|To: E_K_S who wrote (15363)||5/24/2020 5:38:56 PM|
|Related - |
Google bans Podcast Addict app after 9 years for letting users play podcasts that reference COVID-19
By Tom Parker Posted 7:38 pm
Google says it needs to be authorized by official government entities or public health organizations.
If you're tired of censorship subscribe to Reclaim The Net.
One of the most popular podcast players on the Google Play store, Podcast Addict, has been suspended on the grounds that some of the podcasts it indexes reference the coronavirus but aren’t approved by government entities or public health organizations.
According to the now-removed Google Play store listing for Podcast Addict, it was the number one podcast app on Android with over 9 million downloads and over 2 billion podcast episodes downloaded through the app.
It had over 500,000 user ratings with an average rating of 4.7/5 and had been on the Google Play Store for more than nine years.
Podcast Addict was one of the most popular podcast apps in the Google Play store (Wayback Machine – Google Play store – Podcast Addict)Google told Xavier Guillemane, the developer of Podcast Addict, that the app had been removed due to its references to COVID-19:
“Pursuant to Section 8.3 of the Developer Distribution Agreement and the Enforcement policy, apps referencing COVID-19, or related terms, in any form will only be approved for distribution on Google Play if they are published, commissioned, or authorized by official government entities or public health organizations.“In order to get the app restored, Google is asking Guillemane to remove references to COVID-19 and keywords related to COVID-19 from the app.Google told Podcast Addict the app was removed for referencing the coronavirus without authorization from official government entities or public health organizations (Twitter @PodcastAddict)Podcasts player apps such as Podcast Addict connect users with podcasts but don’t actually host the content.
The way podcast apps serve up this content is similar to how web browsers connect users with websites but don’t actually host any of the content on those websites.
What Google is asking of Podcast Addict would be comparable to Google asking a web browser app to remove references to all the websites and social media posts that reference the coronavirus unless the reference comes from an official government entity or public health organization.
Additionally, Google isn’t applying these same rules to its own podcast app – Google Podcasts.
Searching for “ coronavirus”or “ COVID-19” in Google Podcasts returns podcasts from mainstream media outlets that aren’t endorsed by government entities or public health organizations.
|RecommendKeepReplyMark as Last Read|
|From: Glenn Petersen||5/25/2020 12:13:24 PM|
|Google drops future AI oil extraction projects|
May 19, 2020
Google will no longer develop new artificial intelligence tools to help oil and gas companies extract crude, the company announced Tuesday.
Why it matters: The tech giant is breaking away from Microsoft and Amazon, both of which have also developed AI in recent years to expedite oil production and make services more efficient for companies like Chevron and GE Oil & Gas. Google's 2018 contract with Total was in place as of February, a Total spokesperson confirmed to Axios at the time.
What they're saying: Google Cloud "will not, for instance, build custom AI/ML algorithms to facilitate upstream extraction in the oil and gas industry," a Google spokesperson told Axios on Tuesday. OneZero first reported the move.
-- Yes, but: The company will honor its current contracts. The spokesperson declined to say if Google's work with Total would continue, noting that customer contracts are confidential. Total did not respond to a request for comment.
-- Google's 2019 revenue from oil and gas came out to roughly $65 million, which accounted for less than 1% of the company's total revenue in that period, the spokesperson said.
-- On renewable energy, Google says it is applying algorithms from its own data centers to improve efficiency in buildings.
-- Of note: Google's announcement dropped alongside a Greenpeace report released Tuesday that detailed cloud computing and AI contracts between Google, Microsoft and Amazon and oil and gas companies.
Between the lines: Microsoft and Amazon have said that working with the oil industry isn’t at odds with their climate commitments. In some cases, they're working with Big Oil on clean energy plans — like BP giving AWS renewable power.
What we're watching via Axios' Amy Harder: It's too soon to tell whether Google proves to be an outlier or an early indicator of a trend, but one thing is clear now: Environmentalists will be ramping up the pressure on other tech giants and companies in other sectors to sever ties with oil and gas firms now that they've had success with one.
Go deeper: Climate activists target Big Tech over fossil fuel work
|RecommendKeepReplyMark as Last Read|
|From: Glenn Petersen||6/5/2020 11:05:13 AM|
|How Google is using emerging AI techniques to improve language translation qualit|
Kyle Wiggers @Kyle_L_Wiggers
June 3, 2020
Google says it’s made progress toward improving translation quality for languages that don’t have a copious amount of written text. In a forthcoming blog post, the company details new innovations that have enhanced the user experience in the 108 languages (particularly in data-poor languages Yoruba and Malayalam) supported by Google Translate, its service that translates an average of 150 billion words daily.
In the 13 years since the public debut of Google Translate, techniques like neural machine translation, rewriting-based paradigms, and on-device processing have led to quantifiable leaps in the platform’s translation accuracy. But until recently, even the state-of-the-art algorithms underpinning Translate lagged behind human performance. Efforts beyond Google illustrate the magnitude of the problem — the Masakhane project, which aims to render thousands of languages on the African continent automatically translatable, has yet to move beyond the data-gathering and transcription phase. And Common Voice, Mozilla’s effort to build an open source collection of transcribed speech data, has vetted only 40 voices since its June 2017 launch.
Google says its translation breakthroughs weren’t driven by a single technology, but rather a combination of technologies targeting low-resource languages, high-resource languages, general quality, latency, and overall inference speed. Between May 2019 and May 2020, as measured by human evaluations and BLEU, a metric based on the similarity between a system’s translation and human reference translations, Translate improved an average of 5 or more points across all languages and 7 or more across the 50 lowest-resource languages. Moreover, Google says that Translate has become more robust to machine translation hallucination, a phenomenon in which AI models produce strange “translations” when given nonsense input (such as “Shenzhen Shenzhen Shaw International Airport (SSH)” for the Telugu characters “? ? ? ? ? ? ? ? ? ? ? ? ? ? ?,” which mean “Sh sh sh sh sh sh sh sh sh sh sh sh sh sh sh sh sh”).
Hybrid models and data miners
The first of these technologies is a translation model architecture — a hybrid architecture consisting of a Transformer encoder and a recurrent neural network (RNN) decoder implemented in Lingvo, a TensorFlow framework for sequence modeling.
In machine translation, encoders generally encode words and phrases as internal representations the decoder then uses to generate text in a desired language. Transformer-based models, which Google-affiliated researchers first proposed in 2017, are demonstrably more effective at this than RNNs, but Google says its work suggests most of the quality gains come from only one component of the Transformer: the encoder. That’s perhaps because while both RNNs and Transformers are designed to handle ordered sequences of data, Transformers don’t require that the sequence be processed in order. In other words, if the data in question is natural language, the Transformer doesn’t need to process the beginning of a sentence before it processes the end.
Still, the RNN decoder remains “much faster” at inference time than the decoder within the Transformer. Cognizant of this, the Google Translate team applied optimizations to the RNN decoder before coupling it with the Transformer encoder to create low-latency, hybrid models higher in quality and more stable than the four-year-old RNN-based neural machine translation models they replace.
Image Credit: Google
Beyond the novel hybrid model architecture, Google upgraded the decades-old crawler it used to compile training corpora from millions of example translations in things like articles, books, documents, and web search results. The new miner — which is embedding-based for 14 large language pairs as opposed to dictionary-based, meaning it uses vectors of real numbers to represent words and phrases — focuses more on precision (the fraction of relevant data among the retrieved data) than recall (the fraction of the total amount of relevant data that was actually retrieved). In production, Google says this increased the number of sentences the miner extracted by 29% on average.
Noisy data and transfer learningAnother translation performance boost came from a modeling method that better treats noise in training data. Following from the observation that noisy data (data with a large amount of information that can’t be understood or interpreted correctly) harms translation of languages for which data is plentiful, the Google Translate team deployed a system that assigns scores to examples using models trained on noisy data and tuned on “clean” data. Effectively, the models begin training on all data and then gradually train on smaller and cleaner subsets, an approach known in the AI research community as curriculum learning.
On the low-resource language side of the equation, Google implemented a back-translation scheme in Translate that augments parallel training data, where each sentence in a language is paired with its translation. (Machine translation traditionally relies on the statistics of corpora of paired sentences in both a source and a target language.) In this scheme, training data is automatically aligned with synthetic parallel data, such that the target text is natural language but the source is generated by a neural translation model. The result is that Translate takes advantage of the more abundant monolingual text data for training models, which Google says is especially helpful in increasing fluency.
Above: Google Maps with Translate.
Translate also now makes use of M4 modeling, where a single, giant model — M4 — translates among many languages and English. (M4 was first proposed in a paper last year that demonstrated it improved translation quality for over 30 low-resource languages after training on more than 25 billion sentence pairs from over 100 languages.) M4 modeling enabled transfer learning in Translate, so that insights gleaned through training on high-resource languages including French, German, and Spanish (which have billions of parallel examples) can be applied to the translation of low-resource languages like Yoruba, Sindhi, and Hawaiian (which have only tens of thousands of examples).
Translate has improved by at least 1 BLEU point per year since 2010, according to Google, but automatic machine translation is by no means a solved problem. Google concedes that even its enhanced models fall prey to errors including conflating different dialects of a language, producing overly literal translations, and poor performance on particular genres of subject matter and informal or spoken language.
The tech giant is attempting to address this in various ways, including through its Google Translate Community, a gamified program that recruits volunteers to help improve performance for low-resource languages by translating words and phrases or checking if translations are correct. Just in February, the program, in tandem with emerging machine learning techniques, led to the addition in Translate of five languages spoken by a combined 75 million people: Kinyarwanda, Odia (Oriya), Tatar, Turkmen, and Uyghur (Uighur).
Google isn’t alone in its pursuit of a truly universal translator. In August 2018, Facebook revealed an AI model that uses a combination of word-for-word translations, language models, and back-translations to outperform systems for language pairings. More recently, MIT Computer Science and Artificial Intelligence Laboratory researchers presented an unsupervised model — i.e., a model that learns from test data that hasn’t been explicitly labeled or categorized — that can translate between texts in two languages without direct translational data between the two.
In a statement, Google diplomatically said it’s “grateful” for the machine translation research in “academia and industry,” some of which informed its own work. “We accomplished [Google Translate’s recent improvements] by synthesizing and expanding a variety of recent advances,” said the company. “With this update, we are proud to provide automatic translations that are relatively coherent, even for the lowest-resource of the 108 supported languages.”
|RecommendKeepReplyMark as Last ReadRead Replies (1)|
|To: Glenn Petersen who wrote (15366)||6/5/2020 2:33:12 PM|
|From: Glenn Petersen|
|States are leaning toward a push to break up Google’s ad tech business|
Published Fri, Jun 5 20201:22 PM EDT
Lauren Hirsch @laurenshirsch
Megan Graham @megancgraham
-- The state attorneys general investigating Google are considering pushing for a breakup of the tech giant’s ad technology business as part of an expected lawsuit, people familiar with the situation told CNBC.
-- Critics have said that Google bundles its ad tools so that rivals can’t afford to match its offerings and that its operation of search results, YouTube, Gmail and other services to hinder ad competition.
-- Once the attorneys general file their expected lawsuit, they have a number of tools at their disposal to signal their intent to push for a breakup of Google’s ad technology business.
The state attorneys general investigating Google for potential antitrust violations are leaning towards pushing for a breakup of its ad technology business as part of an expected suit, people familiar with the situation told CNBC.
Fifty attorneys general have been probing Google’s business practices for months, alongside a similar probe being led by the U.S. Department of Justice. Both the states and the DOJ are looking to file a suit against the internet giant as soon as within the next few months, people familiar with the situation told CNBC.
The states and the Justice Department have not yet officially decided whether to combine their expected suits, the people said, though they have been collaborating closely. Both have been investigating Google’s search, ad technology and android business.
The attorneys general investigating Google, which is owned by Alphabet, haven’t yet definitively ruled out pushing for alternatives for its ad technology business, like imposing restrictions on how it runs its business, one of the sources said. A suit may also include a push for both that option and breaking up the ad tech business.
Should the attorneys general aggressively pursue a break up of Google’s ad technology business, it would be notable. While regulatory enforcement agencies have recently favored “structural remedies” - break-ups and divestitures - regulators have less onerous solutions available, like barring certain behaviors through a consent decree.
Once the attorneys general file their expected lawsuit, they have a number of tools at their disposal to signal their intent to push for a breakup of Google’s ad technology business. That includes what they allege, the evidence they introduce, pre-trial briefings and press conferences.
In Google’s case, pushing for a break up of its ad technology business may be difficult, some lawyers say, because it does not exist as a stand-alone unit easily hived off. And its two main deals, DoubleClick in 2007 and AdMob in 2009, were years ago.
“Courts are very concerned that by ripping a company apart, it hurts consumers and make it worse for people that don’t have the expertise to do that,” said Stephen Houck, one of the government lawyers in the Microsoft antitrust case two decades ago.
While Google generates the majority of its roughly $161 billion in revenue from ad sales, the revenue it gets from the software and technology that serve as the backbone of that business is far smaller. Its Network Members business, which includes AdMob, AdSense, and Google Ad Manager, generated about $22 billion in sales the last fiscal year.
Google retired the DoubleClick name in 2018, putting its DoubleClick products for advertisers together with Google Analytics 360 to become the “Google Marketing Platform.” Then it put its DoubleClick products for publishers and the DoubleClick ad exchange into the “Google Ad Manager.”
While the government has successfully fought for a break up of corporate giants in the past, including Standard Oil in 1911 and AT&T in the 1980s, more recent cases have imposed weaker remedies. Both IBM in the 1980s and Microsoft in 2000 concluded antitrust suits without breaking up the respective companies.
Still, political winds have more recently seemingly turned against big technology companies.
President Donald Trump has alleged, along with other Republicans, that companies like Google censor conservative content, a claim Google has denied. He earlier this month signed an executive order vowing to crack down on the liability protections internet companies like Google through section 230 of the Communications Decency Act.
Attending the signing of that executive order was Attorney General William Barr, whose deputy recently took leadership of the Justice Departments’ antitrust investigation into Google, after the country’s top antitrust official, Makan Delrahim, recused himself.
Presidential candidate Joe Biden served as vice president to Barack Obama, who has been criticized for allowing tech companies to become too powerful on his watch. Biden has said that, as president, he would set up a new department within the Justice Department to go back and look at the mega-mergers that have occurred and those who are being proposed to occur.
A spokesperson for Google told CNBC in a statement, “We continue to engage with the ongoing investigations led by the Department of Justice and Attorney General Paxton, and we don’t have any updates or comments on speculation.”
“The facts are clear,” she added, “our digital advertising products compete across a crowded industry with hundreds of rivals and technologies, and have helped lower costs for advertisers and consumers.”
A spokesperson for Texas Attorney General Ken Paxton, who is leading the ad tech part of the probe, declined to comment. A spokesperson for DOJ did not respond to a request for comment.
What’s at stake
Critics have said that Google bundles its ad tools so that rivals can’t afford to match its offerings and that its operation of search results, YouTube, Gmail and other services to hinder ad competition. They also say that Google owns all sides of the “auction exchange” through which ads are sold and bought, giving it an unfair advantage.
Google has argued it competes with many vertically integrated players including AT&T, Comcast, and Verizon. Data from the St. Louis Federal Reserve, meantime, show that the price of digital advertising has fallen by more than 40% since 2010.
While it remains unclear in what fashion the attorneys general might push to break up Google’s ad tech business, Google’s $3.1 billion acquisition of DoubleClick provided it the crucial foothold into advertising technology. The Federal Trade Commission decided in a 4-1 vote it would not seek to block the deal, ruling that it was not anti-competitive.
The dissenting vote, Pamela Jones, argued at the time she worried the deal would give Google too much power by way of the data that DoubleClick provides.
“By purchasing DoubleClick, Google will acquire data that will contribute to, and exacerbate, network effects,” she wrote. “As a result, the Google/DoubleClick combination is likely to ‘tip’ both the search and display markets in Google’s favor, and make it more difficult for any other company to challenge the combined firm.”
Jones added that the combined Google and DoubleClick could get access to “unparalleled data sources” that would allow it to match up buyers and sellers of ads in a way its competitors could not.
A lawyer for Google said in a Texas court earlier this year the attorneys general probe was seeking “very detailed information” on the names of Google’s ad tech customers and details pertaining to product pricing, according to a court transcript obtained by CNBC. Inquiries focused on Google’s DoubleClick and AdMob businesses, he said.
For its part, Google argued in a blog post last year the ad technology industry is “famously crowded,” citing competitors like Telaria, Rubicon Project and The Trade Desk. It also said publishers use its technology “to access demand from hundreds of partners,” while advertisers use its technology to buy ad space “on more than 80 exchanges.”
|RecommendKeepReplyMark as Last Read|
|From: Sr K||6/5/2020 10:33:28 PM|
|My question is about lifespan, but there could be others.|
Why did Alphabet set up a structure of Class B to control the company forever and have it lsst forever even when they're gon?. All I remember from 2004 is Do no evil, and $85/share.
What are they doing now? Taking a Sabbatical? Creating a new company that goes into one of the areas Alphabet pursued, or ignored?
When they decided to leave, was it based on Google data, before March 23? Did they see anything that's been mentioned in the conference calls? Maybe I could listen to the last few.
I think it's a ripoff of the public shareholders, and it should have smoothly faded out after about 2024, or 20 years.
What is the outcome of Proxy Votes? High vote from others besides Class B? Compared to others, I don't even watch their quarterly conference calls, because the connection to engineering is gone, and all that remains is marketing and untrustworthy ads to milk the cow.
|RecommendKeepReplyMark as Last Read|
|From: Glenn Petersen||6/21/2020 12:00:10 PM|
|Google got rich from your data. DuckDuckGo is fighting back|
In July 2018, Google was fined €4.34 billion for limiting search on Android phones. Almost two years later, its rivals claim little has changed and the company is as dominant as ever
By Matt Burgess
Monday 8 June 2020
From this summer, when you set up a new Android phone or tablet in Europe, you will be presented with an extra step: a choice screen to select your default search engine.
The screen is simple. Under a search icon and a short blurb are four options. One is always Google; the others vary depending on which country you’re in. Pick one of Google’s rivals, and its results – not Google’s – will appear in a home screen search widget and in the Chrome web browser. Its app will also be downloaded to the device.
The change is so subtle that many people may not notice it. But across the continent, it marks a fundamental shift. The choice screen appears in the wake of the European Commission fining Google €4.34 billion (£3.81bn) for breaching EU antitrust rules. On July 18, 2018, competition commissioner Margarethe Vestager hit Google with the fine – the largest of three she has issued against the company – for abusing the dominance of its Android operating system.
Being big isn’t illegal, but the European Commission says dominant companies have a "special responsibility" not to abuse their powerful market position by restricting competition. At the end of an investigation spanning more than three years, the Commission concluded that Google had acted illegally in using Android to ensure that traffic went to Google Search instead of its competitors. The Commission accused Google of three unfair tactics. Google, it said, had required manufacturers to pre-install the Google Search and Chrome apps if they wanted to license the Google Play app store; had paid mobile network operators and phone manufacturers to exclusively pre-install Google Search; and had prevented manufacturers from creating alternative versions of Android without its approval if they wanted to pre-install any Google apps.
Google’s actions, the Commission said, helped it to cement its dominance in the mobile search industry. In Europe, more than 95 per cent of all searches on mobile are made through Google. The market is mostly Android devices, but Google also pays billions to Apple to be the default search engine in the Safari browser on iOS. According to Vestager, its Android practices “denied rivals the chance to innovate and compete on the merits.” Google is appealing the decision.
The way many European antitrust fines work puts the onus of fixing a problem on the guilty party. Google has re-written the contracts it holds with phone makers and relaxed limits on how Android can be developed by others. On new phones and tablets shipped into the European Economic Area, it has introduced the search engine choice screen.
The choice screen gives Google’s rivals an unprecedented presence on Android devices. Smaller players have never had such an opportunity to directly reach users on the world’s biggest mobile operating system.
Chief among those set to benefit is DuckDuckGo, a US-based search engine founded in 2008 by CEO Gabriel Weinberg. DuckDuckGo will appear on all 31 choice screens across Europe when it is first implemented, a number matched only by Info.com. The company has been competing with Google for more than a decade, offering a search engine that emphasises privacy and doesn’t collect user data, but it is still tiny compared to Larry Page and Sergey Brin’s creation. “We're not trying to topple Google,” Weinberg, 40, says. “Our goal is for consumers who want to choose a private option, they should be able to do so easily.”
The idea for DuckDuckGo came partly from a stained glass class. In 2007, a couple of years after completing his masters in technology policy at MIT, and having already sold a social networking startup, Opobox, for $10 million, Weinberg decided to try his hand at a new hobby. The class teacher handed out a print-out of the best websites to learn more about creating stained glass. When Weinberg tried to search for information on Google, none of the recommended sites appeared in the top results.
The sheet of paper led to the creation of I’ve Got a Fang, a crowd-sourced site where people could contribute authoritative URLs on particular subjects. Weinberg, who lives in Philadelphia, believed the knowledge from people’s heads could be better than Google’s algorithms.
I’ve Got a Fang was a flop: few people added links to the site and it failed to get any traction. But the idea of human curation stuck, and Weinberg combined this notion with the remnants of another failed side gig – one that had already run into trouble with Google. Tldscan was a series of websites that crawled structured information and published it to the web – think sports statistics available in one place. In a now-archived 2010 blog post, “My history of (mostly failed) side projects and startups”, Weinberg wrote that he was getting 50,000 visitors and $500 revenue per day from Tldscan. “Then Google blacklisted all of my sites,” he wrote.
The combination of I’ve Got a Fang’s human expertise and Tldscan’s efforts to surface useful information became key pillars of DuckDuckGo, which launched a year later in September 2008. “My original spark was: could there be a better user experience in search?” says a bespectacled, clean-shaven Weinberg, speaking via a Zoom call in March. (While many companies had to radically adapt their way of working as the Covid-19 outbreak spread across the world, little changed for DuckDuckGo; its 89 staff work remotely all the time, and there is no main office.)
Initially, Weinberg started building the technology needed to crawl, index and rank the web himself, but he stopped when Yahoo! introduced a service called Build Your Own Search Service (BOSS). Using this, he focused on refining Yahoo!’s results and adding extra services on top. Today, DuckDuckGo combines data from more than 400 sources into its search results: Microsoft’s Bing is predominantly used to surface relevant pages, but Wikipedia, Apple Maps, TripAdvisor and DuckDuckGo’s own web crawler also contribute. There are more than 1,000 different sources for the “Instant Answer” snippets DuckDuckGo shows alongside its results.
It wasn’t until 2010 that the search company set out its unique selling point: it wasn’t going to track users. The company committed to throwing away IP address information and not storing details of what people searched for. “At the beginning, privacy wasn't the main draw,” Weinberg says. “It wasn't as known as an issue.”
Initial growth was slow; in April 2010 the site was getting around 30,000 to 40,000 search queries a day, barely appearing as a blip on the search radar. Weinberg didn’t hire his first full-time employee until 2011, and up until then he was primarily focused on being a stay-at-home dad for his first child, born the year after DuckDuckGo launched. During some early investor meetings, he says, he was rocking his second child out of his webcam’s field of vision. The company raised $3 million in a Series A round in 2011, led by Union Square Ventures.
DuckDuckGo’s real breakthrough came thanks to Edward Snowden. The former NSA contractor’s explosive government surveillance revelations in 2013 prominently included Google among the data sources used by British and US spy agencies. People were newly interested in seeking out privacy-first alternatives, and DuckDuckGo’s daily search queries swelled from 1.5 million a day to 4.1 million a day in the space of a year.
At the time, many people switching from Google wouldn’t have been happy with the experience their new search engine offered, and it’s likely that plenty quickly switched back. Weinberg says that it wasn’t until 2014 that DuckDuckGo could really compete with its Silicon Valley rival. It was a crucial year for the company: it hit its tenth employee; incorporated news, video and image verticals in a big site redesign; and was included as an alternative search engine option in Apple’s Safari browser on iOS. “Apple is effectively a mainstream endorsement,” Weinberg says.
It was in 2014 that DuckDuckGo also made its first profit; it’s been profitable ever since, and the company forecasts it will pass the $100 million revenue barrier for the first time in 2020 (thought this may be affected by the Covid-19 pandemic).
It makes its money in the same way as Google: through advertisements. Each time a user clicks on an ad that is shown alongside search results, the company takes home a small amount of money. But where the two differ is that DuckDuckGo uses contextual advertising over the data-heavy behavioural advertising that helps Google earn billions each quarter.
Contextual adverts are shown to users based on the searches they make: if you DuckDuckGo “Mercedes”, you’ll be served adverts for cars on that page. Google “Mercedes”, however, and adverts for cars will follow you off the page and around the web, because Google adds that data to the other information it knows about you. Behavioural advertising is much more lucrative, as each time an ad is clicked on there’s a higher likelihood the person will buy. Google’s expertise in knowing what users will be most likely to click on is rivalled only by that of Facebook.
People’s increased awareness of privacy issues has continued to fuel DuckDuckGo’s growth. As with Snowden’s revelations, the 2018 Cambridge Analytica scandal propelled Weinberg and DuckDuckGo into the mainstream. Around the same time, the company launched desktop browser extensions that stop web tracking, as well as its own mobile web browser for Android and iOS.
The cumulative effect was significant. On March 17, 2018, the day The Observer first reported on Cambridge Analytica, 17 million DuckDuckGo searches were made; exactly two years later, 57 million searches were completed. Weinberg says DuckDuckGo’s key users are people who both care about their privacy and are willing to act upon it. “The caring really went up after Snowden,” he says. “After Cambridge Analytica the acting percentage has really skyrocketed.”
Google’s new Android choice screen could give another boost to DuckDuckGo, finally propelling it into the mainstream as millions of people replace their old phones over the coming years. “It's really a levelling, if it's done right, for competition,” Weinberg says. But he, and other search engines, have issues with Google’s implementation.
Multiple European search startups say they are in favour of a choice screen. They argue that users should be allowed to pick which search tool they use – even if they end up choosing Google (in most cases they will). But Google’s execution of the choice screen has angered its rivals.
To appear, companies have to pay Google, via an auction that is held every three months. Bid enough, and you win a coveted slot as one of three options alongside Google. A similar auction approach was introduced as a fix to a 2017 EU antitrust case around Google Shopping. In search, companies say they don't think an auction is a fair way of encouraging competition or innovation.
Google argues that an auction is a “fair and objective” way of smaller companies being able to compete, and that it allows them to decide “what value they place on appearing in the choice screen”. Every time an Android user picks a search engine to use – even if they switch back to Google minutes later – that search engine is charged. Google issues the companies an invoice each month, and the money goes back into the development of Android.
Ahead of the first round of bidding towards the end of 2019, Google set a minimum bid. Multiple sources, who are not able to speak publicly, say it set prices of above $20 per user for some European countries. This amount would be unrealistic for small companies, the sources say. “I actually think they saw this as a price we would easily be able to pay,” one person familiar with the bidding says, adding that it was “ridiculous”.
Ultimately, the minimum bid was scrapped in favour of a system where all the winners pay the price set by the first losing bidder. Google says it changed the format after receiving feedback and to make it easier for rivals to participate. One source says companies will now be paying the “lowest number to zero that you can do besides being zero.” Another company says the only reason it bid in the process was because it was cheaper than advertising on Facebook. “We will get our name out more than I think we will directly get users from the choice screen,” the company CEO says.
DuckDuckGo is one of the auction’s big winners, appearing as an option in all 31 countries. But it isn’t happy about it. “You're incentivised to bid your profit,” Weinberg says. He predicts the auction prices will keep rising until it’s harder for search engines to make any money. “Google is taking away all the profit of these search engines and profiting from it themselves,” he says. “They're effectively running other people's search and taking other people's profit.”
“It plays the competitors against each other,” says Christian Kroll, the CEO of Ecosia, a German search engine that plants trees when searches are made on its platform. “So instead of collaboration, we have now kind of a toxic environment between smaller search engines as well.” Ecosia decided to boycott the auction. Kroll says in the long term he doesn’t believe the company can win, because 80 per cent of its profits are spent on tree planting. “I think this was very well thought out by Google,” he says.
DuckDuckGo has loudly criticised Google’s efforts. Weinberg and DuckDuckGo general counsel Megan Gray, who previously worked at the US Federal Trade Commission in consumer protection, argue that Google’s process is flawed. They say there should be more than four options shown to users, and that people should be given more information about the choice they’re making. If there was one fix? “Getting rid of the damn auction,” Gray says. “It should be a genuine choice – it shouldn't be pay to play. That would be the best outcome here.”
Paul Gennai, a product management director at Google, says that Android has “created more choice for everyone, not less” and that the operating system's openness has meant “developers of apps that compete against Google have precisely the same ability to reach users”.
“From day one, Google developed Android to be open: both to the manufacturers who build devices, as well as to the people who buy them – we think both should have the flexibility to set them up just as they want,” Gennai says. “This has taken billions of dollars of investment over more than ten years.” He adds that Google also invests in its Play app store, which includes safety features such as scanning apps for malware. “With the changes required in Europe, we implemented an auction to ensure that it would continue to be viable for Google to continue its investments in the open Android ecosystem,” he says.
A spokesperson for the European Commission says it is closely monitoring how the choice screen is being introduced and the mechanisms behind it. "We have been discussing the choice screen mechanism with Google, following relevant feedback from the market, in particular in relation to the presentation and mechanics of the choice screen and to the selection mechanism of rival search providers," the spokesperson says. (Since this story went to press, the results of the second Android choice screen auction have been published. Fewer companies are included as winners and Microsoft's Bing has no presence. DuckDuckGo says prices have increased, too).
Aside from controversies around the auction, there are questions over how much difference the choice screen will make. A 2010 EU antitrust investigation saw Microsoft launch a choice screen for web browsers – the Commission had said that Microsoft used the dominance of the Windows operating system to direct people to its Internet Explorer browser, which was included as the default on Windows. Its choice screen ran for four years, but at the end of this time smaller browsers had not gained any significant market share. (The biggest market shift was the rise of Google's Chrome browser, which had only recently launched and was rising in popularity anyway.)
And in search, Google nearly always wins. One major exception – aside from in China, where Google search is banned – is in Russia. There, Yandex is king. Launched in 1997, a year before Google, Yandex acts as the country’s homegrown Uber, Spotify, Amazon and favoured search engine all rolled into one. At the start of 2015, Russians searching the web on their desktop machines were overwhelmingly likely to be using the company – 63 per cent of all searches were made using Yandex. But on mobile it was a different story. In an Android-dominant country, the tables were turned, with 62 per cent of all mobile searches being made via Google.
Staff at Yandex were aghast when they discovered Google had struck deals with phone makers that prohibited them from accepting payments to pre-install other search engines if they wanted to use the Android operating system.
As in the EU, regulators in Russia were scrutinising the power Android gives Google search. In March 2017, the Federal Antimonopoly Service (FAS) fined Google 438,067,400 roubles (£4.4 million) and then reached a settlement with the company that said Android would offer a choice screen for search engines, in this case on both new and existing devices (Google says technical changes mean the EU choice screen can only apply to new phones).
Smartphone users in Russia started to see choice screens from August 2017. That month, search on Android was 60 per cent Google. The gap closed and a year later it was split 50/50 with Yandex. At the start of March 2020, Yandex had 56 per cent of all Android searches, with Google’s share dropping to 42 per cent.
But while Yandex was also one of the bigger winners in the EU Android auction, appearing on device screens in 16 countries, Preston Carey, a senior vice president at Yandex, says he doesn’t expect to see any large market shifts. “Putting a choice screen up in the UK or Germany is not going to have the same result as putting one up in Russia,” he says. He argues that no other search company has the brand recognition that Yandex does in Russia in any other country – meaning people will be less likely to switch away from Google.
In its own tests, DuckDuckGo created a choice screen with 18 different options and placed Google at the end of the list. It found people still scrolled all the way to the bottom to find the one they were most familiar with.
While Android’s new choice screen is unlikely to end Google’s dominance in search, even a one per cent shift in market share over a period of several years would be colossal growth for firms such as DuckDuckGo. Weinberg predicts his firm could one day have a market share of ten per cent (it currently sits at 0.32 per cent of global market share). This would still pale in comparison to Google but be a big boost to the 1.5 billion monthly DuckDuckGos already taking place.
And to DuckDuckGo, the battle for search represents just a small part of a bigger struggle over how advertising on the web works, and what this means for personal data and privacy.
The vast ad networks of Facebook and Google are fed with personal data collected across the web. Trackers from both companies gather information about online activity from almost all of the web’s most popular sites. By collecting billions of data points, the companies are able to build revenues in excess of $20 billion every three months. The vast majority of this money comes from advertising – in particular behavioural advertising.
Trackers and cookies are present across almost all websites; they monitor and track people’s activities online and log them under unique identifiers connected to individuals (Wired.co.uk uses trackers to collect and share reader data as part of business practices). But at the start of April this year, the data protection regulator for Ireland, which oversees all the big US tech companies headquartered in the country, released a report saying many websites fail to give “basic information” to people that visit them and that “most ordinary users will not be aware of the extent to which they may be tracked”.
This tracking allows companies to collect data on people’s interests and activities, which they can feed into advertising models and use to serve people behavioural adverts linked to their interests; Facebook’s Pixel, for example, is on more than eight million websites. “Strategically, Google and Facebook have all the user data,” Weinberg says. DuckDuckGo’s web browser, like its search engine, does not store personal data. In addition, it blocks the invisible trackers hidden on web pages. The company wants to see less tracking elsewhere, too.
In 2009, security researchers proposed a Do Not Track setting for browsers. In theory this is a checkbox, or an option in a browser’s settings menu, that allows a user to opt out of being tracked as they move around the web. Firefox, Internet Explorer, Opera and Chrome all use it, but it's largely ineffective. (Apple's Safari stopped using it in 2019). Websites across the web don’t have to honour the setting; Medium and Pinterest are two of the most prominent that do. Twitter originally supported it but doesn’t anymore, and efforts from web standards bodies have failed to make it more effective. Despite having the feature in Chrome, Google’s support pages say the company’s own websites “don't change their behavior” when Do Not Track is turned on.
In 2018, consulting firm Forrester found that around a quarter of Americans had the largely-ineffective setting switched on. A year later, lawyers for DuckDuckGo proposed “The Do-Not-Track Act of 2019”, a draft piece of legislation designed to make the Do Not Track setting be backed by law: websites would have to respect a user’s wishes to not have their behaviour online recorded.
The idea has not been put into practice anywhere in the US but a similar concept is marginally closer to reality in Europe. For the last three years, European legislators have been debating the ePrivacy Regulation, a complimentary law to GDPR which could include Do Not Track provisions. The regulation has been subject to intense lobbying and is currently in a state of stagnation. But it’s the closest anywhere has got to limiting how people are tracked online.
Weinberg believes its enactment would open up the online advertising market and thus potentially reduce the dominance of Facebook and Google in the space. If a quarter of the web was suddenly using a setting that meant it was harder to serve them adverts targeted at their interests, traditional contextual advertising would rise again. This would be good for privacy, but it could also help open up competition, Weinberg says; if Google and Facebook didn’t have such a data advantage, perhaps smaller companies could compete on a more even footing. “If everyone competes on contextual advertising, now all of a sudden they actually have to compete with a lot of other players,” he says. Google has already been experimenting further with contextual advertising, which one of its product managers wrote in a blog post was to help preserve user privacy.
Competition regulators are also shifting their focus to how data is used by big tech companies. On both sides of the Atlantic, examinations are underway into how Alphabet, Amazon, Apple, Facebook and Microsoft have acquired smaller companies. The EU is also looking more broadly at how Google and Facebook collect data and the implications for competition.
Ultimately, Weinberg wants people to understand that they don’t need to give away their personal information to make the most of technology. “We really want to combat people's learned helplessness in privacy,” he says. “Everyone's concerned, but a large percentage of people just think there's nothing you can really do about it. We want to prove that you can do something.”
Updated: This article first appeared in print. Prior to online publication Megan Gray's role at the FTC has been updated and Gabriel Weinberg was rocking his second child while talking to investors
Matt Burgess is WIRED's deputy digital editor. He tweets from @mattburgess1
|RecommendKeepReplyMark as Last Read|
|From: Glenn Petersen||6/22/2020 9:40:43 AM|
|Google's U.S. Ad Revenue Is Expected to Decline in 2020, eMarketer Says |
Research firm eMarketer says decline would be the first since it started tracking Google’s ad revenue in 2008
By Keach Hagey
Wall Street Journal
June 22, 2020 9:00 am ET
Google’s U.S. advertising revenue will decline this year for the first time since eMarketer began modeling it in 2008, the research firm said, largely because Google’s core search product is so reliant on the pandemic-battered travel industry.
As the world’s largest digital-advertising company, the Alphabet Inc. GOOG 0.02% unit has heretofore been a money-printing machine, expanding its overall advertising revenue at double-digit rates in nearly every year of its two-decade existence, save for the 2008-09 financial crisis, when it only grew 8%. The firm doesn’t break out its U.S. revenue, but eMarketer’s model found that even in that crisis, Google’s U.S. ad revenue grew.
What is different this time is the way the coronavirus pandemic obliterated marketing spending in some of Google search’s biggest advertisers, and the way some of these advertisers have been public about rethinking their plans to return to the platform even after the pandemic passes.
“The biggest single culprit here is the travel industry, which has been both hardest hit by the pandemic generally, and has concentrated spending on Google in the past,” said Nicole Perrin, principal analyst at eMarketer. “We have already heard statements from major travel companies, like Expedia, that normally spend billions of dollars on Google, mostly on search, that they are pulling back spending on Google and search, and that will continue for the rest of this year.”
Travel represented about 11% of search ad revenue in 2019, Needham analyst Laura Martin estimated.
Expedia Group Inc., which owns brands such as Travelocity, Orbitz and Vrbo, has historically been one of Google search’s biggest advertisers, according to data from the advertising-research company Kantar.
Big SpendersLargest ad spenders on Google search in 2019:
AmazonWalmartHome DepotWayfairExpediaBest BuyeBayconsumersadvocate.orgProgressiveBooking.comSource: Kantar
On an earnings call in May, Expedia CEO Peter Kern said the pandemic offered the company an “opportunity of an entire reset” on its traditionally search-heavy advertising spending. Expedia has warned its investors that a risk for the company is the way Google has launched travel products in recent years that compete directly with its largest advertising partners, and then uses its search function to drive users to its own products.
“As we wade back in, we’re able to be more precise, be more constrained, watch and learn and grow into it, and not just dive back in head first and spend back to the levels we were at,” he said.
EMarketer also pointed to Amazon, another big search spender, pulling back its search spending sharply as the pandemic hit as it struggled to fulfill orders.
EMarketer estimates that Google’s gross U.S. advertising revenue will decline 4%, while its net U.S. revenue—accounting for the payments it makes to website owners to acquire traffic—will drop by 5% this year.
Google declined to comment.
That is still less of a drop than for the advertising market overall, which eMarketer estimates will shrink by about 7% this year.
Despite the pandemic, digital advertising continues to steadily take share from traditional media. And the “triopoly” of Google, Facebook and Amazon is sucking up an ever-greater share of digital spending.
EMarketer estimates that digital advertising will increase its share of the overall advertising market by about 5 percentage points this year, to 60% from 55%. But in absolute terms, the pie is shrinking. The firm expects digital to grow by nearly 2%, television ad spending to decline by 15% and print to drop by 25%.
The triopoly’s share will also inch up this year, albeit at a much slower rate than before, from 62% to 62.2%. Most notable, though, is the changing face of this triopoly: Google, long the biggest player, is shrinking, while both Facebook and Amazon are growing.
While the pandemic hurt all advertising players, and eMarketer had to revise its estimates of Facebook’s performance downward like everyone else, it still expects net U.S. revenue at Facebook to grow by 5% this year to $31 billion.
“The factors that have propelled Facebook to be a digital advertising superpower in the first place are the same ones keeping advertisers spending during the pandemic: huge reach, effective targeting at scale and performance ad products that tie spending to results,” the report said.
It also helped that Facebook has less exposure to travel, and more exposure to categories that have thrived under lockdown, such as direct-to-consumer retail and gaming, eMarketer noted.
“They were pretty resilient,” Ms. Perrin said.
In recent days, civil-rights groups including the Anti-Defamation League and NAACP have called on advertisers to pull spending from Facebook for the month of July to protest the platform’s approach to policing hate speech and misinformation. Densu Group Inc.’s digital firm 360i advised clients to support the boycott. On Friday, outdoor-gear maker the North Face said it would join.
|RecommendKeepReplyMark as Last Read|
|From: Glenn Petersen||6/26/2020 10:26:21 PM|
|Justice Department expected to file antitrust suit against Google |
The suit, expected to be filed in the coming months, would involve allegations that the search giant has monopolized the advertising technology market.
By LEAH NYLEN
06/26/2020 07:31 PM EDT
Justice Department prosecutors expect to file an antitrust lawsuit against Google in the coming months focused on the company’s dominance in online advertising and search, two individuals familiar with the discussions said Friday.
DOJ lawyers and state antitrust officials met online Friday and discussed contours of the expected complaint, according to the people, who weren’t authorized to speak on the record because the investigation is ongoing. People knowledgeable about the case had previously described litigation as likely.
The suit is expected to involve allegations that the search giant has monopolized the advertising technology market. It is also expected to include allegations that Google has taken steps to extend its monopoly over search, such as through contracts with Apple and cellphone makers who use the Android operating system that require it be the default search engine.
The people cautioned that Attorney General William Barr, who did not attend the meeting, has yet to make a final decision on whether to sue, a judgment he could make in the coming weeks. The department would also need to decide what remedy it would seek, such as trying to break up the company or placing limits on its behavior. Whether the state attorneys general would also sign on to the DOJ complaint isn't yet determined.
Prosecutors are still discussing whether to include other aspects of Google’s conduct related to search, the people said.
A DOJ spokesperson declined to comment.
"While we continue to engage with ongoing investigations, our focus is on creating free products that lower costs for small businesses and help Americans every day,” Google spokesperson Julie Tarallo McAlister said.
The department has shown signs in recent months of seriously pushing forward with litigation based on the antitrust probe it launched a year ago. DOJ and the states have separately recently retained economic experts to help with a potential suit and testify at trial if needed, according to three individuals familiar with the probes. DOJ has also been seeking a litigator to help with its case, they said.
The suit would mark the agency’s first significant monopolization case against a major U.S. company in decades. In the late 1990s, DOJ sued Microsoft over what it called the company’s efforts to use its Windows operating system to stifle the Netscape internet browser. DOJ and states won at trial, a decision that was largely upheld on appeal, but Microsoft later reached a settlement with the government.
Europe’s primary competition authority has brought three cases against Google and imposed fines totaling about $9 billion over the search giant’s conduct. Google has filed appeals in all three cases. The U.S. Federal Trade Commission previously investigated Google over allegations that it biases search results to favor its own products. The agency closed the probe in 2013 without taking action after Google agreed to "voluntary commitments" to alter some of its practices.
|RecommendKeepReplyMark as Last Read|
|From: Sr K||7/2/2020 5:46:52 PM|
With $122 billion net of debt, Alphabet raises YouTube TV 30%.
YouTube, the video-streaming outlet owned by Google parent Alphabet Inc., announced Wednesday the latest price increase for its TV service that competes with traditional cable packages. YouTube TV will now cost $64.99 a month, up from $49.99 previously. The company has raised the price every year since launching the service in 2017, though the latest 30% rise is the biggest so far.
|RecommendKeepReplyMark as Last Read|
|From: J.F. Sebastian||7/12/2020 7:47:17 PM|
| The game is rigged: A former marketer shows you how Big Tech’s advertising practices harm us all|
by Lisa Macpherson
July 10, 2020
As of this writing, it appears the U.S. Justice Department and a group of state attorneys general likely will file antitrust lawsuits against Alphabet Inc.’s Google for an array of anti-competitive practices in its search and ad technologybusinesses. At the same time, the Federal Trade Commission, Justice Department and a group of 47 state attorneys general have opened inquiries into Facebook, looking at antitrust behavior as well as possible privacy law violations.
It’s about time.
For 23 years, I held a front-row seat to the advent and evolution of digital marketing, witnessing its inevitable rise and twisted consequences. I specialized in helping companies adapt to the digital revolution but abandoned the field when I realized that the capabilities we marketers supported, like harvesting data, profiling consumers and personalizing content, were damaging society and our democracy. This is why I no longer help brands, including several who made their mark bringing people together, tear our communities apart.
I also left because I realized the game is rigged. The field is tilted, in favor of the two dominant players — Google and Facebook — who set the rules. This is the story of how we got here and what we can do about it.
Advertising abandons context for eyeballs: In the 1990s, advertisers negotiated directly with online publishers to place ads on websites in context that was relevant to their products, just as they had with traditional publishers. This was generally a real negotiation on an even playing field. But as we gained more data about our customers’ behavior, we started placing ads wherever we could reach their eyeballs — regardless of the content or location.
As the number of websites exploded, including thousands of small sites reaching niche audiences, we had to automate the process. By the early 2010s, a “negotiation” had transformed into a blind bidding war facilitated by layers of software programs — and every software provider levied a fee. These layers of advertising technology often meant advertisers did not even know where their ads were running or who was actually seeing them.
Ad costs rise while ad value and consumer benefits plummet: In 2017, a scathing book about digital advertising called “ Bad Men” explained how “the amazing world of ‘ad tech’ magically turns a dollar of online advertising into three cents of value.” Fraud provided one explanation for this value loss. Ads could end up on any of thousands upon thousands of small, unknown sites. We were told we were gaining “reach” and finding new audiences. Then we discovered that many of these sites were gaining traffic from bots — software programs designed to repeatedly load webpages — and not actually people. To combat this, we added “verification services” to our ad-tech stack — for another fee.
We also learned that these unknown sites could be toxic. We coined a new phrase, “brand safety,” to refer to managing the risk of advertising on sites that spread conspiracy theories, hate speech and propaganda. Now we needed “white lists” and “black lists” of content and sites where our advertising could or couldn’t appear. And thanks to the opacity and complexity of the supply chain of ad inventory, these lists didn’t always work.
They still don’t. Recently, we’ve seen reports of how Google’s ad tech continues to place advertisements of major, unwitting consumer brands on websites with content pushing coronavirus (and other) conspiracy theories. Advertising also lost value as consumers embraced ad blockers to prevent websites from loading advertisements. Ironically, Google launched its own ad blocker for its Chrome browser, essentially enabling the tech giant to set advertising standards, charge advertisers to place ads in content that may violate those standards and then block those same ads from view.
These issues in the supply chain became so egregious that some of the most prominent marketing executives in the world began calling them out. In 2017, Marc Pritchard, chief brand officer of Procter & Gamble, the world’s largest advertiser with a $7.5 billion annual advertising budget, described the digital display advertising supply chain as “ murky at best, fraudulent at worst.” A year later, Keith Weed, the chief marketing officer of Unilever, at the time the world’s second largest advertiser, referred to the digital media supply chain as the “ digital swamp.”
Advertising technology had become so complex and fragmented that Luma Partners, a consulting firm, invented the LumaScape chart to help marketers figure out all the companies that sold trading services, social media monitoring, verification services and more. This iconic chart was used by other clients to determine which service markets had the greatest opportunities for profitable acquisition or consolidation.
And that happened: Google went on a buying spree, and today their share of the key services in a marketer’s ad tech stack ranges from 40% to 90%. The company also holds a significant portion of the supply of all display ad space (20% is held by their YouTube unit alone). Despite the consolidation — or maybe because of it — things haven’t changed much since “Bad Men.” A recent report from the U.K. found that publishers receive only 51% of the money spent by advertisers to reach readers — leaving ad tech fees to swallow up most of the rest (and about 15% seems to just … disappear). Additionally, each of the 15 advertisers in the study appeared on an average of 40,524 websites, most of them classified as being “non-premium” ad inventory.
In other words, now Google is doing what it used to take thousands of ad tech companies to do: take premium ad dollars and somehow magically drain them of most of their value.
Facebook forces marketers to play its game while invading your privacy: That brings me to Facebook, another advertising-fueled business model and one unrivaled in its ability to target consumers at scale. Combined with its holdings Instagram and WhatsApp, Facebook has a 75% share of the social network market, and mostly thanks to Facebook’s 2 billion users, a 50% share of the total display ad supply. For any marketer that needs scale in social media, Facebook is the only game that matters. And since it has 100% control of access and pricing, marketers have to play by Facebook’s rules.
Some of the ways Facebook plays are deeply troubling. For example, due to its reliance on advertising for its business model, the company has perfected user experience design intended to increase “engagement” — that is, keep people scrolling because Facebook’s “inventory” consists of users’ time and attention. People are more likely to “engage” — like, comment on and share — information that creates a visceral emotional response. So highly partisan content, hate speech and conspiracy theories get upranked by Facebook’s algorithms, spreading far more quickly than dry but authoritative information. Ultimately, the ability to microtarget content — including ads — to more precise slices of the user base means the most inflammatory ideas — some designed to foment divisions in society — are virtually invisible to those who might counter them with the usual antidote to false or inflammatory speech: more speech.
Over time, Facebook even took on the creation of our ads: We would provide their ad sales team with hundreds of individual images, headlines and lines of copy, and the algorithm would optimize combinations that made consumers share, click or buy. Since advertisers pay more when a user takes some sort of action, Facebook has strong incentives to predict and then control users’ behavior. Frustratingly, we often had no idea what our “ads” even looked like or who saw them until a campaign was over.
In fact, advertisers have little visibility into or understanding of any of Facebook’s algorithms. We could communicate with our own customers (a “custom audience”), find customers like them (a “look-alike audience”) or describe what kind of customers we wanted to show our ads. Then the algorithm took over, compounding what we had described. Sometimes, the algorithm predicted what kind of customer you were from your activities on (or off) Facebook, even things about you that you might have thought were private.
We came to realize that the optimization combined with the predictive capability created the potential — and the reality — of bias, discrimination and exploitation. (Recently, a list of more than 500 advertisers, including Unilever, Coca-Cola, Levi’s, Starbucks and Ford, announced they would withdraw their advertising from Facebook — some included other social media platforms — for the month of July or longer, responding to a boycott campaign organized by civil rights groups. I see this is a welcome sign that advertisers are finally beginning to apply their leverage. And a series of civil rights audits — the final one published this past week — showed that the company still has a long way to go to address discrimination.)
Facebook’s voracious data collection practices — not only on its own platform but across many platforms and devices — and how it analyzed, packaged and sold access to that data earned them billions of dollars in government penaltiesfor violating users’ privacy rights. Eventually, thanks to Cambridge Analytica and other scandals, consumers came to understand how their privacy was being violated, and how they were losing autonomy and freedom. It took a while, because Facebook’s terms of service — which theoretically spelled all this out — were so hard to find, to read and to understand that it wasn’t really clear. They certainly weren’t clear enough to make you leave Facebook for another big social network that might have treated your privacy with more respect. And soon, there really weren’t any other big social networks.
To compete, advertisers must play by Google and Facebook’s rules: Today, Google and Facebook combined get just over 60% of every U.S. digital marketing dollar. As you might imagine, that kind of power and dominance in a market makes it difficult for advertisers of any size to negotiate.
To be fair, the ad community was complicit in some of this. We chased our customers in their inexorable migration to the internet for information, entertainment and socializing. We loved and needed the cost efficiency of highly targeted digital media and craved more reach. We wanted to influence our customers at every step of their “purchase journey” (a framework we invented to feel good about tracking consumers across devices and platforms).
So it’s been painful to realize that the very same capabilities we helped create — extracting information about users and their online behaviors; segmenting and profiling users for precision targeting; using algorithms to customize content to individual users; and designing user interfaces to increase time and engagement — are being used to distribute and amplify hate speech and propaganda, negate privacy and autonomy, manipulate and exploit behavior, and contribute to social isolation and addiction. No wonder consumers are losing trust in brands.
Regardless of how antitrust law deals with Google and Facebook, there’s one thing that’s crystal clear. Both companies have accrued so much power that they’ve had a lasting negative impact on advertisers, their relationships with their customers and consumers themselves.
We need a strong referee to change the advertising game and protect consumers: How do we solve the problems perpetuated by the dominance of Facebook and Google in the advertising market? It will probably take a system of solutions. The answer to many market problems is more competition and more choice, and I’m hopeful that antitrust enforcement will bring that about. A dedicated U.S. regulatory agency with specialized expertise and the agility it requires could also set and enforce new rules as a referee. Consumers also seem to be waking up to the harms of the digital platforms and may change some of their social behaviors. And given that 98% or more of these companies’ revenues come from advertising, I hope more of my advertising colleagues wake up to their power, too.
Lisa Macpherson is a senior policy fellow at the Washington, D.C.-based nonprofit Public Knowledge. She is a former consumer marketing executive.
|RecommendKeepReplyMark as Last Read|