What latency feels like on Google’s Stadia cloud gaming platform

After peppering Google employees with questions regarding Stadia’s latency, pricing and supported devices to mostly no avail, I got my hands on one of their new controllers and pressed play on the Doom Eternal gameplay they were showing off on a big-screen TV.

Things started off pretty ugly. The frame rate dropped to a fast-paced PowerPoint presentation, the resolution dipped between 4K crispness and indecipherable blurriness and latency seemed to be as much as a half-second. As the Google employees looked nervously at each other, someone grabbed the controller from me and restarted the system.

After a system restart, things moved along much, much more smoothly. But what the situation sums up is that when it comes to game-streaming things can be unpredictable. To give Google credit, they stress-tested their system by running Stadia on hotel WiFi rather than taking me down to Mountain View and letting me play with Stadia under much more controlled conditions.

Stadia is Google’s cloud game-streaming service and while there’s a lot we don’t know the basic tenants are clear. It moves console-level gaming online into your Chrome browser and lets you access it from devices like smartphones that wouldn’t be able to handle the GPU-load initially.

Google’s Stadia game-streaming platform kills downloads and lets you play anywhere

Despite the initial hiccup, my experience with Stadia was largely positive. Doom Eternal was in crisp 4K and I was able to focus on the game without thinking about the service I was playing it on, which is ultimately the best endorsement of a new platform like this.

This will likely be a great service for more casual gamers but might not be the best fit for the most hardcore users playing multi-player titles. While you may be launching this service directly from YouTube feeds of eSports gamers, this is something they probably wouldn’t use. That’s because the latency between input and something being displayed onscreen isn’t imperceptible, though it’s probably good enough for the vast majority of users (myself included) which is still a worthy prize for the company’s efforts to take on the massive gaming market.

Google wouldn’t give me a proper range of where exactly latency fell, but they did say it was less than the time it took for a human to perceive something and react — which a Google employee then told me differed person-to-person but was generally 70ms-130ms — so I suppose the most official number we’ll get is that the latency is probably somewhere less than 70ms.

There is no hard truth here though because latency will really depend on your geographic proximity to the datacenter. Being in San Francisco, I connected to a data center roughly 50 miles away in San Jose. Google confirmed to me that not all rural users in supported countries will be able to sign-up for the service at launch because of this.

Other interesting things to note:

• Google said they’d confirm devices later, but when asked about iOS support at launch they highlighted that they were focused on Pixel devices at launch.
• It doesn’t sound like you’ll be able to restore purchases of games you’ve previously gotten, you’ll unsurprisingly have to buy all of your Stadia titles on the platform.
• You’ll be able to access games from YouTube streams, but there will also be an online hub for all your content and you can access games via links.
• The controller was nice and probably felt most similar to the design of Sony’s DualShock controller.

We’ll probably be hearing a lot more at Google I/O this summer, but with my first hands-on demo, the service certainly works and it certainly feels console-quality. The big freaking question is how Google prices this, because that pricing is going to determine whether it’s a service for casual gamers or hardcore gamers, and that will determine whether it’s a success.

Read Full Article

Reducing the Need for Labeled Data in Generative Adversarial Networks

Posted by Mario Lučić, Research Scientist and Marvin Ritter, Software Engineer, Google AI Zürich

Generative adversarial networks (GANs) are a powerful class of deep generative models.The main idea behind GANs is to train two neural networks: the generator, which learns how to synthesise data (such as an image), and the discriminator, which learns how to distinguish real data from the ones synthesised by the generator. This approach has been successfully used for high-fidelity natural image synthesis, improving learned image compression, data augmentation, and more.

Evolution of the generated samples as training progresses on ImageNet. The generator network is conditioned on the class (e.g., "great gray owl" or "golden retriever").

For natural image synthesis, state-of-the-art results are achieved by conditional GANs that, unlike unconditional GANs, use labels (e.g. car, dog, etc.) during training. While this makes the task easier and leads to significant improvements, this approach requires a large amount of labeled data that is rarely available in practice.

In "High-Fidelity Image Generation With Fewer Labels", we propose a new approach to reduce the amount of labeled data required to train state-of-the-art conditional GANs. When combined with recent advancements on large-scale GANs, we match the state-of-the-art in high-fidelity natural image synthesis using 10x fewer labels. Based on this research, we are also releasing a major update to the Compare GAN library, which contains all the components necessary to train and evaluate modern GANs.

Improvements via Semi-supervision and Self-supervision
In conditional GANs, both the generator and discriminator are typically conditioned on class labels. In this work, we propose to replace the hand-annotated ground truth labels with inferred ones. To infer high-quality labels for a large dataset of mostly unlabeled data, we take a two-step approach: First, we learn a feature representation using only the unlabeled portion of the dataset. To learn the feature representations we make use of self-supervision in the form of a recently introduced approach, in which the unlabeled images are randomly rotated and a deep convolutional neural network is tasked with predicting the rotation angle. The idea is that the models need to be able to recognize the main objects and their shapes in order to be successful on this task.

An unlabeled image is randomly rotated and the network is tasked with predicting the rotation angle. Successful models need to capture semantically meaningful image features which can then be used for other vision tasks.

We then consider the activation pattern of one of the intermediate layers of the trained network as the new feature representation of the input, and train a classifier to recognize the label of that input using the labeled portion of the original data set. As the network was pre-trained to extract semantically meaningful features from the data (on the rotation prediction task), training this classifier is more sample-efficient than training the entire network from scratch. Finally, we use this classifier to label the unlabeled data.

To further improve the model quality and training stability we encourage the discriminator network to learn meaningful feature representations which are not forgotten during training by means of an auxiliary loss we introduced previously. These two advancements, combined with large-scale training lead to state-of-the-art conditional GANs for the task of ImageNet synthesis as measured by the Fréchet Inception Distance.

Given a latent vector the generator network produces an image. In each row, linear interpolation between the latent codes of the leftmost and the rightmost image results in a semantic interpolation in the image space.

Compare GAN: A Library for Training and Evaluating GANs
Cutting-edge research on GANs is heavily dependent on a well-engineered and well-tested codebase, since even replicating prior results and techniques requires a significant effort. In order to foster open science and allow the research community benefit from recent advancements, we are releasing a major update of the Compare GAN library. The library includes loss functions, regularization and normalization schemes, neural architectures, and quantitative metrics commonly used in modern GANs, and now supports:

• Training on GPUs and TPUs.
• Lightweight configuration via Gin (examples).
• A plethora of data sets via the TensorFlow datasets library.

Conclusions and Future Work
Given the growing gap between labeled and unlabeled data sources, it is becoming increasingly important to be able to learn from only partially labeled data. We have shown that a simple yet powerful combination of self-supervision and semi-supervision can help to close this gap for GANs. We believe that self-supervision is a powerful idea that should be investigated for other generative modeling tasks.

Acknowledgments
Work conducted in collaboration with colleagues on the Google Brain team in Zürich, ETH Zürich and UCLA. We would like to thank our paper co-authors Michael Tschannen, Xiaohua Zhai, Olivier Bachem and Sylvain Gelly for their input and feedback. We would like to thank Alexander Kolesnikov, Lucas Beyer and Avital Oliver for helpful discussion on self-supervised learning and semi-supervised learning. We would like to thank Karol Kurach and Marcin Michalski for their major contributions to the Compare GAN library. We would also like to thank Andy Brock, Jeff Donahue and Karen Simonyan for their insights into training GANs on TPUs. The work described in this post also builds upon our work on “Self-Supervised Generative Adversarial Networks” with Ting Chen and Neil Houlsby.

Movius raises $45M for its business communications service

Atlanta-based Movius, a company that allows companies to assign a separate business number for voice calls and texting to any phone, today announced that it has raised a $45 million Series D round led by JPMorgan Chase, with participation from existing investors PointGuard Ventures, New Enterprise Associates and Anschutz Investment company. With this, the company has now raised a total of $100 million.

In addition to the new funding, Movius also today announced that it has brought on former Adobe and Sun executive John Loiacono as its new CEO. Loiacono was also the founding CEO of network analytics startup Jolata.

“The Movius opportunity is pervasive. Almost every company on planet Earth is mobilizing their workforce but are challenged to find a way to securely interact with their customers and constituents using all the preferred communication vehicles – be that voice, SMS or any other channel they use in their daily lives,” said Loiacono. “I’m thrilled because I’m joining a team that features highly passionate and proven innovators who are maniacally focused on delivering this very solution. I look forward to leading this next chapter of growth for the company.”

Sanjay Jain, the chief strategy officer at Hyperloop Transportation Technologies and Larry Feinsmith, the head of JP Morgan Chase’s Technology Innovation, Strategy & Partnerships office are joining the company’s board.

Movius currently counts more than 1,400 businesses as its customers and its carrier partners include Sprint, Telstra and Telefonica. What’s important to note is that Movius is more than a basic VoIP app on your phone. What the company promises is a carrier-grade network that allows businesses to assign a second number to their employee’s phones. That way, the employer remains in charge, even as employees bring their own devices to work.

Read Full Article

Apple announces new AirPods

Apple has just announced the second-generation AirPods.

The new AirPods are fitted with the H1 chip, which is meant to offer performance efficiencies, faster connect times between the pods and your devices, and the ability to ask for Siri hands-free with the “Hey Siri” command.

Because of its performance efficiency, the H1 chip also allows for the AirPods to offer 50 percent more talk time using the headphones. Switching between devices is 2x faster than the previous generation AirPods, according to Apple.

Here’s what Phil Schiller had to say in the press release:

AirPods delivered a magical wireless experience and have become one of the most beloved products we’ve ever made. They connect easily with all of your devices, and provide crystal clear sound and intuitive, innovative control of your music and audio. The world’s best wireless headphones just got even better with the new AirPods. They are powered by the new Apple-designed H1 chip which brings an extra hour of talk time, faster connections, hands-free ‘Hey Siri’ and the convenience of a new wireless battery case.

The second-gen AirPods are available with the standard wired charging case ($159), or a new Wireless Charging Case ($199). A standalone wireless charging case is also available for purchase to $79. We’ve reached out to Apple to ask if the wireless case is backwards compatible with first-gen AirPods and will update the post once we know more.

Update: Turns out, the wireless charging case works “with all generations of AirPods,” according to the Apple website listing.

The new Airpods are available to order today from Apple.com and the Apple Store app, with in-store availability beginning next week.

Read Full Article

The Best Podcasts on Spotify in 2019

Spotify is becoming an increasingly important player in the world of podcasting. The company’s acquisition of Gimlet Media only added to an already growing repertoire of podcasts.

Unlock the "Spotify Keyboard Shortcuts" cheat sheet now!

This will sign you up to our newsletter

Enter your Email

Read our privacy policy

As well as hosting the best podcasts available, Spotify boasts original podcasts and exclusive podcasts too. Which is what we focus on in this rundown of the best podcasts on Spotify in 2019.

1. Showstopper

Showstopper is a Spotify original podcast that explores the use of memorable music in TV shows.

The host is The Fader’s Editor-in-Chief, Naomi Zeichner. The episodes generally revolve around her interviewing artists, music supervisors, showrunners, and other members of the production staff who were involved in creating iconic music-meets-screen moments.

Episodes run for about 20 minutes, with a new one released every two weeks.

2. The Joe Budden Podcast with Rory and Mal

Former rap artist Joe Budden has hosted a music chat podcast since February 2015. Originally called “I’ll Name This Podcast Later,” the show has ballooned into a media juggernaut in the last couple of years.

The show isn’t short on controversy. There have been public feuds with Eminem, Complex, Offset, Chance The Rapper, Pusha T, and more.

In September 2018, Joe and his co-hosts signed an exclusive deal with Spotify. This means that Spotify is now the only place to grab the latest episodes as they drop twice a week.

3. Dissect

Dissect is another music-themed original podcast from Spotify Studios. The premise of Dissect is simple: Pick one album per season and discuss one track on the album per episode.

The first season looked at “To Pimp a Butterfly” by Kendrick Lamar. Season 2 discussed “My Beautiful Dark Twisted Fantasy” by Kanye West, and Season 3 was about Frank Ocean’s “Blonde”.

There’s also been an eight-episode mini-series about “The Miseducation of Lauryn Hill” to celebrate the album’s 20th anniversary.

4. Microphone Check

Microphone Check is one of the best podcasts on Spotify for exploring the boundaries between music and culture. Specifically, this podcast primarily focuses on the interactions between hip-hop and the outside world.

The podcast is chat-based. Co-hosts Frannie Kelley and Ali Shaheed Muhammad interview various guests each week. Kelley is a journalist; Muhammad is one of the founding members of A Tribe Called Quest.

The release schedule is a bit erratic. Sometimes there’s only a week between new episodes, while sometimes you will have to wait a full month.

5. Fest and Flauschig

Fest and Flauschig is a German-language podcast hosted by comedians Jan Böhmermann and Olli Schulz.

We appreciate that many of our readers don’t speak German, but Fest and Flauschig was the most-streamed exclusive Spotify podcast on the entire platform in 2018, so it’s worth mentioning.

If you do speak German, let us know what you think of the show in the comments at the end.

6. MotoGP Podcast

Motorsport fans should check out the new MotoGP Podcast. It is a joint venture between Spotify Studios and Dorna Sports.

The series began in time for the 2019 MotoGP championship. At this early stage, the podcast features an interview with the legendary racer Casey Stoner and a summing up of the pre-season testing in Sepang and Qatar.

7. Are and Be

Are and Be is an R&B Spotify podcast (“Are & Be,” get it?!). The first season was published under the name R&B Matters in 2017. The second series in 2018 saw the name change. The third series will be out later in 2019.

The podcast is supposed to act as a bridge between the people who curate Spotify’s Are & Be playlist and the listening audience; the podcast is managed by Spotify’s Global Head of R&B and Soul.

Today, the Are and Be podcast is a platform for artists on the playlist to share their experiences and stories. Each episode lasts about half an hour.

8. David’s Out For A Good Time

David’s Out For A Good Time is a Spotify original podcast that’s aimed at the LGBTQ community.

Hosted by David Olshanetsky—the creator of TheShitneySpears.com—the show discusses a range of topics, covering everything from LGBTQ issues to what it was like growing up in the age of the internet. There’s also a healthy dose of pop culture in every episode.

A new episode is available most Tuesdays, though there are occasional short breaks of a couple of months.

9. Under Cover

Under Cover centers on artists who have released cover versions of popular songs.

The podcast delves into the cover songs, as the artist responsible shares the backstory behind the track, along with what inspired them to record the cover and why they chose the musical approach they did.

Features artists have included Shawn Mendes talking about his cover of “Use Somebody,” Meghan Trainor discussing her recording of Sasha Sloan’s “Normal,” and Def Leppard revealing why they recorded Depeche Mode’s “Personal Jesus”.

10. Amy Schumer Presents: 3 Girls, 1 Keith

Amy Schumer is an American stand-up comedian. In June 2018, she launched a new Spotify original podcast called Amy Schumer Presents: 3 Girls, 1 Keith.

Along with Amy, the other three hosts are Rachel Feinstein, Bridget Everett, and Keith Robinson—all three of whom are fellow comedians. The chat show sees them discuss everything from sex and love to politics and the world’s comedy scene.

All four presenters aren’t shy to drop the curtain and reveal highly personal information about themselves, all of which makes for a highly enjoyable show.

There have been two series so far, with the third imminent.

Alternative Ways to Listen to Podcasts

Spotify is up against some stiff competition when it comes to podcast players. Sure, there are some solid reasons to use Spotify (fewer apps on your device, a centralized audio experience, and a simple way to cast your audio to other speakers in your home), but some of the alternatives are unquestionably more feature-rich.

If you would like to learn more, check out our articles on the best podcast players for Android and the best podcast players for iOS.

Read the full article: The Best Podcasts on Spotify in 2019

Read Full Article

What Deactivating or Deleting Facebook Really Means for Privacy

You’ve heard a lot about privacy concerns on Facebook. You’ve decided enough is enough. You’re going to deactivate your account.

But what does that actually mean to your privacy? Is your personal data completely wiped? And is it better to deactivate or delete your profile?

Facebook Deactivation vs. Facebook Deletion

We need to differentiate between deactivation and deletion. You might’ve seen them used interchangeably, but they’re not the same.

Deactivating your Facebook account is a temporary measure. Deletion is more permanent. We’ll cover both here as they have different implications for your personal privacy. Generally, however, deactivation means there’s still personal information about you online. Deleting Facebook doesn’t mean all your details will be erased, though.

Does Deactivation Erase All Your Personal Data?

No.

Facebook deactivation puts your data in stasis. If you want to return to Facebook, your profile still exists. Your timeline becomes invisible, but your personal information will still be held by the company. That’s because Facebook wants you back. Everything is restored if you sign back in.

All you’re doing when you deactivate Facebook is stopping your profile from being publicly visible. It’s a good move if all you’re concerned about is your peers (and potential strangers searching for you) knowing too much about you.

Your comments on other people’s posts won’t disappear. However, your profile picture will be replaced by a default Facebook image (i.e. the gray silhouette) and no one will be able to click on your name to get to your timeline.

So what happens to Messenger? When you deactivate Facebook, it’ll ask whether you want to stop using the instant messaging service too. Yes, you can use Messenger without having a Facebook account. Your messages won’t vanish, even if you deactivate. In fact, they won’t vanish if you delete Facebook either: instead, they’ll simply read “Facebook User”.

Does Deleting Facebook Erase All Your Personal Data?

Deactivation won’t erase everything—so what can you do? Deleting Facebook sounds like a solid option… except some data persists.

It takes 30 days after you request account deletion for your profile to entirely vanish. This period gives you leeway in case you decide you don’t want to permanently leave Facebook. However, it can take 90 days for the company to delete your data; no other parties can access your data in the time it takes to get rid of your account.

Your timeline will no longer exist. You’ll automatically be taken off any pages you “like”. You won’t be a member of any groups. Your online persona will effectively disappear.

However, that doesn’t mean everything will be deleted. Some information will be kept, albeit stripped of personal identifiers. You become a demographic. Facebook and its third-parties use those statistics.

Your comments aren’t deleted either. Think of them as flotsam and jetsam. Let’s say you commented on a MakeUseOf Facebook post a few years ago—that will still exist.

Crucially, your log data (a record of all you do, including times you log in and groups you frequent) remains. Your name isn’t attached to that data though.

Essentially, this is the same thing that happens if you deactivate your account. Your profile picture is replaced by the default icon. Normally, if you click on someone’s name in the comments section, you’ll be redirected to their timeline. The link vanishes once you delete Facebook because there’s nothing for it to lead to.

What Happens to Tags in Facebook Photos?

It doesn’t matter if you deactivate or delete Facebook; the same thing happens to any photos you’re tagged in. Deleting your account doesn’t affect other people’s.

how to avoid being tagged with embarrassing photos on facebook on your birthday?

deactivate. pic.twitter.com/lIvp3FAP4t

— h0n3y (@abywanders) March 7, 2019

Tags will revert to Plain Text. The link to your profile will expire. Obviously, if you reactivate your account, these links (including comments) will become active again.

What does this mean for your privacy? Based on tagged photos, Facebook can still recognize you. How? Using facial recognition software called DeepFace. This tool stores your face as a neural network by noting distinct features and relative measurements. Your face becomes a data composite, which Facebook uses to suggest matches in images.

Once you’re tagged in photos, Facebook knows what you look like. Because those tags aren’t deleted, your data set will be kept.

Fortunately, such tags aren’t always right. Scroll through your tagged pictures and you’ll see an array of inaccuracies. Friends might have tagged you in something just to make sure you see it. Or they’ll have added the tag anywhere, muddying the water.

And that’s fine because Facebook does this automatically. They just use smart algorithms to suggest tags; no one will actually be looking for you individually.

If a friend’s account (or your own, if you’ve only deactivated Facebook) is hacked, however, a cybercriminal could, in theory, use their intuition to work out exactly what you look like. But why? Personally Identifiable Information (PII)—like your name, address, and date of birth—can be useful to hackers, but recognizing your face won’t do them much good.

No, you won’t be untagged from photos, but generally, there’s very little cause for concern.

What Happens to Third-Party Facebook Logins?

Many services now let you sign in using a Facebook account. These include popular apps like Spotify, Etsy, and LinkedIn, so what happens when you delete or deactivate Facebook? You’ll automatically sign out of these services, and further steps will hinge on individual apps.

We advise you do your research before deleting permanently.

I quit the Facebook yesterday. When looking for instructions to delete (it’s 5 screens deep) they first display scary syntax like the below. That’s exactly why I’m backing away, @facebook. I want my own friendships and social life, not impacted by you. pic.twitter.com/afcJ6SEpnh

— Allison (@Dustytext) March 7, 2019

It’s worth contacting the third-parties you intend to continue using in the future. In many cases, you’ll have to create new accounts.

Can they still access your data after deleting your profile? Facebook asserts that, once you begin the deletion process, no other services can see your personal information. That’s definitely a positive.

But third parties aren’t obliged to get rid of their data. It just means they can’t get more private information from your social media page. Deleting Facebook certainly doesn’t equate to deleting other accounts.

You might wonder, then, what details they get from Facebook. They can at least access what’s on your public profile, meaning your name, profile picture, and anything else you’ve not hidden when managing your Facebook privacy settings.

It also depends on the third-party. They harvest information most pertinent to their business. Trip Advisor, for instance, can present adverts to you based on where your friends have recently traveled to or reviewed. Tinder judges your interests based on “likes”, so you might not get the most accurate matches without connecting to Facebook.

How Else Can Facebook Spy on You?

Want to escape Facebook? The firm now owns a few services which gobble up data. You’ll need to cleanse your smartphone if you’re serious about your privacy.

Many of Facebook’s acquisitions have been conglomerated into the social network itself, like Face.com and Bloomsbury AI. You don’t need to worry about these if you delete Facebook.

But the company’s other purchases can carry on collecting data about you, with that information ending back at Facebook Inc. Some are obvious: Facebook Messenger carries its parent company in its name. If you don’t want Facebook knowing much about you, limit what personal details you share there.

Then there’s Instagram, which seems like a great way to share your photography portfolio and chat with friends. Don’t be fooled: Instagram is spying on you.

Oculus VR, also owned by Facebook, reserves the right to share information with Facebook and third-party advertisers. Yet Oculus maintains that it doesn’t share with advertisers, and any other private data is shared in a limited capacity. Its Privacy Policy allows Oculus to share private information with its owner at any stage in the future, however.

That’s without mentioning Facebook’s costliest ever acquisition, WhatsApp!

If you really want to escape the Facebook umbrella, you’ll likely need to uninstall a few useful apps. It’s up to you to decide whether it’s really worth it.

Don’t Have Facebook? You’re Still Not Safe!

You don’t need a Facebook account for the company to create a shadow profile of you.

Check out Facebook’s Privacy Policy, which states they “receive and analyze content, communications, and information” if someone should “upload, sync or import your contact information”. If you’re in someone’s address book and they allow Facebook access to it, you already have a shadow profile.

This won’t be as extensive as an actual Facebook user.

This isn’t your sole concern. The company got in trouble for tracking and transmitting cookies collected by the social plug-in used to “like” pages across millions of websites.

Deleting Facebook restricts what data is collected, yet the network extends beyond its own site.

Do You “Like” Your Own Privacy?

Leaving the platform entirely is a big step. If you’re not sure about deleting Facebook, try deactivation first.

If you deactivate, your data isn’t publicly available, but Facebook still stores it. Most of your details will be wiped within 90 days if you delete your account, though traces will remain. So if you’ve made up your mind, here’s how to delete or deactivate Facebook.

Read the full article: What Deactivating or Deleting Facebook Really Means for Privacy

Read Full Article

How to Create Amazing Webcam Effects Using Java and Processing

Processing is a powerful tool which allows the creation of art through code. It is the combination of a Java library for working with graphics, and an integrated development environment (IDE) which allows you to write and run code easily.

There are many graphics and animation beginner projects which use Processing, but it is also capable of manipulating live video.

Today you’ll be making a live video slideshow of different effects controlled by the mouse, using the Processing video library. As well as flipping the live video, you’ll learn to resize and color it, and how to make it follow the mouse cursor.

Project Setup

To begin, download Processing and open a blank sketch. This tutorial is based on a Windows system, but it should work on any computer with a webcam.

You may need to install the Processing Video library, accessible under Sketch > Import Library > Add Library. Search for Video in the search box, and install the library from The Processing Foundation.

Once installed, you are ready to go. If you want to skip the coding, you can download the complete sketch. It is much better to make it yourself from scratch, however!

Using a Webcam With Processing

Lets begin by importing the library, and creating a setup function. Enter the following into the blank Processing sketch:

import processing.video.*;

Capture cam;

void setup(){
  size(640,480);
  cam = new Capture(this, 640, 480);
  cam.start();
}

After importing the video library, you create a Capture instance called cam to store the data from the webcam. In setup, the size function sets up a 640×480 pixel sized window to work in.

The next line assigns cam to a new instance of Capture, for this sketch, which is the same size as the window, before telling the camera to turn on with cam.start().

Don’t worry if you don’t understand every part of this for now. In short, we have told Processing to make a window, find our camera, and turn it on! To display it we need a draw function. Enter this below the code above, outside of the curly brackets.

void draw(){
  if (cam.available()){
    cam.read();
  }
  image(cam,0,0); 
}

The draw function gets called every frame. This means that many times each second, if the camera has data available you read the data from it.

This data is then displayed as an image, at the position 0, 0, which is the top left of the window.

Save your sketch, and press the play button at the top of the screen.

Success! The data stored by cam is correctly being printed to the screen every frame. If you are having problems, check your code thoroughly. Java needs every bracket and semi-colon in the right place! Processing can also require a few seconds to access the webcam, so if you think it isn’t working wait a few seconds after launching the script.

Flipping the Picture

Now that you’ve got a live webcam image let’s manipulate it. In the draw function, replace image(cam,0,0); with these two lines of code.

scale(-1,1);
image(cam,-width,0);

Save and rerun the sketch. Can you see the difference? By using a negative scale value, all of the x values (the horizontal pixels) are now reversed. Because of this, we need to use the negative value of the window width to position the image correctly.

Flipping the image upside down requires just a couple of small changes.

scale(-1,-1);
image(cam,-width,-height);

This time, both the x and y values are flipped, turning the live camera image upside down. So far you’ve coded a normal image, a horizontally flipped image, and a vertically flipped image. Let’s set up a way to cycle between them.

Making It Cycle

Instead of rewriting your code every time, we can use numbers to cycle through them. Create a new integer at the top of your code called switcher.

import processing.video.*;

int switcher = 0;
Capture cam;

We can use the value of switcher to determine what happens to the camera image. When the sketch starts, you give it a value of 0. Now we can use logic to change what happens to the image. Update your draw method to look like this:

void draw(){
  if (cam.available()){
    cam.read();
  }

  if(switcher==0){
    image(cam,0,0);
  }
  else if(switcher == 1){
    scale(-1,1);
    image(cam,-width,0); 
  }
  else if(switcher == 2){
    scale(-1,-1);
    image(cam,-width,-height);
  }
  else{
    println("Switcher = 0 again");
    switcher = 0;
  }
}

Now, all three variations of the code will trigger depending on the value of switcher. If it doesn’t match one of our if or if else statements, the else clause gets reset to 0. Logic is an important beginner skill to learn, and you can find out about them and a lot more with an excellent YouTube Programming Tutorial!

Using the Mouse

Processing has built-in methods for accessing the mouse. To detect when the user clicks the mouse, add the mousePressed function at the bottom of your script.

void mousePressed(){
  switcher++;
}

Processing listens for any mouse clicks and interrupts the program to carry out this method when it detects one. Every time the method gets called, the value of switcher gets bigger by one. Save and run your script.

Now, when you press the mouse button, it cycles through the different orientations of videos, before returning to the original. So far you have just flipped the video, now let’s do something a little more interesting.

Adding More Effects

Now, you will code a four-color live image effect similar to the famous Andy Warhol works of art. Adding more effects is as simple as adding another clause to the logic. Add this to your script between the last else if statement, and else.

else if(switcher == 3){
  tint(256, 0, 0);
  image(cam, 0, 0, width/2, height/2);

  tint(0, 256, 0);
  image(cam, width/2, 0, width/2, height/2);

  tint(0, 0, 256);
  image(cam, 0, height/2, width/2, height/2);

  tint(256, 0, 256);
  image(cam, width/2, height/2, width/2, height/2);
}

This code uses the image function to create four separate camera images in each corner of the screen and to make them all half sized.

The tint function adds color to each camera image. The numbers the brackets are red, green, and blue (RGB) values. Tint colors all the following code with the chosen color.

Save and play to see the result. Try changing the RGB numbers in each tint function to change the colors!

Making It Follow the Mouse

Finally, let’s make the live image follow the mouse position using helpful functions from the Processing library. Add this above the else part of your logic.

else if(switcher==4 ){
  image(cam, mouseX, mouseY, width/2, height/2);
}

Here, you are positioning the image from your camera at mouseX and mouseY. These are built in Processing values which return which pixel the mouse is pointing at.

That’s it! Five variations of live video through code. However, when you run the code, you’ll notice a couple of problems.

Finishing Up the Code

The code you’ve created so far works, but you’ll notice two issues. Firstly, once the four-color variation shows, everything afterward is tinted purple. Secondly, when you move the video with the mouse, it leaves a trail. You can fix it by adding a couple of lines to the top of the draw function.

void draw(){
  tint(256,256,256);
  background(0);

  //draw function continues normally here!

At the start of each frame this code resets the tint color to white, and adds a background color of black to stop the video leaving trails. Now when you test the program, everything works perfectly!

Webcame Effects: Art From Code

Processing is very powerful, and you can use it to do many things. It’s an excellent platform for making art with code, but it is equally suited to controlling robots!

If Java isn’t your thing, there is a JavaScript library based on Processing called p5.js. It’s browser-based, and even beginners can use it to create fantastic reactive animations!

Image Credit: Syda_Productions/Depositphotos

Read the full article: How to Create Amazing Webcam Effects Using Java and Processing

Read Full Article

Master the New Mobile Gmail With These 10 Tips

The Best Practices for Instagram Affiliates

Affiliate marketing has long been dominated by bloggers turning clicks into cash. Today, Instagram is the real money-maker in the game, with influencers functioning as a friendly bridge between brands and customers.

To generate results, influencers must strike a balance between promotional content and authentic recommendations, tips, and tricks. Below, we’ve compiled a list of the best practices for Instagram affiliates.

1. Affiliate Posts Must Fit Your Feed

With Instagram affiliate marketing, both the influencer and the brand should get value out of the relationship. As such, it’s important to remember that success comes from establishing trust. This means that you’ll need to introduce affiliate products in a way that feels genuine.

Here are some examples of how you can ensure that your posts don’t come off as overtly “promotional”.

Make Sure You Believe in the Product

It’s hard to promote something you’ve never used before and still make a convincing case for making the purchase. As such, it’s important to be selective about the products you choose to promote.

If you partner with clothing brands, take pictures that show how it works with your existing wardrobe. Beauty and makeup stuff? Try offering a short tutorial in your next Instagram Story.

Show How the Product Has Made Your Life Better

To convince people that buying this product is worthwhile, you’ll want to show off its value in your feed. Essentially, this means you’re promoting products in a way that sounds like you’re just helpful. Would you tell your friends or family about this product, if not, maybe sit this one out.

This example from Glad to be in First does a good job promoting this cookbook. She’s enthusiastic (see the avocado toast mention), so it sounds like a genuine endorsement. Still, she points toward the affiliate link and shouts out the brand.

Don’t Spam Your Followers

Your followers shouldn’t feel pressured into buying products. Post after branded post will take a toll on your follower count. Patience pays off here. Your best bet is interweaving affiliate content into your feed with unbranded posts. Followers are quick to drop their influencer allegiance when they feel that they’ve “sold out”.

2. Use Paid Promotions Sparingly

Sponsored posts are only available to those users who have Instagram for business accounts. If you haven’t already upgraded to business class, head over to Instagram and make the change.

Sponsored posts allow you to embed a clickable link inside your posts and promote your posts to the exact audience you’d like to reach. It’s a powerful way to grow your following and increase sales.

But you shouldn’t pay to boost your content all of the time, or else you’re just giving Instagram your commissions. Make sure you set a budget that you’re comfortable with and track your results.

For more ways to grow your audience, read our guide to help you gain more followers on Instagram.

3. Make the Most of the One Link in Your Bio

Instagram only allows one live link at a time. Which can get in the way of promoting every product to your audience at once.

When you market the product in an individual post, don’t forget to point followers back to the link in your bio. Because there’s an extra step on Instagram, as opposed to a blog where you can embed personalized links anywhere, you’ll need to make sure you prompt your audience.

Use Instagram Stories to Share Links

Instagram Stories are a great way to promote affiliate links. You’ll need a business account to make the most out of this feature, otherwise you can’t add live links to your Stories.

If you have a business account, you can apply to add shoppable tags to your stories and link to products that way. Business accounts do not automatically come with the shopping feature. We recommend checking out Instagram’s Help page for more on who qualifies as a merchant.

If you don’t have shopping permissions, you can use the “swipe up” trick instead. Smalltalk Social created this Stories campaign in Canva. Shoppers can swipe up to buy the books on Amazon and Smalltalk receives the commission.

Use Link-in-Bio Apps

Later’s Linkin.bio is one example. It allows you to add links to individual Instagram posts scheduled through their tool. The idea is to turn your page into an optimized landing page.

There’s also LinkTree, which supports up to five links at a time. According to their marketing page, one bio link can support multiple destinations. Another example is lnk.bio, which offers a free entry-level plan for new influencers who want to try out the tool before committing.

Link-in-bio apps are typically a paid service, but they do offer an efficient workaround for influencers promoting more than one product at a time.

These apps track which posts users are clicking on and will present a custom link each time. This means you’re not constantly changing the link to match your latest campaign, thus not losing out on commissions.

Cross-Promote Affiliate Links on Other Channels

Don’t limit yourself to Instagram alone. Instead, leverage your other networks like Twitter and Facebook to share the same links. The idea isn’t to cut and paste the same content across all channels.

Instead, tailor each post to according to the platform you’re posting on and link to your original Instagram post. Take, for example, the tweet below. It highlights an Instagram giveaway and points followers toward a specific action.

Another idea is running a blog. The blog should fall within the same niche as your Instagram page, offering another platform for connecting with your audience and promoting your product links.

Focusing on creating good content will help you rank on Google Search, which, in turn, can help you grow your Instagram following. This is only a viable option if you have the time to write regular content and you’re a competent writer. If that’s not the case, stick with the socials.

4. Use Hashtags the Right Way

Hashtags might seem simple, but they’re actually the best way to increase your reach organically. Like Twitter, Instagram hashtags work to organize video and photo content.

Relevant, targeted hashtags allow you to get discovered by new audiences, thus adding up to more followers, more engagement, and more referrals.

As you can see below, a short caption, followed by several hashtags that relate to vegan foods, healthy eating, and smoothies, can attract a wide range of foodies.

You’ll want to use a combination of hashtags, some which boast a high posting volume and others that boast comparatively less. It’s easier to rank for less popular hashtags.

Start your research process by looking at the hashtags you’re following. What are other people in your niche posting? What about your competitors?

From there, type relevant examples into an app like Leetags or Hashtagify.me. Both options will help you narrow down some good options for getting the word out. Instagram allows you to use up to 30 hashtags, but the optimal number is closer to 10.

Stay away from generic tags like #blogging or #socialmedia. These unimaginative phrases won’t do anything but attract bots or people outside of your target audience.

5. Disclose Everything (and We Mean Everything)

Transparency is everything. And in the affiliate game, transparency is required by law. It’s important to note that if you are posting something for financial gain on social media, you need to disclose this partnership to your audience.

The FTC guidelines state that branded hashtags and #sponcon should be “easily noticed”.

The Best Affiliates Focus on Being Themselves

Affiliate influencers of all sizes help brands connect to their customers by being transparent and helpful. Yes, you’re selling from your account, but your audience needs to believe that you care about the products you promote.

The most important consideration is selecting a program that offers the products that resonate with your followers and from there, creating compelling content that gently promotes your links. Skip the hard sell and embrace authenticity instead.

Finally, affiliate marketing isn’t the only way to make money on Instagram. You can also sell products, services, or even pictures.

Read the full article: The Best Practices for Instagram Affiliates

Read Full Article

How to Fix Ghost Touch on iPhone: 9 Potential Fixes to Try

Keyboard Shortcuts You’re Getting Wrong and YouTube Channels You Must See

If you don’t know about a keyboard shortcut, you can end up feeling pretty silly when you accidentally hit one. Something crazy is likely to happen, from the keyboard beeping to your display rotating.

In this week’s Really Useful Podcast, Ben Stegner joins Christian Cawley to discuss accidental keyboard shortcuts. He also has a selection of interesting YouTube channels covering product unboxing, video game music, cookery, and chill out music and video.

We take a look at the latest news, covering Twitter’s upgraded camera features and the new Windows 10 feature of rolling back a buggy update. And Christian Cawley reveals the results of the week he spend using a Raspberry Pi as his primary computer…

Really Useful Podcast Season 2 Episode 7 Shownotes

This week, we talked about:

• Twitter updates camera feature
• Windows 10 will automatically rollback buggy updates
• Is Kodi legal?
• What happened when Christian used a Raspberry Pi as a PC for a week?
• YouTube channels:
  • Chill out
  • Unique cookery channels 
  • Product unboxing
  • Video game music on YouTube
• Christian’s first YouTube upload

Hosts for S2E7 of the Really Useful Podcast are Christian Cawley and Ben Stegner, who you will find on Twitter as:

• @thegadgetmonkey
• @stegnersaurus

Know anyone who would benefit from having tech topics broken down into simplified language? Share our podcast with them, or suggest they to subscribe.

Enjoy the show? Subscribe to the Really Useful Podcast on:

• iTunes
• Spotify
• Player.FM
• Google Podcasts
• Stitcher.com
• YouTube

Catch the tech podcast for technophobes again next week!

Read the full article: Keyboard Shortcuts You’re Getting Wrong and YouTube Channels You Must See

Read Full Article

Seven Africa-focused startups present at Y Combinator’s Demo Day

The seven African-focused companies which presented as part of Y Combinator’s 200 strong cohort of Winter 2019 class of 200 startups may seem like a small percentage for such a large class, but it represents the growing significance of African ventures in YC’s universe.

Since 2016, the Silicon Valley accelerator—that provides seed funds and mentorship for early stage startups—has backed 25 companies located in Africa and another 10 with an Africa product focus, according to YC spokesperson Lindsay Amos.

Past YC Africa alumns covered here at TechCrunch include payments startup Paystack, logistics firm Kobo360, and VOD startup Afrostream (now shuttered).

Of the 7 Africa-oriented YC class who presented at demo day 2019, 5 originated in Nigeria and 1 in Tanzania. All 7 are fintech ventures with products targeted across currency trading, agriculture, healthcare, and education.

Here’s the skinny on the Africa focused startups that presented at Demo Day 2019.

Nala (Tanzania): A fintech service building internet-free mobile payments for Africa. Nala lets users pay bills, do p2p payments, and purchase cell phone minutes without an internet connection. The startup’s app is one of Google PlayStore’s Top Free Finance Apps in Tanzania, according to founder Benjamin Fernandes. Nala plans to leverage the seed funding and exposure from YC participation to raise additional capital to close its first seed round, he told TechCrunch.

54gene (Nigeria): 54gene aims to be the 23andMe for Africa. The company says that competitor data is limited because their users are mostly white. By focusing on Africa, the company can help detect and identify DNA markers that might otherwise go overlooked. Co-founder Abasi Ene-Obong has a PhD in cancer biology from the University of London.

VertoFX (UK): VertoFX is a B2B currency marketplace and international payment provider for frontier markets with an initial target market of Africa. The platform allows businesses to gain easy access to foreign currencies and make international payments seamlessly, providing clients with real-time and cheaper exchange rates, liquidity, and faster settlement periods. VertoFX’s founders are ex-bankers Anthony Oduwole and Ola Oyetayo. In an email to TechCrunch Oduu said there are currently 18 currencies on the platform and the ability to settle in 120 countries, including China and the U.S. 

Thrive Agric (Nigeria): The company helps smallholder farmers in Nigeria access crowdfunded loans to help grow their crops, as well as help them sell their produce. The company has worked with 14,000 farmers to date, with plans to reach 1,000,000 farmers across Africa by 2022 in what it says is a $54 billion market opportunity.

CredPal (Nigeria): A credit card company that allows consumers in Africa to make purchases across online and offline merchants and pay in fixed monthly installments. CredPal touts it cards as having the convenience of a credit card and the simplicity of fixed monthly repayments. The purpose is reducing the financial pressure faced by consumers in Africa and also improving their standard of living. The scope is Africa and it is currently operational in Nigeria. Founders are Fehintolu Olaogun and Olorunfemi Jegede.

Schoolable (Nigeria): Provides simple finance solutions for schools in Africa. There are 65 million students in Africa attending private schools, but tuition payments can be a pain point for parents paying for their child’s education. That’s because tuition payments are often due upfront and it’s more difficult than it should be for schools to keep track of payments. Schoolable is creating an invoicing app that helps ensure parents make payments on time, while also using the app to save directly for tuition.

Wallet.NG (Nigeria): A fintech app that allows people to send money to friends or family and pay bills with only the recipient’s phone number. Wallet helps users keep track of their spending through financial reporting available on the app.

 

Read Full Article

Google fined $1.49BN in Europe for antitrust violations in search ad brokering

The European Commission has just announced another antitrust fine for Google.

The latest fine — $1.49BN — relates to its search ad brokering business which competition commissioner Margrethe Vestager noted today is “by far” the company’s main source of revenue.

“Today’s decision is about how Google abused its domiance to stop website’s using brokers other than the AdSense platform,” she said, noting that the Commission looked at more than 200 agreements and found at least one clause that harmed competition.

“There was no reason for Google to include these restrictive clauses in its contracts other than to keep rivals out of the market,” she added.

The Commission found three types of anti-competitive restriction in Google’s contracts — including the company requiring its ads to have premium placement.

We’ve reached out to Google for comment.

This story is developing… refresh for updates

It’s the third Commission antitrust penalty for Google, following the $5BN fine for anti-competitive behaviors attached to Android last summer and a $2.7BN penalty for Google Shopping antitrust violations in mid 2017.

In recent years Vestager has also flagged concerns about several other Google products, including travel search, image search and maps. Though no formal probes have yet been announced.

The latest EU antitrust decision against Google relates specifically to Google AdSense ads that appear on third party sites as a result of a search made on those sites.

The Commission made a formal Statement of Objections against AdSense in 2016, when it identified several practices it believed violated antitrust rules after investigating complaints.

Its objections included that Google required exclusivity in AdSense site search deals, mandating that third parties do not source search ads from its competitors; that Google required third parties to take a minimum number of search ads from it, with premium placement for them; and that Google required sites to seek approval from it before making any changes that might involve competing ads.

The Commission noted at the time that exclusivity practices had been in place since 2006, before being gradually replaced by Google from 2009 in most contracts — with the requirement of premium placement/minimum ads and the right for Google to authorise competing ads.

“The Commission is concerned that the practices have artificially reduced choice and stifled innovation in the market throughout the period,” it wrote then. “They have artificially reduced the opportunities for Google’s competitors on this commercially important market, and therefore the ability of third party websites to invest in providing consumers with choice and innovative services.”

You can read more on the background to the AdSense complaint (and Google Shopping) in our report from 2016 here.

Since being hit with a wave of antitrust scrutiny and action in Europe Google has been forced to tweak product and make some changes to its business practices. Though it’s far from clear it’s done enough to stave off further regulatory intervention if the end-goal is to promote effective competition.

Google’s latest Android tweaks — announced just yesterday — attempt to address complaints that its default browser and search settings on the smartphone platform box out competition by preferring its own browser over rival browsers.

In a blog post dated March 19 Google said it will start asking Android users to choose which browser and search apps they wish to use, with the change slated as coming in the next months.

Last fall it also made changes to the licensing terms of Android to allow device makers to unbundle its apps — for a fee.

While in fall 2017 Google tweaked how it displays search results for products in Europe.

Although it was later accused of creating a ‘fake’ price comparison site scheme to create the allusion of a thriving market.

In all these EU antitrust cases complaints against Google have not gone away.

Rather complainants continue to couch the company’s self-styled compliance ‘remedies’ as a joke.

Yet with antitrust and political scrutiny of Google and other tech giants stepping up domestically as well as internationally, any self-serving competition ‘fixes’ are operating on borrowed time.

Moreover figleafs will likely increase pressure for more radical regulatory intervention — such as a break up of Google…

Elizabeth Warren wants to break up Google, Amazon and Facebook

It’s therefore likely no coincidence that — in a more recent browser-related update — the search giant quietly expanded search engine choices in Chrome, adding privacy-focused rival DuckDuckGo’s search engine to the lists of options it promotes to users in more than 60 markets with the latest Chromium stable release, as well as French pro-privacy Qwant as an option in its home market.

This change was welcomed by both DuckDuckGo and Qwant.

Though the latter told us it still recommends its users use Firefox or Brave, rather than Google’s Chrome browser.

Read Full Article

Markforged raises $82 million for its industrial 3D printers

3D printer manufacturer Markforged has raised another round of funding. Summit Partners is leading the $82 million Series D round with Matrix Partners, Microsoft’s Venture Arm, Next47 and Porsche SE also participating.

When you think about 3D printers, chances are you’re thinking about microwave-sized, plastic-focused 3D printers for hobbyists. Markforged is basically at the other end of the spectrum, focused on expensive 3D printers for industrial use cases.

In addition to increased precision, Markforged can manufacture parts in strong materials, such as carbon fiber, kevlar or stainless steel. And it can greatly impacts your manufacturing process.

For instance, you can prototype your next products with a Markforged printer. Instead of getting sample parts from third-party companies, you can manufacture your parts in house. If you’re not going to sell hundreds of thousands of products, you could even consider using Markforged to produce parts for your commercial products.

If you work in an industry that requires a ton of different parts but don’t need a lot of inventory, you could also consider using a 3D printer to manufacture parts whenever you need them.

Markforged has a full-stack approach and controls everything from the 3D printer, software and materials. Once you’re done designing your CAD 3D model, you can send it to your fleet of printers. The company’s application also lets you manage different versions of the same part and collaborate with other people.

According to the company’s website, Markforged has attracted 4,000 customers, such as Canon, Microsoft, Google, Amazon, General Motors, Volkswagen and Adidas. The company has shipped 2,500 printers in 2018.

With today’s funding round, the company plans to do more of the same — you can expect mass production printers and more materials in the future. Eventually, Markforged wants to make it cheaper to manufacture parts at scale instead of producing those parts through other means.

Read Full Article

Google Trends Maps

Google Trends Maps

Ahead of third antitrust ruling, Google announces fresh tweaks to Android in Europe

Google is widely expected to be handed a third antitrust fine in Europe this week, with reports suggesting the European Commission’s decision in its long-running investigation of AdSense could land later today.

Right on cue the search giant has PRed another Android product tweak — which it bills as “supporting choice and competition in Europe”.

In the coming months Google says it will start prompting users of existing and new Android devices in Europe to ask which browser and search apps they would like to use.

This follows licensing changes for Android in Europe which Google announced last fall, following the Commission’s $5BN antitrust fine for anti-competitive behavior related to how it operates the dominant smartphone OS.

tl;dr competition regulation can shift policy and product.

Albeit, the devil will be in the detail of Google’s self-imposed ‘remedy’ for Android browser and search apps.

Which means how exactly the user is prompted will be key — given tech giants are well-versed in the manipulative arts of dark pattern design, enabling them to create ‘consent’ flows that deliver their desired outcome.

A ‘choice’ designed in such a way — based on wording, button/text size and color, timing of prompt and so on — to promote Google’s preferred browser and search app choice by subtly encouraging Android users to stick with its default apps may not actually end up being much of a ‘choice’.

According to Reuters the prompt will surface to Android users via the Play Store. (Though the version of Google’s blog post we read did not include that detail.)

Using the Play Store for the prompt would require an Android device to have Google’s app store pre-loaded — and licensing tweaks made to the OS in Europe last year were supposedly intended to enable OEMs to choose to unbundle Google apps from Android forks. Ergo making only the Play Store the route for enabling choice would be rather contradictory. (As well as spotlighting Google’s continued grip on Android.)

Add to that Google has the advantage of massive brand dominance here, thanks to its kingpin position in search, browsers and smartphone platforms.

So again the consumer decision is weighted in its favor. Or, to put it another way: ‘This is Google; it can afford to offer a ‘choice’.’

In its blog post getting out ahead of the Commission’s looming AdSense ruling, Google’s SVP of global affairs, Kent Walker, writes that the company has been “listening carefully to the feedback we’re getting” vis-a-vis competition.

Though the search giant is actually appealing both antitrust decisions. (The other being a $2.7BN fine it got slapped with two years ago for promoting its own shopping comparison service and demoting rivals’.)

“After the Commission’s July 2018 decision, we changed the licensing model for the Google apps we build for use on Android phones, creating new, separate licenses for Google Play, the Google Chrome browser, and for Google Search,” Walker continues. “In doing so, we maintained the freedom for phone makers to install any alternative app alongside a Google app.”

Other opinions are available on those changes too.

Such as French pro-privacy Google search rival Qwant, which last year told us how those licensing changes still make it essentially impossible for smartphone makers to profit off of devices that don’t bake in Google apps by default. (More recently Qwant’s founder condensed the situation to “it’s a joke“.)

Google still claimed to be blocking search rivals on Android, despite Europe’s antitrust action

Qwant and another European startup Jolla, which leads development of an Android alternative smartphone platform called Sailfish — and is also a competition complainant against Google in Europe — want regulators to step in and do more.

The Commission has said it is closely monitoring changes made by Google to determine whether or not the company has complied with its orders to stop anti-competitive behavior.

So the jury is still out on whether any of its tweaks sum to compliance. (Google says so but that’s as you’d expect — and certainly doesn’t mean the Commission will agree.)

In its Android decision last summer the Commission judged that Google’s practices harmed competition and “further innovation” in the wider mobile space, i.e. beyond Internet search — because it prevented other mobile browsers from competing effectively with its pre-installed Chrome browser.

So browser choice is a key component here. And ‘effective competition’ is the bar Google’s homebrew ‘remedies’ will have to meet.

Still, the company will be hoping its latest Android tweaks steer off further Commission antitrust action. Or at least generate more fuzz and fuel for its long-game legal appeal.

Current EU competition commissioner, Margrethe Vestager, has flagged for years that the division is also fielding complaints about other Google products, including travel search, image search and maps. Which suggests Google could face fresh antitrust investigations in future, even as the last of the first batch is about to wrap up.

The FT reports that Android users in the European economic area last week started seeing links to rival websites appearing above Google’s answer box for searches for products, jobs or businesses — with the rival links appearing above paid results links to Google’s own services.

The newspaper points out that tweak is similar to a change promoted by Google in 2013, when it was trying to resolve EU antitrust concerns under the prior commissioner, Joaquín Almunia.

However rivals at the time complained the tweak was insufficient. The Commission subsequently agreed — and under Vestager’s tenure went on to hit Google with antitrust fines.

Walker doesn’t mention these any of additional antitrust complaints swirling around Google’s business in Europe, choosing to focus on highlighting changes it’s made in response to the two extant Commission antitrust rulings.

“After the Commission’s July 2018 decision, we changed the licensing model for the Google apps we build for use on Android phones, creating new, separate licenses for Google Play, the Google Chrome browser, and for Google Search. In doing so, we maintained the freedom for phone makers to install any alternative app alongside a Google app,” he writes.

Nor does he make mention of a recent change Google quietly made to the lists of default search engine choices in its Chrome browser — which expanded the “choice” he claims the company offers by surfacing more rivals. (The biggest beneficiary of that tweak is privacy search rival DuckDuckGo, which suddenly got added to the Chrome search engine lists in around 60 markets. Qwant also got added as a default choice in France.)

Talking about Android specifically Walker instead takes a subtle indirect swipe at iOS maker Apple — which now finds itself the target of competition complaints in Europe, via music streaming rival Spotify, and is potentially facing a Commission probe of its own (albeit, iOS’ marketshare in Europe is tiny vs Android). So top deflecting Google.

“On Android phones, you’ve always been able to install any search engine or browser you want, irrespective of what came pre-installed on the phone when you bought it. In fact, a typical Android phone user will usually install around 50 additional apps on their phone,” Walker writes, drawing attention to the fact that Apple does not offer iOS users as much of a literal choice as Google does.

“Now we’ll also do more to ensure that Android phone owners know about the wide choice of browsers and search engines available to download to their phones,” he adds, saying: “This will involve asking users of existing and new Android devices in Europe which browser and search apps they would like to use.”

We’ve reached out to Commission for comment, and to Google with questions about the design of its incoming browser and search app prompts for Android users in Europe and will update this report with any response.

Read Full Article