in-app subscription, with The New Yorker available now. (Wired subscriptions are coming next month.) Time Inc., which reached a deal with Apple last week, clarified that it won’t include in-app subscriptions, which would be where Apple takes that now-infamous 30% cut. The Financial Times, meanwhile, is still negotiating with Apple.

Forbes’ Jeff Bercovici explained why publishers may be warming to Apple’s deal: Turns out, more people are willing to share their personal data with publishers feared. Still, Mathew Ingram of GigaOM used iFlowReader’s bad Apple experience as a warning to other companies about the dangers of getting into bed with Apple.

Now that Apple-publisher relations have thawed, the New York Times’ David Carr moved to the next issue: Negotiations between publishers and advertisers over how valuable in-app ads are, and how much those ads should cost. Time.com’s Chris Gayomali wondered why magazines are more than giving away app subscriptions with print subscriptions, and concluded that it’s about getting more eyeballs on the print product, not the app, in order to maintain the all-important ad rate base.

In other words, Carr said in another post, publishers are following the old magazine model, where the product is priced below cost and the money is made off advertising instead. He questioned the wisdom of applying that strategy to tablets: “the rich advertising opportunity that will produce may be a less durable and less stable business than grinding out highly profitable circulation over the long haul.”

A postmortem on Bin Laden coverage: It’s now been close to two weeks since the news of Osama bin Laden’s death broke on Twitter, but plenty of folks were still discussing how the story was broken and covered. Gilad Lotan and Devin Gaffney of SocialFlow put together some fascinating visualizations of how the news spread on Twitter, especially the central roles of Donald Rumsfeld staffer Keith Urbahn and New York Times reporter Brian Stelter. Mashable’s Chris Taylor concluded from the data that trustworthiness and having active followers (as opposed to just lots of followers) are more important than ever on Twitter.

Media consultant Frederic Filloux was mostly reassured by the way the traditional news outlets handled the story online: “For once, editorial seems to evolve at a faster pace than the business side.” There were still folks cautioning against going overboard on Twitter-as-news hype, while the Telegraph’s Emma Barnett wondered why pundits are still so surprised at the significant role Twitter and Facebook play in breaking news. (“It’s exactly what they were designed for.”)

New York Times public editor Arthur Brisbane gave the blow-by-blow of how his paper responded to the story, highlighting a few tweets by Times reporters and editors. Reuters’ Felix Salmon chastised Brisbane for not including Brian Stelter’s tweets, which were posted a good 15 minutes before the ones he included. The exclusion, Salmon surmised, might indicate that the Times doesn’t see what Stelter did on Twitter as reporting.

Google News founder Krishna Bharat compared the way Google handled 9/11 and Bin Laden’s death, marveling at how much more breaking-news coverage is available on the web now. The Lab’s Megan Garber used the occasion to glean some insights from Bharat about trusting the authority of the algorithm to provide a rich palette of news, but at Search Engine Land, Danny Sullivan used the Bin Laden coverage to point out some flaws in Google News’ algorithm.

Reading roundup: Lots of interesting little rabbit trails to choose from this week. Here are a few:

— ComScore’s April traffic numbers are out, and there were a number of storylines flowing out of them: Cable news sources are beating print ones in web traffic, the New York Times’ numbers are down (as expected) after implementation of its paywall, and Gawker’s numbers are starting to come back after dropping last year with its redesign.

— Last week, ESPN columnist Rick Reilly told graduating students at the University of Colorado’s j-school to never write for free. That prompted Jason Fry of the National Sports Journalism Center and Craig Calcaterra of MSNBC.com’s Hardball Talk to expound on the virtues of writing for free, though Slate’s Tom Scocca took Reilly’s side.

— Late last week, Google lost an appeal to a 2007 Belgian ruling forcing it to pay newspapers for gaining revenue for linking to their stories on Google News.

— Finally, two thoughtful pieces on brands and journalism: Jason Fry at Poynter on assessing the value of organizational and personal brands, and Vadim Lavrusik at the Lab on journalists building their brands via Facebook.

March 04 2011

16:00

November 16 2010

19:30

Google News experiments with metatags for publishers to give “credit where credit is due”

One of the biggest challenges Google News faces is one that seems navel-gazingly philosophical, but is in fact completely practical: how to determine authorship. In the glut of information on the web, much of it is, if not completely duplicative, then at least derivative of a primary source. Google is trying to build a way to bake an article’s originality into its no-humans-used algorithm.

Today, it’s rolling out an experiment that hopes to tackle the “original authorship” problem: two new metatags, syndication-source and original-source, intended to attribute authorship, via URLs, into the back end of news on the web. Though the tags will work in slightly different ways, Googlers Eric Weigle and Abe Epton note in a blog post, “for both the aim is to allow publishers to take credit for their work and give credit to other journalists.”

Metatags are just one of the many tools Google uses to determine which articles most deserve news consumers’ attention. They work, essentially, by including data about articles within webpages, data that help inform Google’s search algorithms. Google itself already relies on such tagging to help its main search engine read and contextualize the web. (Remember Rupert Murdoch’s so-far-unrealized threats to opt out of Google searches? He would have done it with a noindex tag.)

The tags are simple lines of HTML:

<meta name="syndication-source" content="http://www.example.com/wire_story_1.html">

<meta name="original-source" content="http://www.example.com/scoop_article_2.html">

And they’ll work, Weigle and Epton explain, like this:

syndication-source indicates the preferred URL for a syndicated article. If two versions of an article are exactly the same, or only very slightly modified, we’re asking publishers to use syndication-source to point us to the one they would like Google News to use. For example, if Publisher X syndicates stories to Publisher Y, both should put the following metatag on those articles:

original-source indicates the URL of the first article to report on a story. We encourage publishers to use this metatag to give credit to the source that broke the story. We recognize that this can sometimes be tough to determine. But the intent of this tag is to reward hard work and journalistic enterprise.

(This latter, original-source, is similar to Google’s canonical tag — but original-source will be specific to Google News rather than all of Google’s crawlers.)

Google News is asking publishers to use the new tags under the broad logic that “credit where credit is due” will benefit everyone: users, publishers, and Google. A karma-via-code kind of thing. So, yep: Google News, in its latest attempt to work directly with news publishers, is trusting competing news organizations to credit each other. And it’s also, interestingly, relying on publishers to take an active role in developing its own news search algorithms. In some sense, this is an experiment in crowdsourcing — with news publishers being the crowd.

At the moment, there are no ready-made tools for publishers to use these tags in their webpages — although one presumes, if they get any traction at all, there’ll be a plugins for many of the various content management systems in use at news organizations.

The tags, for any would-be Google Gamers out there, won’t affect articles’ ranking in Google News — at least not yet. (Sorry, folks!) What it will do, however, is provide Google with some valuable data — not just about how its new tags work, but also about how willing news publishers prove to be when it comes to the still-touchy process of credit-giving. That’s a question Google News has been trying to tackle for some time. “We think it is a promising method for detecting originality among a diverse set of news articles,” the tags’ explanation page notes, “but we won’t know for sure until we’ve seen a lot of data. By releasing this tag, we’re asking publishers to participate in an experiment that we hope will improve Google News and, ultimately, online journalism.”

August 24 2010

10:50
Older posts are this way If this message doesn't go away, click anywhere on the page to continue loading posts.
Could not load more posts
Maybe Soup is currently being updated? I'll try again automatically in a few seconds...
Just a second, loading more posts...
You've reached the end.

Don't be the product, buy the product!

Schweinderl