New York University professor Jay Rosen has spent almost 15 years working on this idea, work that has included projects like NewAssignment.net in 2006 and a joint venture with The Huffington Post called OffTheBus, which originally launched in 2008 and had at least one spectacular success). More recently, he has built a kind of real-time journalism lab at NYU called Studio 20, and is helping his students not only develop new ideas for networked reporting, but work with a number of media companies to actually implement those ideas.

The shock of inclusion is not as severe

Rosen isn’t just leaving this to his students: he himself is also working on a joint venture with Quartz, the business site that is part of Atlantic Media, to explore the best ways to do “networked journalism” in real time — a venture he launched on Monday night. In a somewhat unusual partnership that seems more like a consulting arrangement than a typical journalism-school role, Rosen asked Quartz for the “specs” of what they were looking for, and then tried to meet them.

Sort of like a consultancy that gets paid in puzzles. My idea of journalism research: these "specs" from @qz editors. studio20nyu.tumblr.com/post/503459375…—
Jay Rosen (@jayrosen_nyu) May 14, 2013

In the specifications, Quartz says it wants “to put together a suite of tools and techniques for quickly booting up a network around a fast-moving, ongoing global news story that cuts across traditional beat boundaries.” Gideon Lichfield, the site’s global news editor, has written in the past about how Quartz sees its reporters and writers as indulging in or exploring “obsessions” rather than typical beats, and Rosen said it saw the need for new tools to do that.

In an IM interview (which is embedded in full below, with edits made for clarity) Rosen said that he believes the cultural barriers to seeing the crowd as having something to contribute to journalism — what media theorist Clay Shirky has called the “shock of inclusion” — have been lowered somewhat, so there is less of a sales job for journalists who want to experiment with these approaches.

“That is less of a factor than it was years ago. There are enough people who know what ‘readers know more than I do’ means, and they have experience with the reality of it.”

Remember the 90-percent rule

Rosen also said that there are enough journalists and others even in traditional newsrooms and media entities who are interested in new ways of reaching out to what Rosen calls “the people formerly known as the audience,” and are just looking for help. So Studio 20 has partnerships with outlets as varied as the Wall Street Journal, ProPublica and Mashable in which students work with the partner to develop and implement new tools and methods.

In terms of what media outlets need to know before they begin this process, Rosen said one important factor is knowing that whatever they do will be governed by the “90-percent rule” — a rule of thumb in social media that suggests most crowdsourcing projects will see about 1 percent of the participants contribute heavily and 9 percent contribute somewhat, with 90 percent just “lurking.”

“90 percent will never participate, so what do we have for them? 10 percent might engage, but you have to have the right ask, the right incentives, the right UI. One percent are your core contributors, but you have to find them, deeply engage them, compensate them. That is way harder than ‘let’s crowdsource this!’”

Sources can now go direct

In some cases, compensation might be monetary, Rosen says — or it might take the form of other rewards (The Guardian and ProPublica have both talked about their experiments with crowdsourcing projects in the past, and what they have learned about how to structure them so that people are encouraged to participate). Mayhill Fowler eventually left the Huffington Post project in part because she wasn’t compensated for her work.

Rosen also said that crowdsourcing doesn’t always have to involve building tools: for example, two of his students used Reddit threads (called sub-Reddits) and extracted information about specific topics that later turned into stories for Mashable.

The NYU journalism professor agreed that good beat reporters have always used some form of crowdsourcing in their work, but says it is much easier now to reach out and find high-quality sources of information in real time. And he added that there is one major difference between now and then: namely, that sources can publish themselves and “go direct,” as blogging pioneer Dave Winer has described it, and that changes the balance of power for journalists. If anything, he says, this makes the need for effective crowdsourcing even more acute.

Post and thumbnail photos courtesy of Flickr users Christian Scholz and Jan Arief Purwanto

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Frenemy mine: The pros and cons of social partnerships for online media companies
• Content monetization: News licensing and syndication still need marketplaces and infrastructure
• Best practices in optimizing content for social engagement

Open Garden is trying to build a mesh network for mobile devices that does away with the notion that a single device must connect to the internet through a pre-defined path. Instead it’s created tablet, smartphone and PC software that will allow their host devices to crowdsource their connections by forming ad hoc mesh networks with other Open Garden devices. Collectively those clients determine which device or devices have the best connections to the internet – whether its 3G, 4G, Wi-Fi or Ethernet – and route traffic through those links.

It’s a hard notion for many people to grasp since we tend think of our connections as individual property, but Open Garden is trying to change that mindset. If everyone shares their connection, then everyone’s link to the network is optimized. Open Garden co-founder and CEO Micha Benoliel believes that Google Glass is an ideal candidate to demonstrate that principle.

Open Garden has succeeded in installing its networking software for Glass, which means the wearable can now act as a node in a larger mesh. The setup has an immediate advantage for Glass owners: they don’t need to pair Glass to their Android devices through Bluetooth or subscribe to a tethering plan from their operators. The Open Garden client makes the connection automatically over Wi-Fi.

But the bigger implication is that Glass could eventually become part of a larger networked community ranging far beyond its owner’s personal constellation of devices. A Glass user, for instance, might be able to leave his smartphone at home or in his office. Instead of losing connectivity, the Google headgear would search out other Open Garden devices – even other Glass units — hopping from node to node until it wended its way onto a mobile or broadband network.

To make its software work on Glass, Open Garden had to do more than just tap into the Mirror API. Benoliel said Open Garden’s developers had to delve deeper into the Android kernel to gain access to its networking functions, which will make distributing its software all but impossible when Glass becomes generally available to the public. But Benoliel is hoping Google will see the advantage of making Glass part of the larger mesh and is lobbying the search giant to clear the way for its app.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• What to watch in mobile in 2013
• Survey: How apps can solve photo management
• Why retailers should forget showrooming and turn to in-store Wi-Fi

Some of the commentary about Reddit and the bombings has made it seem as though all of Reddit was engaged in a massive “witch hunt” to find the identity of the suspects in Boston. But the reality is that other parts of Reddit were doing things that were much more valuable, and I think we shouldn’t lose sight of that. So here are a few things that I think Reddit got right:

• It collected verified information: There were multiple Reddit threads that did nothing but curate or aggregate information about the bombings, including links to police reports, news articles and other sources. These threads also helped collect photos and video clips of the Boston marathon that might have contained useful information — and asked anyone with that information to also send those photos and clips to the authorities.
• It helped people who wanted to help: A number of the threads early on in the aftermath contained lists of all the things that users could do if they wanted to assist not just the investigation but the people who had been injured — from links to Google’s Person Finder and the Red Cross help line to information on where to pick up bags left at the scene, or airlines who had changed their policies on cancelling flights as a result of the attacks.
• It helped to verify facts: In most of the information-gathering threads, there is real-time verification of the info occurring, as users challenge other users to prove their claims. It is almost identical to the discussion that occurs on a Wikipedia “talk” page, in which editors try to verify the information that is being posted to an entry. Multiple updates occur within minutes of each other, and each one is marked with the time and any edits that took place.

Is Reddit capable of journalism? Yes

Even Reddit itself posted a disclaimer on one of its threads that said it isn’t trying to be a media entity, and that what it does isn’t journalism. And the user who created the “Find Boston Bombers” sub-Reddit or thread told The Atlantic that he doesn’t think of it as journalism either, and that no one should ever rely on such threads as a source because there is so much conflicting information flying around. He also admitted that the attempt to identify the bombers from photos was “a disaster.”

So if even Reddit itself doesn’t claim to be producing journalism, why do I keep saying it is? Because I think Reddit and Twitter and other social tools are broadening the concept of journalism. Some, like my friend Raju Narisetti from News Corp., believe that we should call this kind of thing something else — like that horrible term “user-generated content” — and leave the term journalism for things that are produced by professionals who are held to standards (although some might question whether the New York Post fits that description).

@mathewi you should fear. Find a new definition for non-journalism and use it. Why call ugc, crowds as journalism. It isnt.—
Raju Narisetti (@rajunarisetti) April 22, 2013

In a nutshell, I believe that journalism is being atomized — that is, broken down into its component parts. One of those is the news-gathering function, whether it’s from eyewitnesses or just on-the-ground observation. This part of journalism can and is being done by anyone, thanks to what Om has called the “democratization of distribution,” and it can be hugely valuable. And the verification function has also been outsourced, so that people like Eliot Higgins can play a key role in identifying Syria weapons without leaving their apartment.

Reddit may have failed badly in one specific thread, and that is unfortunate. But other parts of the site have and continue to perform valuable functions that I see as part of the broader landscape or ecosystem of networked journalism. Instead of focusing just on the downside of that community, we should be thinking about how to take advantage of it — how to turn a negative feedback loop into a positive one.

Post and thumbnail photo courtesy of Shutterstock / wellphoto K

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Content monetization: News licensing and syndication still need marketplaces and infrastructure
• Frenemy mine: The pros and cons of social partnerships for online media companies
• Sector RoadMap: Content personalization in 2013

Within hours of the explosions in Boston, members of the Reddit community had created a thread (or sub-Reddit) about the incident, in an attempt to identify potential suspects. Users posted photos that had been published online or submitted by onlookers and analyzed video clips, piecing together clues like a specific kind of zipper that was used on a backpack found at the scene. Eventually, two potential suspects were identified — including one who posted a message on Facebook about his innocence.

Plenty of mistakes to go around

After some more investigation and crowdsourced information gathering, users on the Reddit thread seemed more or less convinced that the two were not likely to be the actual bombers, and eventually declared them “cleared.” Meanwhile, the New York Post identified the same two people as potential suspects and published their photos on the front page (both suspects have now been identified — one was reportedly shot by police on Friday and as of mid-afternoon on Friday the other was said to be on the run).

Alexis Madrigal at The Atlantic wrote that the process taking place on Reddit amounted to “vigilantism,” and was reprehensible, and warned against encouraging untrained people to try and determine the validity of forensic evidence after such an event. But is what happened on Reddit so bad? And is it any worse than what the traditional media have done in similar situations? I’m not convinced.

Yes, users of Reddit made mistakes — plenty of them, including identifying the wrong person as a suspect a second time on Thursday after erroneous information emerged from police scanners and other sources, something which caused a considerable amount of grief for a young man’s family and led to an apology posted on Reddit by a moderator.

But it should be noted that CNN and the NY Post have made plenty of mistakes as well, something Ryan Chittum of the Columbia Journalism Review doesn’t really mention in his post about how brilliant the traditional media was and how wrong Reddit has been. The larger point is that this isn’t an either/or situation — crowdsourcing is valuable, and has been valuable for journalism and will continue to be. This is admittedly not an example of it at its finest.

Remember when we didn’t think random people putting together an encyclopedia would ever work? And yet it has — in part because it has a lot more structure than Reddit or 4chan. And those sites would probably be a lot more useful in these cases if people spent more time thinking and less time typing. But that doesn’t negate the value they can provide. The idea of using the knowledge and resources of the crowd is the whole point behind Guardian editor-in-chief Alan Rusbridger’s “open journalism,” and it is a force we need to figure out how to tame, not dismiss as irrelevant based on one incident.

Open journalism works better

Am I calling what Reddit has been doing since the Boston bombings journalism? Yes. It may not encompass the entirety of what we know as journalism, and it is clearly flawed, but it is certainly an important aspect of it — just as Eliot Higgins, an unemployed British accountant, is performing a valuable journalistic act (one that New York Times writer C.J. Chivers has recognized) in verifying smuggled weapons in Syria by watching hundreds of hours of YouTube videos every day, even though no one is paying him to do so.

Will Oremus at Slate makes a fairly persuasive argument that Reddit has in some cases been *more* responsible in its attempts to identify the individuals than some traditional sources, including the Post. This kind of crowdsourced fact-checking and verification of evidence has been going on for years — it’s just more mainstream now. And anyone looking for evidence of someone jumping the gun and encouraging vigilantism doesn’t have to look any further than CNN.

When I wrote recently about the benefits of having journalism occur out in the open, journalism teacher Steve Fox and others said I didn’t spend enough time on the need for verification, and maybe I didn’t, but I believe this also should be done out in the open. In fact, one of the benefits to doing so is the ability to have more eyes on the information at hand — thereby making it easier to filter out the noise and find the signal, or triangulate the truth. As Jay Rosen has said, journalism gets better the more people there are doing it. And that includes Reddit.

Post and thumbnail photo courtesy of Flickr users Christian Scholz and Eva Blue and Jan-Arief Purwanto

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Frenemy mine: The pros and cons of social partnerships for online media companies
• Content monetization: News licensing and syndication still need marketplaces and infrastructure
• Best practices in optimizing content for social engagement

When he headed off to college at the University of Pennsylvania, Kulkarni tried out for the team. He was 5-feet, 9-inches and weighed 130 pounds. He didn’t make it.

But Kulkarni would find another path to basketball success. He never got to don a Dallas Mavericks or Houston Rockets hat on draft day, but he did get a shot to hobnob — and impress — Mavericks owner Mark Cuban and Rockets General Manager Daryl Morey at this year’s MIT Sloan Sports Analytics Conference.

Kulkarni technically studied computer engineering and entrepreneurship but, he joked when I asked the question, “I like to say basketball.” Even though it doesn’t appear on his transcripts, there’s no denying he did his hardcourt homework.

No budget, no problem

Although he didn’t make the varsity basketball team, Kulkarni did play junior varisty ball at Penn — and in doing so spotted a golden opportunity. Penn is the ninth-winningest college basketball program of all time, but it’s no longer a powerhouse and, Kulkarni noted, Ivy League schools simply do not fund athletic programs like some other Division I universities do. There wasn’t a whole video department dedicated solely to processing and analyzing game film. Kulkarni watched the head coach prep for games largely by himself, poring over film to find a few shining examples of good or bad plays that could serve as teaching points.

And then a lightbulb went off for Kulkarni: There must be a whole lot of small colleges and high schools suffering from the same problem. So, in 2008, he launched Krossover.

The company’s flagship service is pretty self-explanatory. Coaches upload their game film after each game. Krossover’s team of hundreds of college students gets to work breaking it down. When the coach wakes up, the last night’s game is online and he can examine just about anything he wants — statistics, individuals plays, where on the field or court most of the action took place. It quite literally analyzes everything that’s quantifiable, and in some cases visualizable, by studying game film. (You can experiment with it yourself here.)

Krossover’s shot chart.

That’s really cool if you’re a high-school or lower-level college coach — and Kulkarni said more than 1,000 of them currently use Krossover — but it’s probably not going to impress too many professional or big-time college coaches. It’s certainly not going to impress the sports-stats superstars that flock to the MIT conference every year. To reach this audience, Krossover needed something new.

Are you smarter than Kevin Durant? (Hint: No)

Hence the startup’s latest idea, an iOS app for testing people’s on-the-court sports knowledge, called sIQ (or sports IQ). Kulkarni said the inspiration for sIQ came while watching Krossover board member and then-head of analytics for the Oklahoma City Thunder NBA franchise Ben Alamar talk about the differences in brain activity between weekend warriors and professional athletes. The biggest is that while wannabe pros make conscious decisions about what to do next, professional athletes just react — they see and they do without hesitation.

Can Kevin Durant do this? Source: Krossover

When you’re measuring someone’s sports IQ, the faster and more-accurately they react, the smarter — in theory — they are. Kulkarni’s first question to Alamar upon hearing this: “Can I take this test and if I do better than Kevin Durant can I get a 10-day contract?”

Funny, yes, but also telling about the mentality of guys who watch a lot of sports — and, like Kulkarni, might still play a lot — and hold out hope in the back of their minds they’re in some way on par with professional athletes. Kulkarni had a hunch the app would go over well at the MIT Sloan event, which is full of, as he described them (and, by proxy, himself and me) “a bunch of sports nerds who probably can’t play worth a damn but know everything about sports.” They could settle a lot of bets over who’s a smarter athlete by watching plays unfold on an iPhone, then when video pauses, predicting the outcome.

So, he told one his developers to take two weeks earlier this year and get something ready for the conference, which took place the first week of March. While there, the Krossover team met Daryl Morey, a former Wall Street quant turned Houston Rockets general manager — an extreme version of Moneyball inspiration Billy Beane in that Morey has no basketball experience and relies almost solely on numbers to make his decisions. He loved the app and quickly dragged over Dallas Mavericks owner (and entrepreneur and investor and blogger) Mark Cuban to play with it. At NBA All-Star Weekend in February, ESPN basketball columnist Bill Simmons played sIQ. Poorly.

By contrast, when Kulkarni gave sIQ to a 15-year-old basketball phenom at the gym where they both play, the kid got the highest scores Kulkarni had ever seen. And not only did he answer correctly, his average response time was about 1 second — about 4 times faster than the guys at Sloan were able to respond.

The fact that it works, as evidenced by the teenage natural’s performance, has some NBA and NFL executives already spotting an opportunity in sIQ. Professional teams — including Alamar’s new employer, the Cleveland Cavaliers — are testing it out as a method to gauge college players’ sports IQs as part of the draft preparation and to train players to react better by using with the app as a way to predict what will happen next in any given situation.

Maybe, Kulkarni suggested, sIQ could become part of the highly publicized NFL Combine and replace the controversial Wonderlick test for gauging draft prospects’ intelligence. It should almost certainly provide a more-compelling platform for high-school coaches trying to get their athletes to study game film.

The positive feedback has Krossover building sIQ as a platform rather than as a just an app. Yes, sports fans will be able to download it and test their IQs, but the company also hopes to build versions specifically for applications like testing and training real athletes. It will start off in basketball, but sports like football, volleyball, soccer, wrestling, boxing and maybe even mixed martial arts are on the horizon.

Taking analytics from the front office to the field

And whether or not sIQ turns out to be the game-changer Krossover hopes it will be, the company seems to be on the right path. Sports analytics is becoming a huge business, but primarily in the front office where executives are trying to figure out who they want on their teams and how much they’re willing to pay. Handicapping the annual NCAA March Madness basketball tournament is a popular pastime, too. (As evidence of how hot the space is right now, Kulkarni said Krossover has raised $4.5 million primarily through angel investors — including some professional athletes — who want to get in on what they see as a sexy business.)

Statwing breaks down the NFL.

However, as I noted when profiling Statwing recently (its full-time-statistician and part-time-sports-geek founders recently uploaded a trove of NFL data for people to play around with), the analytic mindset has yet to trickle down to the coach’s office in most situations and affect decisions such as what type of play to call in what situations. Aside from golf, perhaps, it certainly hasn’t made its way onto the field of play to actually improve players’ performance.

And that was Kulkarni’s major takeaway from the MIT Sloan conference this year: while teams and companies like his are collecting “an obscene amount” of data on every single aspect of nearly every single sport, they’re struggling to find ways to make sense of it.

“The thing everyone’s trying to figure out is: Is there a way for you to find the two or three or four things that will guarantee you a win or at least tip the scale in your favor at any given time,” he said.  “I don’t think anyone has cracked the code.”

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Defining work in the digital age: an analysis by GigaOM Pro
• How data warehousing is now a cost-effective solution for businesses
• Sector RoadMap: Social customer service in 2013

OpenSignal has recorded more than 1,500 signal strength readings in eight metropolitan areas from devices connected to T-Mobile’s network: Denver, Kansas City, Las Vegas, New Orleans, New York City, San Diego, Seattle and the Bay Area. OpenSignal has mapped those signal readings on its blog, though it represents multiple signal readings as single data points, largely to respect the privacy of the testers.

Though the network isn’t commercially launched, OpenSignal CEO Brendan Gill told me that its crowdsourced app is popular with engineers at all of the carriers for ad hoc network measurements. What we’re most likely seeing, Gill said, is a bunch of technical workers from T-Mobile and its vendor partners that have the OpenSignal app loaded and running on their test devices. A good indication of this is that one of the devices sending data is a Samsung Galaxy S4, which isn’t yet available to the general public.

T-Mobile CTO Neville Ray has already confirmed that its LTE network is complete in Las Vegas and Kansas City, so test data from those cities is hardly a surprise. We’ve also seen evidence of the network in NYC: A GigaOM reader mapped out a cluster of cells in Astoria, Queens, and T-Mobile has given several live 4G demos to reporters in Manhattan.

The other five areas are new, but because of their importance you would expect them to show up early on T-Mobile’s national rollout schedule. OpenSignal recorded the biggest concentrations of tests in San Jose, Calif., and surrounding Bay Area cities like Mountain View, Santa Clara and Sunnyvale. OpenSignal even tracked tests in the East Bay, but recorded none in San Francisco proper.

The Seattle area had the second highest concentration of readings, many of them around Bellevue, which just happens to be the location of T-Mobile’s national headquarters. Las Vegas yielded many data points as well, though the signal readings in Denver, Kansas City, New Orleans, New York and San Diego were much more sparse.

OpenSignal also recorded about a dozen speed tests on T-Mobile’s LTE network, averaging an impressive 25 Mbps on the downlink and 8 Mbps on the uplink (though keep in mind that the network is largely empty so there’s no congestion). The speed tests were so few because they must be manually initiated, while OpenSignal’s app takes signal strength readings automatically on a regular basis.

Gill said that he’s fairly convinced that these eight markets will be among the first to launch based on the activity OpenSignal is tracking, though he cautioned that his conclusions don’t constitute a scientific study. The results are dependent on a fairly limited pool of people using OpenSignal’s app, so there are likely many more cities with live LTE networks that the company couldn’t track.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Best practices in optimizing content for social engagement
• A look back at mobile in the third quarter
• Mobile 2012 and beyond

Speaking at GigaOM’s Structure:Data conference Wednesday, Harvard Associate Professor Karim Lakhani said crowdsourcing the genomics project did several things. First and foremost it generated dozens of different approaches to tackling the same problem.

Before the TopCoder contest was created, researchers were considering two different paths of investigation. The contest revealed 89 differing approaches to the problem, 20 of which were extreme values — possibilities Harvard and its National Institutes of Health counterparts had never considered.

Second, the contest was able to create motivation to solve problems that you wouldn’t necessarily find in a group of researchers. An institution like Harvard may have brainpower in spades, Lakhani said, but throwing a bunch of geniuses at a problem doesn’t necessarily lead to result if they’re unmotivated to solve it.

“When you go into a self-selection model you don’t have to worry about motivation,” Lakhani said. There are monetary rewards for winning a TopCoder contest, of course, but taking home a prize is not guaranteed. Everyone has their own motivation for participating, whether it’s cash, experience, scoring reputation points or even the free T-shirts given to each participant. “Because there are large numbers of people participating, there’s a greater chance you’ll find the right skills and the right motivation,” Lakhani said.

Ultimately crowdsourcing your science isn’t a replacement for having smart people of your own, Lakhani said, but it certainly helps, something Harvard’s counterparts at NASA have discovered.

CHeck out the rest of our Structure Data 2013 coverage here, and a video embed of the session follows below:

A transcription of the video follows on the next page

Go to page 2 (of 2) on GigaOM .

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Examining the rise of crowd labor platforms in 2012
• Best practices in optimizing content for social engagement
• The quantified self: hacking the body for better health

Grafetee lets you set up a new embeddable, interactive map on the Grafetee website, for free and within a minute or two. This service was already available, but now bloggers and businesses can pay $20 a month for a premium version that includes a channel in the Grafetee mobile app (for Android and iOS), customizable graphics and tools for moderating user submissions. Enterprise and governmental users can also pay $200 a month to get a webpage under their own domain, along with premium support and API access to their content.

The Finnish police force was an early Grafetee user, having set up a map so citizens could, for example, notify the cops that a certain park should be patrolled on weekends due to troublesome teenagers hanging out there. Since then, others picked up on the idea, such as animal shelters wanting people to flag strays, and mobile operators looking to crowdsource coverage data. According to founder Juha Huttunen, someone even set up a Grafetee map for tracking toxic seaweed during the Finnish summer.

And yes, it can be used for offers too. “If you have a chain of cafes or whatever you can send your own offers to your customers,” Huttunen explained to me. “You would have a channel in the Grafetee app and, using the toolkit, the cafeteria would upload their offers. Anyone subscribing to the channel in the app would see those offers whenever they are close by.”

A couple of Finnish startups have already built their services using Grafetee, and they actually provide a pretty good insight into the range that’s on offer. One, kidd.io, is on the simple side, collecting and displaying the locations of child-friendly services around Helsinki. Hoods.fi, a location-based flea market and garage sale service, shows greater complexity.

There are other services in this space — AmigoCloud targets the enterprise and public administrations, Ushahidi’s Crowdmap (currently in closed beta) is for aid groups tracking crises, and Everplaces is all about community-sourced recommendations – but none that I’ve seen have quite the combination of simplicity and breadth that Grafetee offers.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Best practices in optimizing content for social engagement
• Examining the rise of crowd labor platforms in 2012
• The quantified self: hacking the body for better health

Verizon averaged 14.3 Mbps down and 8.5 Mbps up, according to Root’s new report, but what it lacked in raw speed, Verizon made up for in coverage. Of the 77 markets in which Root performed its own measurements, Verizon had an LTE network up in every one. Meanwhile AT&T’s 4G service was present in only 47 of the 77 at the time Root performed its tests last year. These maps, compiled from Root’s crowdsourced data, show just how far Verizon’s LTE network reaches compared to AT&T and Sprint:

What’s more, Verizon’s coverage within its LTE footprint was much more consistent. When in a Verizon LTE market Root testers found themselves connected to a Verizon LTE signal 93.2 percent of the time, while for AT&T the number was 81.7 percent. We’ve started to see that trend in Root’s city-specific reports: Big Red is reaching further into the suburban and exurban regions of its launch markets than Ma Bell.

But AT&T was quick to point that it has added many more cities since Root compiled its data (Root measured different markets at different times in the second half of 2012). Of the 30 cities where Root found no LTE network, AT&T has since launched networks in 26 of them, AT&T spokesman Seth Bloom said.

Root’s most recent results for New York City, where AT&T boasts the fastest LTE connections.

Sprint only started its LTE rollout last summer so it’s still far behind AT&T and Verizon. Root’s staggered testing regime found LTE networks in only five of the 77 markets measured last year and in even in those five markets it caught an LTE signal only half the time. When Root did find LTE, Sprint averaged 10.3 Mbps down and 4.4 Mbps up. Sprint’s speeds are generally lower because it is using half the spectrum for LTE that AT&T and Verizon are tapping for their rollouts.

T-Mobile won’t launch LTE until later this year, but Root did measure its HSPA+ network performance. T-Mobile averaged 7.3 Mbps on the downlink and 1.5 Mbps on the uplink.

RootMetrics uses both crowdsourced data — drawn from smartphones loaded with its CoverageMap iPhone and Android apps — and professional testing conducted both in vehicles and indoors (For a detailed look at Root’s methodology, check out our video of a recent Root test in Chicago). Root is also working with GigaOM this week at SXSW in Austin to measure the impact that a large conference of mobile savvy users has on city’s mobile data networks.

Correction: An earlier version of this post incorrectly stated T-Mobile’s LTE network will launch next year. T-Mobile’s 4G network will actually go live later this year.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

But as we change where we get our news and turn to places like Twitter for information and verification of facts, it’s important to ask how that polarization will translate to social media – if it will at all. Several journalists discussing the future of news dissemination (something we’ll also be discussing at paidContent Live in April) tied these issues to those of crowdsourced news, particularly in the Middle East, when the tensions between accuracy and access are most apparent.

NBC correspondant Ayman Mohyeldin made an interesting argument about verification, arguing that people should be free to select the accounts they want to follow and personally decide whether to trust that information or not, just as they tune into particular cable shows in the United States and apply their own sense of skepticism to Maddow and Hannity.

“You ultimately choose which channels to watch,” he said. “There’s no reason that should be different in who you follow.”

The argument puts a good deal of trust in the user’s judgement and takes some pressure off journalists or random people on Twitter to present accurate information, but it’s an idea that fellow panelist Andy Carvin has popularized to much controversy recently. The idea came under fire during the spread of misinformation on Twitter during Hurricane Sandy, and certainly has its detractors:

The end user is responsible for deciding who they trust, @AymanM says. True, but most don't have the expertise to decide. #nextnews #sxsw—
Eric Carvin (@EricCarvin) March 10, 2013

But it’s a good reminder that even if we think of cable news as being particularly polarizing right now, news consumption and opinions on Twitter might not be all that different.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Best practices in optimizing content for social engagement
• Examining the rise of crowd labor platforms in 2012
• The quantified self: hacking the body for better health