This will make your Facebook feed a little less annoying.

There’s much to love about Facebook and all the new features it’s been rolling out. But notice how your updated News Feed seems all cluttered up? There’s always a whole bunch of posts that read along the lines of  “you’ll never guess what happened next”.

These articles are known as Clickbait – they tend to carry intriguing headlines which suggest there is more to the story than there actually is. They make use of alluring phrases in order to drive web traffic. And they’re horribly annoying to the reader. It’s a waste of a click, a waste of your time, you basically get nothing out of it, except may be a little annoyed.

Facebook has now intensified its crackdown on so-called clickbait news. The social network says it can now automatically suppress these stories. It has built a technology that scans for tell-tale signs of clickbait headlines on its news feed section – phrases such as “you’ll never guess why” and “his response was priceless”.

This scanning technology was created by a team at Facebook who manually reviewed “thousands of headlines” to put together a list of commonly identifiable clickbait traits.  Headlines which are flagged for using such phrasing will be automatically diminished in prominence on Facebook’s news feed. “We’re making an update to News Feed ranking to further reduce clickbait headlines in the coming weeks,” read a post on Facebook’s blog. “With this update, people will see fewer clickbait stories and more of the stories they want to see higher up in their feeds.”

Image copyright SCREENGRAB/WALESONLINE
Good riddance to stories like these!

Facebook previously tried to tackle clickbait through another means – by timing how long users remained on an external website after clicking on a headline. But it said that this process wasn’t accurate enough. It said it now categorizes headlines as clickbait if they withhold information required to understand what the content of the article is and also if they exaggerate the article to create misleading expectations. “People have told us they like seeing authentic stories the most,” the company added. Let’s hope this move by Facebook will mean we get to read more relevant and meaningful content in our feed, it would be a welcome relief from all the cheap quality gibberish content making its rounds on the internet. Thoughts?


Anisha Sawant

 

European Commission signs Social Media “Code of Conduct” with Silicon Valley’s finest

More worrisome news for free speech advocates as government agencies continue pushing for, what could inevitably result in, online censorship. There is, of course, no place for hate on the internet. And it is absolutely unfair to discriminate against religious/ethnic minorities. But if the government gets to control expressions of hate/anger on the biggest platform connecting the world, what could this mean for freedom of expression in general – which has been a core European value for ages? The European Commission begs to differ..

This does not affect the right to freedom of expression; rather, it refers to conduct publicly inciting to violence or hatred directed against a group of persons or a member of such a group defined by reference to race, colour, religion, descent or national or ethnic origin.

European Commission via LinkedIn

Yesterday, Facebook Inc, Google’s YouTube, Twitter Inc and Microsoft signed an agreement to block illegal hate speech from their services in Europe within 24 hours. This is yet another move that shows the mounting pressure on tech companies to monitor and control content.

The new European Union “code of conduct on illegal online hate speech” states that the above mentioned tech giants will review reports of hate speech and remove or disable access to the content if necessary.

European governments were acting in response to a surge in antisemitic, anti-immigrant and pro-Islamic State commentary on social media.

The companies made light of the deal, saying it was a simple extension of what they already do. Unlike in the US, many forms of hate speech, such as pro-Nazi propaganda, are illegal in some or all European countries, and the major internet companies have the technical ability to block content on a country-by-country basis.

But people familiar with the complicated world of internet content filtering say the EU agreement is part of a broad and worrisome trend toward more government restrictions.

“Other countries will look at this and say, ‘This looks like a good idea, let’s see what leverage I have to get similar agreements,'” said Daphne Keller, former associate general counsel at Google and director of intermediary liability at the Stanford Center for Internet and Society.

“Anybody with an interest in getting certain types of content removed is going to find this interesting.”

Source: Reuters

EC No Place For Hate speech.png

European authorities have been putting tremendous pressure on social media companies to be more aggressive in targeting hate speech online. For example, this celebratory hashtag (#Brusselsisonfire) that surfaced after the March bombing in Belgium. Or when Twitter said in February that it had deleted more than 125,000 ISIS-affiliated accounts.

The Code of Conduct is a “self-regulatory” measure, which means that it’s not legally binding. The code of conduct enumerates a few specific commitments to address the problem:

  • The companies need to have clear and accessible ways of identifying and removing hateful content, and that they must review the majority of the content reported within 24 hours.
  • They have to work more with “civil society organizations” (nonprofits, advocacy organizations, etc.) to target such content.
  • They must train their staff “on current societal developments and to exchange views on the potential for further improvement.”

Despite the agreement signed today, these companies have a pretty tense relationship with European regulators, especially Google and Facebook. The EU is currently leveling antitrust charges against Google, and German regulators have begun looking into Facebook’s practices.

Source: Recode

You can read more about the code of conduct here. Leave us a comment and let us know what you think!

 


Anisha Sawant

Is AI getting ready to outperform humans? Zuckerberg sure thinks so!

Technology is advancing at a tremendous rate. Everything seems to be getting smarter – homes, cars, shoes.. you name it! A lot of attention is being focused on AI – with tech giants now investing heavily in incorporating bots into their services. But are we really looking at a not-so-distant future where AI will surpass humans in doing things that, well, make us human?

Recently, Facebook Founder and CEO, Mark Zuckerberg was asked about the future of the machine learning technology that powers its Messenger bots. Here’s what he said..

“So the biggest thing that we’re focused on with artificial intelligence is building computer services that have better perception than people… So the basic human senses like seeing, hearing, language, core things that we do. I think it’s possible to get to the point in the next five to 10 years where we have computer systems that are better than people at each of those things.”

Source: The Verge

 

Bit scary, isn’t it? The thought of computer systems with abilities that make us biologically human – senses and skills we rely on to interact with our world. Makes it seem rather inevitable that a wide variety of jobs currently done by people will no longer require expensive human labor, because artificial intelligence will be far superior.

However, Zuckerberg goes on to say..

“That doesn’t mean that the computers will be thinking or be generally better…”

That isn’t really a great amount of praise for the species he belongs to, but we’ll take whatever little we can get. Thinking here means something more complex than understanding how to carry on a conversation or tell a cat from a dog. It’s about being able to do several different things well, and learning on your own in an unpredictable and unsupervised environment. Generally better is being used in the sense that we are generalists, capable of handling a wide universe of tasks.

Source: The Verge

Human and animal learning is mostly unsupervised. We know how to program machines to do certain tasks, to follow a routine, to understand data and extract meaningful information from it. But we cannot train a machine to master unknown skills on their own. Until unsupervised learning cant be “taught” to machines, humans, as natural beings, will still have the upper hand. There cannot be true AI without unsupervised learning.

Zuckerberg sees bots stepping in for customer service reps and personal assistants.

“One way that I think you’re going to see bots work, between people who are actually driving the businesses directly will need to in some way train or answer questions for people, but we can build artificial intelligence that can learn from people how to automate a lot of that and make that experience a lot faster for people who want to interact with businesses and public figures.”

How should we prepare for this new reality? One possibility is universal basic income (more on this another time). The other is to redesign our educational system with a focus on new skills that will equip our future generations to find meaningful jobs in a world of advanced AI.

But when you look at the less-than-impressive launch of recent smart bots from Microsoft and Facebook  — they turned out to be slow (Messenger bots that don’t respond for several minutes), racist (Tay’s mean tweets) and in general, not the smartest — you can’t help but wonder if this is just another one of those AI hype rounds that will soon lose traction. What are your thoughts on this? Will machines ever really be better than humans, at being human? Leave us a comment below!

 


Anisha Sawant

 

Check out Facebook’s latest gadget — it lets you shoot 3D video!

Yesterday, at F8 – Facebook’s annual global developer conference, Founder & CEO Mark Zuckerberg announced that they have successfully built a 360-degree camera called Surround360. The company has spent the past year working on a stereoscopic camera that will allow us to produce 3D video.

Zuckerberg’s adorable 4-month old daughter, Max, is reaching the age where she learns to walk; and when the time comes, this doting daddy wants to go all out. “When Max takes her first step, we’ll be able to capture that whole scene, not just write down the date or take a photo or take a little 2-D video,” he says. “The people we want to share this with…can go there. They can experience that moment.”

 

FB surround360

 

This is all part of Zuckerberg’s effort to move the Internet beyond text and photos and video to a new mode of communication. The hope is, what begins with 360-degree video will extend to the kind of virtual reality offered by the company’s Oculus headset. “Over time, people get richer and richer tools to communicate and express what they care about,” Zuckerberg said in an interview with the folks over at WIRED. “What’s next? You’re clearly going to be able to experience whole scenes, whether that’s captured through some kind of 360-degree camera or it’s computer-generated, as games are.”

Companies like Nokia and Google share his vision. But what sets Facebook apart is they’re giving the designs away – yes, for free – they’ll be posted to Github this summer.

Zuckerberg sees this as another part of the company’s mission to Connect the World. His idea is to build Facebook around this mission and try all sorts of things to get closer to achieving it. “The real goal is to build the community. A lot of times, the best way to advance the technology is to work on it as a community.”

We love this thought. After all, the greatest technological breakthroughs have come from people around the world, working together, focused on a common goal. How do you feel about Facebook’s new strategy to bring the world together? Let us know in the comments!

 


Anisha Sawant

 

 

Facebook just did something awesome for the visually impaired!

Mark Zuckerberg’s mission with Facebook has been about connecting the world. Over a billion uses across the globe use the social network on a daily basis to keep in touch with their loved ones. Our News Feed is our daily source of everything that’s going on with our Friends. It contains pictures, videos, status updates – and we can like, share, comment on or react to these updates. When you think about it, almost every single post contains some visual information.

People who are blind or visually impaired use something called a Screen Reader to know more about the visual aspects of a website they’re on. A screen reader is a software application that attempts to identify and interpret what is being displayed on the screen. This interpretation is then re-presented to the user with text-to-speech, sound icons, or a Braille output device. Screen readers are a form of Assistive Technology potentially useful to people who are blind, visually impaired, illiterate or learning disabled, often in combination with other AT, such as screen magnifiers.

Source: Wikipedia

Screen readers essentially make use of something called alternative text – text associated with an image that serves the same purpose and conveys the same essential information as the image. In situations where the image is not available to the reader, perhaps because they have turned off images in their web browser or are using a screen reader due to a visual impairment, the alternative text ensures that no information or functionality is lost. Absent or unhelpful alternative text can be a source of frustration for visually impaired users of the Web.

Hence, for people who are blind or visually impaired, it has been difficult to access what has become the biggest platform connecting the world – until now. Thanks to Facebook’s accessibility team – whose sole purpose is to help people with disabilities have a seamless experience on the social network – another big leap has been made in Facebook’s journey of bringing people together.

“You just think about how much of your news feed is visual — and most of it probably is — and so often people will make a comment about a photo or they’ll say something about it when they post it, but they won’t really tell you what is in the photo,” said Matt King, Facebook’s first blind engineer, in an interview with TechCrunch. “So for somebody like myself, it can be really like, Ok, what’s going on here? What’s the discussion all about?

Until now, blind and visually impaired people who have access to screen readers could listen to what people are writing on Facebook, but they had no way to figure out what’s going on in the millions of photos shared on Facebook every day.

Yesterday, Facebook introduced Automated Alternative Text (AAT) that uses AI to identify the visual components that make up a photo, which can then be read out aloud. Before AAT, people using screen readers would only hear the name of the person who shared a photo, along with any accompanying text that the person wrote on Facebook. Now, someone could hear “image may contain two people, smiling, outdoors”, or “image may contain pizza, food”.

 

Image depicting Facebook's AAT at work

 

Zuckerberg made this announcement on his Facebook Page yesterday:

“We just launched a tool that uses artificial intelligence to help people who are blind experience photos on Facebook.

When blind or visually impaired people use the internet, they use something called a screen reader that turns text into spoken words. But until now, screen readers haven’t been able to describe what’s in a photo.

Using artificial intelligence, we’re able to understand what’s in a photo and describe that for someone using words. This is a great use of AI technology, and it’s an important step towards making sure everyone has equal access to information and is included in the conversation. Check out this video to see how it works.”

Facebook’s AAT is based on a neural network having billions of parameters, and one that is trained with millions of examples. Neural networks are a type of model for machine learning. For images,  a neural network can be though of as a pattern recognition system.

It recognizes images and words in transportation (“car”, “boat”, “motorcycle”), nature (“outdoor”, “mountain”, “wave”), sports (“tennis”,”swimming”), food (“ice cream”,”sushi”) and descriptive words for appearance (“baby”,”selfie”).

AAT is currently available for iOS screen readers set to English because that’s where Facebook sees the most use from blind and visually impaired people. It is coming soon to other platforms and languages.

How awesome do you think this new feature is? It is Facebook’s way of giving dignity to every person with a seeing disability in the world, and helping that connect with everyone else.

“Facebook’s mission is to connect the world. And that goes for everyone.”

 


Anisha Sawant