The fact there's a Google News Labs team is only one of the many points I have to conclude that the space is heating up, so maybe that's not fair. But it's definitely (almost tangibly) different than it was even a six months ago. The level of interest in the space compared to last year is seriously something along the lines of 10x. Two years ago nobody cared.
As far as newspapers go, you're right - the important thing is controlling the channel. Google, FB, and Twitter all want to control that channel, but they need the data in order to do so. Reliable, verified breaking news data is much harder to come by than one would think, especially at scale.
That's very interesting. What are some of the problems with getting reliable news data at scale? And what's the value of getting it? Specifically, how will you make money?
Sorry for the prying questions. I'm not dismissing. Just the opposite: I'm quite interested to see whether it could work.
I understand companies might pay for access to that, but why? When most people get their news from Reddit, and when techies get their news from HN, what's left? It seems hard to beat a community of upvotes.
To phrase it more precisely, why is access to news within 60 seconds more valuable than access within 20 minutes? Are stock traders going to be the customer, or who? There doesn't seem to be a way to deploy advertising with this, so it's hard to understand why it will make money.
Is the goal to sell this to TV stations who then show ads after they tell people about breaking news?
> When most people get their news from Reddit, and when techies get their news from HN
"Most people" have never heard of Reddit or HN.
The difference isn't 60 seconds vs. 20 minutes; it's <=20 minutes vs 6 hours. Do a Google News search for a breaking news event and for a few hours you only find yesterday's articles.
I wonder who would pay for that and why? Since advertisements can't be deployed through such a service, the service will need customers. Are those customers companies, or individuals? Are you going to sell to people involved in the stock market, like day traders? Or what type of customers would find this valuable?
The notion of accuracy being "overrated" in any aspect of news is very troubling to me. Wars are fought, people are killed, and lives are destroyed as a result of inaccurate news.
For some news events up to 30% of the content being shared (and going viral) across social networks is verifiably false. That sways beliefs and opinions and affect lives. To conclude otherwise is completely irresponsible and incredibly dangerous. That's dangerous for society, dangerous for democracy, and... I can't believe I'm actually arguing that inaccurate information being shared as if it were accurate is dangerous.
Hyperbole aside, you are conflating two things: The accuracy of breaking news with the dissemination of that news. Breaking news is very often inaccurate. No amount of fact checking can fix it because in most cases no other news is available to counter the initial report. Unless you have reporters on the ground who can provide a more accurate report, you can't hope to improve that report. When a later report becomes available it becomes the newest report.
Like Infobitt, you appear to be aiming for a metric of quality ("accuracy") that the market doesn't care about that much. Take the recent plane crash in Taiwan as an example. At first it was reported that there were one or two casualties. Later reports said 13 dead. Most people are like "Oh really?" But, even though the first report is less accurate, it has more value to people than the second (unless you have family on the plane).
As for the distribution of news, unless you are Facebook you have no control over how news is shared. In fact, most people don't care about the casualties, they just want to share the video of the plane crashing across the bridge.
News spreads however it's first reported. Sometimes it's corrected later, but often times it's picked up and spread, even by major news organizations. You are assuming that news will eventually be corrected by news organizations. First, that's a false assumption (https://news.ycombinator.com/item?id=8757421), and second, for many it's already too late by that point.
An anecdote of one false rumor not mattering doesn't change the fact that accurate news makes a difference.
I'm actually surprised by your comment. From my personal experience, it's not "up to 30%", it's more like "up to 80%". From lies in headlines[0] that are debunked in main text, to inaccuracies, to agendas, most of the things I see in news are speculation and/or falsehoods.
And yes, it indeed affects lives. It is dangerous for society and democracy. But citizens don't seem to care, and news outlets earn money this way.
I would sincerely like Google hitting news agencies the way they fight spam and SEO. Cut out as much bullshit as possible.
[0] - as an example of a most blatant lie in a headline I've seen, I recall browsing a news site with my mother, when we saw a headline along the lines of "Poland is the poorest country in the EU [by some metric]!". And the main text said more-less: "... except the following 12 countries which are even poorer". I wonder how much damage this kind of decpetion in headlines does. Most people don't read every article, they just skim headlines - so when you put a false claim there, that's what most people are going to remember.
As far as newspapers go, you're right - the important thing is controlling the channel. Google, FB, and Twitter all want to control that channel, but they need the data in order to do so. Reliable, verified breaking news data is much harder to come by than one would think, especially at scale.