Facebook Accused of Ignoring Government Warnings Before Mob Violence in Sri Lanka

Angry mobs of Buddhists in Sri Lanka last month attacked minority Muslims, burning mosques and killing at least one. Those riots appear to have been triggered in part by false stories spread on Facebook and WhatsApp. And despite efforts by governments and nonprofits to alert them to the mounting risk, Facebook is accused of doing next to nothing to remove clear incitements to violence in the weeks leading up to the attacks.

The sequence of events in Sri Lanka is detailed in a bruising new report by the New York Times, which threatens to undermine Facebook’s longstanding claim to be a force for good in the world. At their heart were allegations of a plot by Muslim Sri Lankans to sterilize the country’s Sinhalese-speaking Buddhist majority, supported by a false story on Facebook saying that police had seized 23,000 sterilization pills from a Muslim pharmacist in the town of Ampara.

In an episode eerily reminiscent of reactions to the Hillary Clinton Pizzagate conspiracy theory, those stories led a mob of Buddhists to storm a Muslim-owned restaurant in the town of Ampara, falsely claiming its food was laced with drugs. The exchange exploded into beatings, rioting, and mosque-burning. Video of those events was also uploaded to Facebook, feeding further violence and the death of a 27-year-old aspiring journalist.

Aside from the brutal violence itself, the most disturbing part of the Times report is the allegation that Facebook, which has no offices in Sri Lanka, ignored or deflected repeated attempts by government officials and nonprofit monitors to intervene in a growing storm of hatred. As early as October of 2017, Sri Lankan officials pleaded with Facebook to better police hate speech, hire more Sinhalese-speaking content screeners, and establish a direct point of contact with local authorities.

Get Data Sheet, Fortune’s technology newsletter.

Instead, Facebook insisted its content-flagging tool would be enough to alert the company to dangerous content. Members of a Sri Lankan group called the Center for Policy Alternatives did as recommended, repeatedly flagging posts including messages such as “Kill all Muslims, don’t even save an infant.” But “nearly every report,” according to the Times, was deemed to not violate Facebook’s standards. According to the Times, Facebook still has not filled around 25 positions for Sinhalese screeners that have been open since June.

The violence in Sri Lanka mirrors similar events in Myanmar, India, Mexico, and even the United States. They strike at the heart of Facebook’s utopian promise to connect people, showing that such connections can spread violent hatred as quickly as cute baby pictures.

Ethnic and religious resentments are not created by Facebook. But as the Times points out, Facebook’s core structure – including an algorithm that prioritizes content that gets the most engagement – may help foment outrage and tribalism. In nations with weak legal systems, citizens may be more likely to take justice into their own hands.

“The germs are ours,” as one Sri Lankan official told the Times, referring to the sectarian devisions in Sri Lanka, “but Facebook is the wind, you know?”

Australia competition regulator flags scrutiny of Uber Eats

SYDNEY (Reuters) – The head of Australia’s competition regulator said on Monday it will scrutinize contracts between Uber Technologies Inc [UBER.UL] and restaurants that use its food delivery app Uber Eats.

FILE PHOTO – A bag of donuts destined for delivery via Uber Eats is rushed to a driver from a kitchen in Sydney August 12, 2016. Picture taken August 12, 2016. REUTERS/Jason Reed

The Australian Broadcasting Corporation reported on Sunday several restaurateurs were disgruntled with contract terms they said left them, and not Uber, responsible for late deliveries despite being charged by Uber for the service.

“Certainly, we’ll have a look at it,” Rod Sims, head of the Australian Competition and Consumer Commission, told ABC radio on Monday when asked about the report.

“We have three bits of the law we can deal with here. One is business to business – are they misleading the people they’re dealing with? Two is: Are they engaged in unconscionable conduct, putting all the conduct together? And thirdly are the terms with which they work unfair? So, there’s a lot to look at there,” he said.

FILE PHOTO – An Uber driver takes delivery of bags of donuts destined for a customer via Uber Eats in Sydney, August 12, 2016. Picture taken August 12, 2016. REUTERS/Jason Reed

Uber, which is backed by Japan’s SoftBank Group, said Uber Eats is merely a “marketplace that connects restaurants with delivery partners” and that its terms and conditions are “consistent with Australian law”.

“Restaurants aren’t locked in, just like delivery partners and eaters. Uber Eats is just one option. If a restaurant does not wish to use the app, they do not have to,” an Uber East spokeswoman said in a statement.

Uber, which is preparing for a potential initial public offering in 2019, lost $4.5 billion last year and is facing fierce competition at home in the United States as well as a regulatory crackdown in Europe where governments have been leery of the company’s impact on the traditional taxi industry.

Reporting by Tom Westbrook; Editing by Daniel Wallis and Stephen Coates

A 15% Yield With A Discounted Buyout Backdoor

We went wandering down the back alleyways of MLP land this week and came across another very interesting buyout deal.

Southcross Energy Partners LP (SXE), a smaller midstream LP, is being acquired by American Midstream Partners LP (AMID). The deal is supposed to close in Q2 ’18, relatively soon. The buyout already has been approved by SXE’s common unitholders.

“Public unitholders of SXE will receive 0.160 AMID common units for each SXE common unit in a unit-for-unit merger, which is anticipated to have minimal, if any, tax recognition for such unitholders.” (Source: AMID site)

There’s currently a discount available for AMID via buying SXE units, which have a conversion factor of 6.25X into AMID units, post buyout.

At today’s price of $1.72 for SXE, this equals a $10.75 price for AMID, which is 7.33% discount:

This $10.75 discount current conversion price gives you a whopper of a yield, at 15.35%, and some upside potential. However, you won’t receive the upcoming distribution from AMID, unless the deal somehow gets closed by next Friday, 4/27/18, AMID’s next ex-dividend date.

AMID pays in the usual MLP Feb/May/Aug/Nov cycle, so, as an SXE converted AMID unitholder, your first payout wouldn’t be until ~8/3/18. AMID’S management is targeting 1.1x distribution coverage post-buyout. Unitholders get a K-1 at tax time.


Deal highlights – Among many other positive attributes, AMID’s management sees this deal as being immediately accretive to AMID’s distributable cash flow.

(Source: AMID site)

This makes sense – SXE used to pay $.40/unit quarterly but eliminated its payout in February 2016. However, SXE generated $2.39/unit in DCF in 2017. Add this to AMID’s $1.75/unit generated in 2017, and it gives you $4.14/unit vs. AMID’s total distributions/unit of $1.65 in 2017.

These numbers won’t be exactly the same in 2018, but that big cushion certainly gives management a lot of leeway moving forward and makes us believe that AMID should be able to at least maintain its current $.4125 quarterly payout and ultimately increase it, if the market rewards the company’s transformation efforts and bids up AMID’s price/unit.

AMID’s management has been selling off non-core assets but has made a series of acquisitions in order to transform the company into a more stable cash flow model.

(Source: AMID site)

The downside to this is that, as with any ongoing divestiture and acquisition process, the company will experience negative growth for a certain period, as it loses the earnings of sold assets, before the new assets start contributing to earnings. This is what happened in 2017:


Take a look at SXE’s current valuations – they’re the lowest we’ve ever seen in the midstream space – a price/DCF of just .72, a price/book of .17, and a price/sales of just .13.

It’s easy to see why AMID management went after this deal:

Analysts’ Targets:

The other plus about buying SXE at ~$1.72 (which equals a post-buyout conversion discounted price of $10.75) is that your cost is under analysts’ lowest price target for AMID, of $11.00. AMID is currently 26.47% below analysts’ average price target of $14.67.


Debt Leverage: AMID’s 2017 10K states that, as of 12/31/17, AMID’s total leverage ratio was 5.23X. However, we came up with a higher figure of 6.8X. One of the distinctions that management makes is not counting non-recourse debt in its presentations – they refer to “compliance leverage”:

(Source: AMID 2017 10-K)

We put together a table which compares our figures with AMID’s post-buyout projections. Management is targeting 4.5X debt leverage and $300M in EBITDA after the buyout, with a goal of reaching 3.5X within 18 months. This implies that they won’t be taking on more debt, and that EBITDA should ramp up to ~$385M within 18 months.


We rate SXE a buy, based upon the current buyout discount price, its very low valuations, and the upcoming, post-buyout yield, which will be well-covered. An additional plus is that a veteran energy investing firm with deep pockets, Arclight Partners, owns a ~27% of AMID’s units.

(Source: NASDAQ)

(All images by Double Dividend Stocks, unless otherwise noted.)

Disclosure: I am/we are long SXE.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Cyber Saturday—How Facebook and GDPR Propelled an Underdog to Victory at RSA Conference

Happy Saturday, dear readers.

Earlier this year I noted that Europe’s General Data Protection Regulation, or GDPR, would be a big topic of conversation at this year’s RSA Conference, the biggest hobnobbing affair in the cybersecurity industry. I could not have foreseen how scandal after data privacy scandal at Facebook would intensify the discussion.

At this year’s ever entertaining “innovation sandbox contest,” a startup competition and hallmark of the conference, a little-known, New York City-based concern called BigID capitalized on the zeitgeist. The company, which had just eight employees as recently as December (mostly engineers in Israel), pitched itself differently than the typical cybersecurity marketing spiel. There was nary a mention of “detection,” “defense,” or “artificial intelligence.”

“I’m with BigID and our big idea is that privacy matters,” said Dimitri Sirota, CEO and cofounder of the firm, taking the stage. He explained that his company’s technology indexes business’s private data, maps out the inter-relationships between databases, and helps identify what companies need to do to comply with data regulations in different parts of the world.

“Ours was understandable,” Sirota told me later on a call. “You didn’t have to have a PhD in computer science to get what we did. It was accessible to the audience and judges.”

Sirota’s clarity of thinking was apparent to me years ago, back when he was heading up the security business at CA Technologies. In 2014, he livened up a panel I moderated at an enterprise security summit. A couple years later, Sirota strolled into Fortune’s offices clad in a black leather jacket and told me his plan to build a business around data privacy and compliance. Looks like he had the right idea at exactly the right time.

“Big data is almost like this atomic collider—smash all this data together to get value from it,” as Sirota put it on our recent call. “No one has been thinking of stewardship or custody or management of that information.”

Now everyone is thinking about it. With British officials raiding the offices of embattled political consultancy Cambridge Analytica, Mark Zuckerberg bending the knee before congress, and GDPR set to go into effect next month, no story holds greater sway in techland. It’s no surprise BigID took home the crown.

Dream big and have a great weekend.

Robert Hackett


[email protected]

Welcome to the Cyber Saturday edition of Data Sheet, Fortune’sdaily tech newsletter. Fortune reporter Robert Hackett here. You may reach Robert Hackett via Twitter, Cryptocat, Jabber (see OTR fingerprint on my about.me), PGP encrypted email (see public key on my Keybase.io), Wickr, Signal, or however you (securely) prefer. Feedback welcome.

How to Build Custom Alexa Skills for the Amazon Echo

When Amazon first introduced developer tools that let people build stuff for Alexa, the company made a conscious decision to call these functions “skills” rather than apps. It was a subtle way of making Alexa seem capable, and also, suggesting to developers that building these skills would be a low lift. With just a “few lines of code,” Amazon promised, “you can build entirely new experiences designed around voice.”

Amazon says most Echo users in the US have tried these third-party skills at least once, but getting them to work can be tricky. Alexa’s voice skills often require super specific queries, and until Amazon started paying attention to the discovery process, taking the time to find new skills felt like a non-essential burden.

Now, Amazon has decided to make Alexa’s skills all about you: your dad jokes, your homework, your birthday. Yesterday the company rolled out a tool called Blueprints, which helps anyone—even non-coders—build custom skills for their Echos.



The announcement felt like a curious little leap in the world of virtual assistance. (Can you imagine Apple letting people program Siri to say whatever they want?) It also seemed like a gimmick, but a good one—one that could encourage more people to embrace skills.

The website for Blueprints lists 21 different skills, categorized by topics like “fun and games,” “home,” and “storytelling.” Annoyingly, the banner in the Alexa mobile app for Blueprints leads you to the mobile web, and at some point you’ll have to log into your Amazon account again. You also have to agree to Amazon’s terms for developers before you can make a new skill. Congrats! You’re a dev.

The Blueprints are not quite as customizable as you might think. Think of them more like Mad Libs for pre-existing Alexa narratives. Using the “family jokes” template, you can type in a joke like, “You know who would like a pet owl?” and then add the punchline, “Who?” You can also start a joke with “When I wake up” and teach Alexa to complete it with “Well you know I’m gonna be, I’m gonna be the man who wakes up next to you.” I’m not saying I did those things, because that would be incredibly corny. But you could.

You can also leave custom voice message for people, like an incoming houseguest or a pet sitter. But they’ll have to know the exact prompt to use in order to get that message. So you might end up texting your guest anyway, or leaving them a written note, and while you’re at it you might as well just mention that you’ve switched the cat’s food again or that the towels are in the third closet down the hall and at that point, why create a custom Blueprint for Alexa?

Still, I could see couples or families using this feature to leave private voice messages for each other. You can also program Alexa to pay you a personalized compliment.

There are a few restrictions: Alexa won’t swear. Try to program a custom response with profanity and the Blueprint will ask you to remove the offending word. (I tried twice.) The Blueprints also leave a lot of room for interpretation. You could, for example, create a question like “Who is the president?” and have Alexa respond that it’s Barack Obama. If you go back to asking Alexa a more generic question, one you haven’t personally programmed, like “Who is the leader of the US?”, Alexa will then correctly say Donald Trump.

And even after you’ve built these skills, you have to ask the questions verbatim. I built a Q&A asking whether the cat is hungry. (The answer is always yes.) By the time I processed that and other skills, I forgot the exact question I had written, and had to go back into my Blueprints to find the right terminology. Those verbal gymnastics can make it difficult to use the skills you’ve created.

But these personalized responses are by far the closest brush I’ve had with non-Amazon skills since I’ve been using Echo devices. Amazon says engagement with Alexa skills grew more than 50 percent from January 2017 to January 2018, and that there are more than 40,000 skills currently available. I have no reason to doubt this; this just hasn’t been the case for me.

I often use Alexa to access timers, music, and news; I’ve also used it to buy household goods, and at one point, to trigger my coffee maker. But I’ve never used Alexa to order a pizza, or to call an Uber—the third-party skills that are supposed to nudge voice into the future. Voice computing just hasn’t made sense when I’ve wanted to see the thing I’m ordering, or track the car as it’s pulling up to my house. But at the same time, I could see how adding custom-made, hyper-personal responses could drive even more of this kind of engagement with Alexa, especially since other smart home speakers don’t really do this.

My Blueprints test was brief, and yet it was one of the most fun experiences I’ve had so far with a virtual personal assistant at home. That’s probably because it enabled the “personal” part of that whole equation. The home bots may someday be our overlords, but at least we can program them to say ridiculous things in the meantime.

Hey, Alexa

DNC Lawsuit Against Russia Reveals New Details About 2016 Hack

The Democratic National Committee Friday filed a lawsuit against a broad slate of people and entities allegedly responsible for the 2016 hack of its email, phone calls, and more. But while the suit claims involvement from a host of headliners—Wikileaks, Julian Assange, Donald Trump, Jr., and Russia among them—its immediate importance lies in the previously unreported timeline it lays out.

While a rough outline of the DNC hack that rocked the 2016 election had previously been established, the 66-page lawsuit, first reported by The Washington Post gives exact dates for the first time. It also asserts coordination among a web of characters affiliated with the Trump campaign, Russia’s GRU intelligence service, and WikiLeaks.

“No one is above the law,” the suit begins. “In the run-up to the 2016 election, Russia mounted a brazen attack on American Democracy.”

The details of when and how that attack occurred, though, are more clear than ever—and may indicate that Russia’s plan to interfere in the US election predated its DNC intrusion.

According to the DNC lawsuit, Russian intelligence group Cozy Bear—the GRU-affiliated hacker group, also known as APT29—infiltrated the DNC network as far back as July 27, 2015, nearly a year before the leaks of the pilfered material began. The suit says that a second Russian group—Fancy Bear, the outfit that has recently tormented the International Olympic Committee as well—hacked the DNC’s systems on April 18, 2016. The DNC wouldn’t notice the presence of either until April 28, 2016, at which point it called in security firm CrowdStrike to help analyze and mitigate the damage.

The remedy was costly. The suit details the necessary fixes; the DNC had to “decommission more than 140 servers, remove and reinstall all software, including the operating systems, for more than 180 computers, and rebuild least 11 servers.” Between repairing and replacing equipment and hiring experts to manage the fallout, the bill came out to over a million dollars.

By then, of course, the worst damage had already been done. The DNC had been devastatingly compromised. The Russians had gained access not only to email systems but also to backup servers, VOIP calls, and chats. They were prepared to make off with “several gigabytes of data,” the suit says, a little over a week before the DNC even knew they were there.

The timeline from there has been a matter of public record. On June 14, the DNC first disclosed the hack. The following day, a persona going by Guccifer 2.0—only recently confirmed to be a Russian intelligence agent—claimed responsibility, leaking a 237-page opposition research report on Donald Trump in the process.

The leaks continued steadily from there, as the suit details. Guccifer 2.0 struck again on June 27, June 30, and July 6. On July 22, WikiLeaks took the wheel, releasing nearly 20,000 internal DNC emails. The following day, according to the suit, multiple DNC employees received an email that said: “I hope your children get raped and murdered. I hope your family knows nothing but suffering, torture, and death.”

The rest of the suit rehashes the connections that have played out in the press over the last several months, alleging Roger Stone, Paul Manafort, George Papadopoulos, and a host of Russians as ingredients in a collusive soup. But for close observers of Russia’s hacking efforts against the US in 2015 and beyond, it’s the timeline that provides the most valuable information.

That’s in part because of how it aligns with two incidents not mentioned in the suit. Many of the early leaks appeared on a site called DCLeaks, which went live in June 2016 but was registered on April 19, which the suit confirms was a day after Fancy Bear broke into the DNC. But the same group that registered DCLeaks had attempted but failed to register ElectionLeaks.com on April 12, nearly a week before the Fancy Bear hack.

The timeline strongly implies that Russia’s aim was to disrupt the election from the start, rather than a reconnaissance mission that rapidly escalated.

“They had already carried out the Podesta intrusion in March, and carried out a pretty large scale attempt to target the campaigns,” says John Hultquist, director of threat intelligence at security firm FireEye, referring to the emails of Hillary Clinton campaign chairman John Podesta, which were ultimately leaked a month before the 2016 election. That, combined with registering ElectionLeaks before the Fancy Bear break-in, “suggests they had this plan prior to even compromising the organization.”

It’s unclear how likely the DNC lawsuit is to succeed, especially in its efforts to hold Russia accountable in a US court. But its revelations shed light on one of the most impactful hacks of recent memory—and maybe the intentions of the country behind it.

Russian Hacks

3 Reasons Why Subscription Businesses Like Blue Apron and Trendy Butler Will Beat out Retail

The earliest subscription services such as newspapers and milk cartons have existed for decades without much attention. However, In the past five years, we have seen an explosion of innovative startups using the subscription business model to dominate their niche, beating out incumbents in the traditional retail sector.

This disruption is happening across a wide variety of industries, with over 2,000 business entities that operate under the subscription model in the United States alone. Meal kit subscription service Blue Apron went public earlier this year at a company valuation of nearly $2 billion. Birchbox, the New York-based startup that sells monthly boxes of beauty samples, is now valued at over $500 million. And Dollar Shave Club, the eccentric shaving brand, was acquired for over $1 billion, just to name a few success storiees.

Success achieved by early players in the industry has inspired a fresh wave of entrepreneurs to apply the subscription box model to new areas. Many of these hot startups are proving you can leverage big data and machine learning models to create extremely lucrative steady sources of recurring revenue.

One example of this is Trendy Butler, which offers a $65/month subscription box that comes with a combination of designer clothing (t-shirts, jackets, pants, etc.). The company, like many of today’s subscription services, uses an algorithm that collects your personal tastes and preferences (like sizes, styles, colors, etc.) to craft the perfect mix of outfits. It’s like Spotify, but for your clothing.

In analyzing the success of startups like Trendy Butler, I dug into why the subscription service model is likely to beat out conventional brick and mortar retail long term. Here are 3 reasons why:

1. Personalization at scale.

The assumption that entrepreneurs must operate under is that their customers are inherently trying to maximize their personal value while doing as little work as possible. In the case of shopping for clothing, we all want to look good, but many of us do not want, have the time, or frankly the talent to pick out the best outfits.

That’s what fueled Trendy Butler’s founders into realizing that we can use technology to completely rethink the way a shopping “experience” is delivered. Rather than randomly recommend products, predictive machine learning algorithms are used that take in a large data set (100+ points) of personalized information. As you expand the scope of the inputs to the algorithm, it gets smarter.

As recommendations are able to improve and become more personalized, the traditional brick and mortar way of doing business will simply not be able to keep up.

2. Predictable revenue sources.

Subscription business models also bring a sense of predictability that the retail industry has been lacking for decades. Since many stores cannot accurately forecast demand, there is often lots of waste, saturated product and overhead costs. These inefficiencies can often mean the difference between success and failure for many retailers.

Subscription companies circumvent these costs by doing much of the work behind the scenes. Additionally, most of if not all of their customers are paying monthly in exchange for a routine service/product. This is an extremely secure source of revenue that companies can develop over time

3. Establishing relationships with customers.

There is something special about opening your door to a new package even just once or twice a month. I personally love it when I receive a shipment from BarkBox and get to see my puppy light up with joy. The surprise in every subscription box is a unique opportunity for a company to delight their customers and provide a unique and memorable experience.

These touch points, which are rare in most other industries, develop customer loyalty. Over time, subscription box businesses tend to develop relationships with their customers because of the recurring nature of the interactions.

Retail is far more transactional as most of the instances are one time exchanges. With subscriptions, there is a constant need to interface with customers and continue the relationship.

As more and more companies infiltrate different industries, the only true competitive advantage startups will have will be in their ability to establish a strong and loyal community of backers. Building defensible relationships with customers is a great method of doing just that.

Time Warner CEO says AT&T merger needed to compete with internet titans

WASHINGTON (Reuters) – Time Warner (TWX.N) Chief Executive Jeff Bewkes on Wednesday defended his company’s planned merger with telecoms firm AT&T (T.N) as necessary to compete effectively for advertising with internet giants like Google and Facebook.

FILE PHOTO: Time Warner CEO Jeff Bewkes arrives ahead of arguments in the trial to determine if AT&T’s merger with Time Warner is legal under antitrust law at U.S. District Court in Washington, U.S., March 22, 2018. REUTERS/Aaron P. Bernstein

Bewkes told Judge Richard Leon, who will decide if the $84.5 billion deal may go forward, that the U.S. Justice Department was wrong to say that AT&T would be reluctant to license Time Warner’s TV and movie content to rivals, causing blackouts, in order to win over new customers to AT&T subsidiary DirecTV.

“I think it’s ridiculous,” said Bewkes, who has been CEO for more than 10 years. “If our channels are not in distribution we lose lots of money (from lost subscriptions and advertising).”

He said that “one percent, less than one percent, maybe two percent” of subscribers would drop their pay TV subscription because of a blackout, far below the 12 percent estimated by an economist for the government who testified earlier in the trial.

Bewkes argued it was in Time Warner’s best interest financially to license its television channels, which range from movies to CNN to sports, broadly online.

He said Time Warner had been hampered in innovating and advertising because it does not have the granular information about viewers held by pay TV and internet companies.

With digital advertising, Chevrolet, for example, can target car ads at people looking to actually buy a car, he said.

AT&T has said a key benefit of owning Time Warner is that it can take data about its 141 million U.S. wireless subscribers and 25 million video subscribers and marry it with Time Warner’s programming to enable advertisers to target TV ads.

Targeted TV ads, also known as addressable TV, have yet to go mainstream because they involve renegotiating carriage deals with programmers and distributors, said Brian Wieser, an analyst at Pivotal Research.

Targeted TV could represent more than $100 billion in revenue by 2030 for companies that offer it, according to an April Credit Suisse report, which called it “a largely overlooked benefit of the AT&T/Time Warner transaction.” The ads can be sold at triple the price of regular ads.

“The Google/Facebook duopoly has such a strong hold on the market, I think it’s important that there is healthy competition and that we aren’t just forced to invest in two places,” said Tim Villanueva, head of media strategy for Fetch, an ad agency focused on mobile, whose clients include eBay and Lululemon. He said he was interested in using the new platform.

Advertisers’ spending on TV ads in 2018 is expected to be around $70 billion, a 1.45 percent increase from three years ago, according to research firm eMarketer.


In cross examination, Justice Department lawyer Claude Scott pointed to efforts that Time Warner was already making to move into targeted advertising and online distribution, including contracting with tech companies, an apparent attempt to call into question the need for the megamerger.

Scott referred to Bewkes’ compensation package, noting that he would be leaving the company when the deal closed and that he owned more than 2 million Time Warner shares. AT&T’s deal for Time Warner is about a 35 percent premium over the market price.

The trial has seen a parade of witnesses testifying about how the merger would affect them. Executives from smaller pay TV companies talked about how important it was to have access to Time Warner’s movies and television shows.

The trial, which began in mid-March in U.S. District Court in Washington, is expected to wrap up this month.

Reporting by Diane Bartz; Additional reporting by Jessica Toonkel; Editing by Bernadette Baum and Rosalba O’Brien

Electric Scooter Startups in Battle with San Francisco 

This is a guest post by Applico CTO and Principal Tri Tran. Prior to joining Applico, Tri was the co-founder and CEO of Munchery. 

After years of public battles with Uber, San Francisco has learned some valuable lessons. 

This time around, three electric scooter rental companies – Bird, Lime and Spin – are trying to roll out their service in downtown San Francisco. But the city is fighting back. As I happen to live in San Francisco and work in downtown, I’ve been able to witness this battle first hand.

Fast Rollout

As a startup entrepreneur, I quickly recognized the strategies that these three companies have taken to maximize business traction in as short a time as possible:

  • Flood a certain limited geography (such as downtown San Francisco) with a lot of scooters.
  • Get as many people as possible using them, very quickly. Once certain critical mass is reached, perhaps leverage them and their loyalty to fight off any regulations.
  • The default is seeking forgiveness afterward instead of seeking permission prior to operating. Let the city take any actions it needs to, assuming it’s historically very slow to react anyway.

I would not be surprised if these three companies also employ “fake” users and have them ride the scooters around town to create buzz, and thus word of mouth referral.

The City Fights Back

To my surprise, the city worked quickly and City Attorney Dennis Herrera issued cease-and-desist orders to all three companies.

The city wants the scooter companies to take actions to:

  1. Keep users from riding scooters on sidewalks
  2. Keep scooters from blocking sidewalks when parked
  3. Ensure that riders use helmets

It’s all in the name of public safety. Until then, the city will impound these scooters and may issues fines of a minimum of $125 for each violation. That is unless the companies can abate the problem within 30 days or prevail in an appeal hearing.

What Will Happen Next

It is not clear what these companies would do next, but here’s my take:

  • These three companies don’t have the operations to meet the city’s demand for the above described points
  • Limiting riders from sidewalks and ensuring unused scooters not block sidewalks is too tall of an order

Technically, it’s entirely possible to track riders on where they ride and whether they had left the scooters on sidewalks. They can even issue fines to riders who disobey such rules. But these rules would be a direct conflict to the convenience of ride-wherever and park-wherever, and thus would greatly reduce the attractiveness of using their scooters in the first place.

So what should they do? San Francisco has clearly established that it can move fast to regulate AND has created a repeatable process to impound these scooters.

Not complying will simply be too costly. More importantly, San Francisco also provides a model for all other cities to copy if/whenever the service rolls out to their cities. That’s bad news for Bird, Lime and Spin.

Ultimately, the service that these scooter companies provide is convenient, but not nearly as convenient as Uber was when it first arrived and replaced the painful taxi experience.

Additionally, the immediate public disruption of a horde of unfamiliar vehicles taking over sidewalks is much more apparent. Scootering on the streets has its own dangers as well when mixed with automobile traffic. The bike accident rate in San Francisco hasn’t changed much in the past few years. Relatively few people will brave their lives for that convenience.

Thus, these scooter companies will not have similar leverage nor political power from their small user bases to what Uber had when it fought against regulation attempts.

Based on the above, my prediction is that this service will go down into history as a fail, at least in major U.S. cities. It’s not necessarily a bad thing as these entrepreneurs will have learned a lot from the experience. They may find better, more sustainable businesses as a result. Cities are still grappling with what the future of transportation looks like, as are entrepreneurs. Scooters may not be it.

Majority of divisive Facebook ads bought by "suspicious groups" – study

(Reuters) – Most of the political ads about divisive issues that ran on Facebook Inc before the 2016 U.S. presidential election were sponsored by “suspicious groups” with no publicly available information about them, according to a study released on Monday and based on a database of five million ads on Facebook.

FILE PHOTO: Silhouettes of mobile users are seen next to a screen projection of Facebook logo in this picture illustration taken March 28, 2018. REUTERS/Dado Ruvic/Illustration/File photo

One in six of those groups was linked to Russia, according to a University of Wisconsin-Madison study here, and the identities of the rest of the 122 groups that are labeled “suspicious” are still unknown, an indication of the influence of “astroturf” or shell companies in U.S. politics.

Over a quarter of the suspicious ads mentioned Donald Trump or Hillary Clinton, two of the presidential candidates in the election, and 9 percent expressly advocated for or against individual candidates.

Most other ads deliberately avoided mentioning candidate names while still getting the message out by doing things like supporting policies pushed by candidates, Young Mie Kim, the lead researcher said. 

The researchers labeled suspicious ad-buyers as groups with pages that have been inactive, inaccessible, removed or banned by Facebook since the election and there was no information available publicly about them.

Project DATA, the research team, also found that voters were also disproportionately targeted in swing states like Wisconsin and Pennsylvania with ads that focused on issues like guns, immigration and race relations.

Facebook Chief Executive Officer Mark Zuckerberg has announced a crackdown on who buys ads about divisive issues, saying this month that the company would require every such advertiser to confirm their identity and location.

Reporting by Shariq Khan in Bengaluru; Editing by Bernard Orr