The Low-Down: 2019-09-29

Oct 5, 2019

Amazon Ordered 100,000 Electric Powered Delivery Vans From A Startup

Which suggests it is trying to appear serious about the environment because it is really, really serious about becoming a major competitor in delivery.

That Amazon is a significant investor in the company may perhaps lessen the reputational benefits from this announcement. JL

Andrew Hawkins reports in The Verge:

Rivian is a new name in the electric vehicle industry, having only debuted its pickup truck and SUV in 2018. Originally founded to compete with Tesla’s first car, Rivian eventually pivoted toward a more action-adventure customer segment. Amazon led a $700 million funding round last February. In April, Ford announced a $500 million investment in Rivian that the companies said would result in a new electric vehicle to be sold by the auto giant.

It might not taste any better - or worse - but it helps alleviate labor shortages and high turnover in restaurant staffing. JL

Melissa Hellmann reports in the Seattle Times:

A worker places a pre-made 16-inch circle of pizza dough into the module. Artificial intelligence uses mathematical formulas to process and assess the size, dimensions and placement of the pizza on the belt, and to adjust the lighting so a computer-vision camera can properly see it. "Food assembly is labor intensive and tedious. It’s hard to keep trained workers who make a consistent product.” The automated system requires one worker to place the dough into the assembly line and replenish the toppings, (addressing) the shift from dine-in restaurants to takeout, where a high volume of food is needed in a short time.

People who did not grow up using the internet and came to it later in their lives are less comfortable communicating mood and tone.

So their interpretation and use of capitalization, punctuation, grammar - and emojis - can be very different from that of digital natives. JL

Abigail Weinberg reports in Mother Jones:

People who started using computers later in life think of internet communication as: “this is a style you use when on a device” “this is how you talk on the internet” Whereas people who grew up using them use different styles: “this is how you indicate irony” “this is playfulness” “this is passive aggression” “this is sincerity” etc. I cannot overstate how much older people have NO IDEA they could possibly be communicating tone of voice, and how much younger people have NO IDEA everyone isn’t constantly communicating tone of voice

Delivery companies dominate certain markets and exercise predatory pricing, in addition to imposing new web marketing techniques that drive customers away from the restaurants and to delivery company sites.

The abuse complaints have become so numerous that government agencies are investigating. JL

David Yaffe-Bellany reports in the New York Times:

Restaurant owners complain about the high commissions charged by Grubhub and other delivery apps, which range from 15 to 30%. Grubhub charges restaurants fees for phone calls that do not result in orders. Grubhub has acknowledged it uses an algorithm to determine whether a phone call resulted in an order rather than reviewing the transcript of every call. Although most customers place delivery orders by tapping their smartphone, Grubhub allows users to make phone calls through the app, and Yelp listings often include a Grubhub phone number alongside a restaurant’s number.

Cable cutting has been rampant anyway, but this explains some of the economic basis for consumers' seeking alternative channels for entertainment. JL

Jon Brodkin reports in ars technica:

A Consumer Reports analysis of cable bills found that companies add $37.11 per month in fees to the average bill, raising consumers' actual costs way above the advertised prices. The $37.11 "in fees created by the cable industry" add 24% to the average base price of $156.71 a month. That doesn't include another $13.28 in government-related taxes and fees, which raise prices even higher.

Incomes for most Americans have stagnated, while car prices have risen, meaning people are having a harder time paying for the cars that symbolize middle class lifestyles.

Car makers and auto dealers have responded by offering loan terms that extend past the probable life of the car being bought, effectively cannabilizing future sales. This is not a sustainable economic model. JL

Ben Eisen and Adrienne Roberts report in the Wall Street Journal:

A third of auto loans for new vehicles taken in 2019 had terms longer than six years. That means monthly payments past when the brake pads give out and beyond when the car gets traded in. A decade ago, that was less than 10%. Incomes have risen at a sluggish pace in the past decade, but car prices have grown. New technological and safety features have made the most basic cars more expensive.Lending makes it possible for more Americans to procure a vehicle by spreading the debt over longer periods. Wall Street snaps up these loans, which are bundled into bonds. Dealers now make more money on the loans than on the cars they sell.

Having grown up with ecommerce, the digital natives find it commonplace. The personal touch offered by in-store shopping is more exciting and genuine - and it is increasingly enhanced by technology - which makes it more accessible to them.

But it is also the case that as urban dwellers - primarily - with little storage space or interest in inventory, they need what they need and they want it immediately, which in-store retail can provide. JL

Janet Stilson reports in Ad Week:

43% of millennials and Gen Z are likely to increase their in-store shopping this year, followed by Gen X (29%) and baby boomers (13%).Emerging tech in retail stores is most attractive to millennials (50%) (But) “the purpose of millennials’ in-store purchases may be more need-based than want-based. Shopping when they need something and being able to take a product home with them is important.” “It’s not about price and promos. It’s about an authentic experience and feeling that [consumers] matter. The days of mass market are rapidly dwindling. You can’t be everything to everybody. But you can be amazing to a niche.”

Is the US Doing Enough To Maintain Its AI Competitiveness?

Apparently not. JL

Edd Gent reports in Singularity Hub:

Governments around the world are pouring money into AI research, developing detailed AI strategies, but the US has been slow to follow. The US government announced the 2020 budget includes $1 billion of funding for non-military R&D in AI. But industry insiders said it was insufficient if the US wants to maintain its dominant position in AI; (it's) far less than the tens of billions China is spending. South Korea pledged $2 billion to its strategy. Matching the synergy between the state and private sectors powering China’s rise is unlikely to fly in the US, but the US can restore R&D funding to historic levels, train new talent, retain foreign talent, boost tech adoption in federal agencies and form tech alliances with like-minded countries.

Any way you measure it, that is a lot of suds. JL

Gary Stoller reports in Forbes:

The craft brewing industry contributed $79.1 billion to the U.S. economy and was responsible for nearly 560,000 jobs in 2018. Of 559,545 jobs provided by craft brewing, 150,000 were located at the breweries. Workers for craft breweries received more than $5 billion in wages and benefits during the year. Craft breweries in California, which has more breweries than any other state, made the biggest economic contribution — $9 billion. Rounding out the Top 5 contributors were breweries in Pennsylvania, $6.3 billion; Texas, $5.1 billion; New York, $4.1 billion, and Florida, $3.6 billion.

These apps harness humans' growing dependence on tech and their growing demand for digitally determined certainty in an increasingly uncertain world. You get what you pay for. JL

Arielle Pardes reports in Wired:

A crop of new apps have arrived, harnessing the powers of artificial intelligence, to offer relationship advice. The app parses text conversations to estimate the compatibility and personality of the individual you're chatting with, scoring openness, emotional control, extraversion, agreeableness, and conscientiousness. The iOS version has a singular function: to suggest the probability, on a 100-point scale, that the contact is romantically interested. Whether a text analyzer reveals anything real or not, using one seems to offer a false sense of predictability and a semblance of control over otherwise messy human relationships.

With ecommerce growing more slowly, Amazon is rolling out more bricks and mortar stores intended to diversify its revenue sources - and perhaps deflect the growing criticism of its digital domination by paying local taxes and creating more jobs. JL

Esther Fung reports in the Wall Street Journal:

Amazon is advancing a plan to open a chain of U.S. grocery stores in Los Angeles, Chicago and Philadelphia. It has signed more than a dozen leases The first few stores are in dense suburban locations, outside urban cores and cater to middle-income consumers. Apart from prepared foods, they will stock mainstream groceries. (At) Amazon Go stores customers can grab ready-to-eat food and groceries checkout-free. It also has four Amazon 4-star stores, which stock products rated 4-stars on the Amazon site, and 18 Amazon Books stores. The new grocery chain isn’t intended to compete with the company’s Whole Foods chain.

AI is being deployed with greater frequency. There is no legal obligation to notify consumers when they are being evaluated - for whatever reason - by an algorithm rather than a human.

There is growing concern that without any sort of transparency or possibility of appeal, these systems could perpetuate inaccuracies and inequities. JL

Dalvin Brown reports in USA Today:

AI can match employees who have the ideal skill sets for a specific work environment with employers who may be too busy to have humans screen candidates."Meaningful bits" of information include "how a person will work, how long they will stay, will they be a top sales performer or a high-quality worker." It is (also) touted as a faster and more accurate assessment of a potential borrower as it sifts through tons of data in seconds.AI is used in predictive analysis, in which a computer reveals how likely a person is to commit a crime. The technique has faced scrutiny over whether it improves safety or simply perpetuates inequities.

There is simply too much money involved now for assurances of efficacy to be taken on faith. JL

Steve Lohr reports in the New York Times:

Publishers routinely complain that the opaque nature of the digital ad pipeline is inefficient and expensive, with middlemen taking an outsize share of ad spending. Newspaper and magazine publishers, by some estimates, collect only 30 to 40 cents of every dollar spent on their ads online, compared with about 85 cents in the pre-internet days.The companies called for more visibility into where each dollar is spent in the online advertising supply chain. They committed to standards and practices for sharing data on fees and authenticating content.

Big tech companies have done a masterful job of optimizing the opportunities presented to them by financial innovation and government-funded research. But the outcomes have not been what was anticipated, with benefits mostly accruing to a few.

The question is whether through a combination of social engineering and economic concentration the inevitable logic of scale and power has become foreordained, or whether those same forces will tip if they become unsustainable. JL

Mariana Mazzucato reports in Project Syndicate:

With personal data becoming the world’s most valuable commodity, will users be the platform economy’s masters or its slaves? By exploiting technologies that were originally developed by the public sector, digital companies have acquired a market position that allows them to extract massive rents from consumers and workers. The widespread use of tax arbitrage and contract workers is eroding the markets and institutions upon which the platform economy relies. Innovation does not just have a rate of progression; it has a direction. Do we really want to live in a society where our innermost desires and manifestations of personal agency are up for sale?

A Tesla Police Car Nearly Ran Out Of Power During A Chase. Not Car's Fault, Sort Of

Technically, it wasn't the car's fault due to the timing of the incident and an incomplete charge, but it does underscore the ongoing issue electric cars have with the perceptions of limited battery life risk.

But this will all be irrelevant once we get to telepathic travel... JL

Phil Helsel reports in CNBC:

A police officer in the Tesla Model S, which the department bought in March, became involved in a police pursuit but radioed dispatch to say that the electric vehicle warned that it had only 6 miles of battery life left and that he may not be able to continue in the chase.The Tesla was not fully charged at the beginning of the officer’s shift at 2 p.m. and that the pursuit didn’t begin until around 11 p.m. The department had other police vehicles behind the Tesla to take over the chase, and the California Highway Patrol was also responding.

Marketers have to go where the consumers are, even if means more subtlety and lesser impact. JL

Kelsey Sutton reports in Ad Week:

Viewers are spending more time watching programming on streaming services, but since many of those OTT offers either don’t carry ads or offer ad-free tiers, brands are finding another entry point through product placements and brand integrations.Ultimately, though, brands want to be where consumers are. As viewing habits continue to shift to streaming, finding a way to make it onto emerging platforms will only become more essential.Subscribers to Hulu’s ad-free plan pay twice as much as limited ad customers to avoid commercials.

If the data are skewed, so will be outcomes, as will solutions based on those data. JL

Linda Nordling reports in Nature:

One of the best predictors for length of stay was the person’s postal code.The postal codes that correlated to longer hospital stays were in poor and predominantly African American neighbourhoods. People from these areas stayed in hospitals longer than did those from affluent, white areas. Perhaps they were less likely to be prescribed the drugs they needed. Machine-learning algorithms could help scientists to tease out which people are likely to respond best to which treatments. But this hinges on the data that are available for these tools to learn from, and those data mirror the unequal health system we see today.

Using attacks on tangible assets to protect intangible assets by hitting them where it really hurts. Which is not all that different from where more traditional capitalists park their money. JL

Catalin Cimpanu reports in ZDNet:

Besides imposing new sanctions on the Internet Research Agency and the private property of six of its employees, Treasury officials today also went after the private possessions of Yevgeniy Prigozhin, the IRA's founder. US officials imposed sanctions on companies Prigozhin uses to manage three planes and a yacht. "By continuing to service these aircraft and the vessel, providers of such services run the risk of facilitating or supporting Prigozhin's nefarious activities and may also be subject to future sanctions."

These two events are connected by a fundamental and growing lack of trust in Facebook.

This is especially true of its management whose response to criticism appears to have been one of superficial gestures overlaying a deeper contempt for national governments and supranational organizations. What goes around comes around. JL

AnnaMaria Andriotis and Peter Rudegeair report in the Wall Street Journal and Adam Satariano reports in the New York Times:

Europe’s top court said an individual country can order Facebook to take down posts, photographs and videos and restrict global access to that material. Visa, Mastercard. and other financial partners that signed on to help build and maintain the Libra payments network are reconsidering their involvement following a backlash from U.S. and European government officials.

Ok, perspective is important: ecommerce growth is slowing to a mere 25% annually...

But since China is the largest retail market in the world, trends beginning there suggest opportunities both in that country and globally. JL

Lambert Bu and colleagues report in McKinsey:

Online retail sales are expected to swell to $1.5 trillion, representing a quarter of China’s total retail-sales volume, and more than the retail sales of the ten next largest markets in the world. But growth of online retail sales is cooling, dipping from the 40 and 50% annual rates seen in the early part of the decade to 25% in the past few years.An increasingly crowded marketplace (is) pushing up the cost of acquiring new customers and retaining them. (Alternatives include) digitally powered physical retail to tap into omnichannel behavior; small-town youth, the next pocket of customer growth; and monetizing social attention and direct-to-consumer channels

Federal government funding of R&D had declined. Corporate R&D is protected by patent law and by increasingly aggressive litigation to sequester intellectual property, even when its derivations are mixed. A decline in support for public universities and restrictions on immigration further stifles the spread of innovation.

The good news is that these problems are largely the result of misguided public policies which can be fixed. JL

Noah Smith reports in Bloomberg:

U.S. federal research funding has stagnated in recent years. The U.S. spends less of its total economy on research and development than Germany, Japan and South Korea. Private spending (is) more product-focused and less generally applicable, and it also spreads to the rest of the economy more slowly thanks to patent law and corporate protection of trade. Increasing the supply of spots at good universities to ensure access for the low-income who all too often miss out on college. Skilled immigrants also boost the rate of innovation.

People Can Now Type On Phones Almost As Fast As On Keyboards

It helps to be under age 19 - and to use your thumbs. Word suggestions actually slow people down because smartphone users have learned they have to think about the suggestions since they're so often wrong. JL

Matthew Gault reports in Motherboard:

A good typist can type 100 words per minute (WPM) on a desktop keyboard, but most only type 35-65 WPM. People using two thumbs can achieve typing speeds averaging 38 WPM on smartphones.Age and digit placement were the two best predictors of typing speed on a phone. 74% of people (who) use two thumbs achieve speeds of 50 WPM. Using a single index finger or single thumb drove averages down to 35 WPM. Age was the stand out metric: people between 10 and 19 typed 10 words faster per minute on their phones than people in their 40s. "Word completion helps, but the time spent thinking about the word suggestions outweighs the time it takes to type the letters, making you slower overall,”

Monitoring behavior from swallowed sensors allows for more granular management of performance and assurance in outcomes. Which could presage where humans are headed. JL

The Economist reports:

Dairy cows have gone digital. A sensor that can be swallowed lodges inside one of a cow’s four stomachs, and stays there for the rest of the animal’s life, monitoring body temperature, movement and stomach acidity, uploading the results when the cow is near a wireless detector.When fed to machine-learning algorithms data can detect when animals are in heat, and spot early signs of calving 15 hours before it happens. They can identify diseases several days before they become obvious to humans, allowing a 15-30% drop in antibiotic use. A new sensor will add the ability to monitor digestion.

UPS decided on a strategy making it the 'anti-Uber.' Instead of fighting authorities and regulators every step of the way, it cooperated by not only seeking to help develop the process but by sharing its own insights to improve that process with real time data.

The result of this cooperative and collaborative stance has been a series of certifications that have given the company unique first mover competitive advantages. JL

Darrell Etherington reports in Tech Crunch:

UPS has been working with the FAA throughout its development and testing process for drone deliveries. It became the first commercial operator to perform a drone delivery for a paying customer thanks to an exemption from the government. The certification has no limits on the size or scope of operations, a certification, no other company has attained. (It) permits the company to fly an unlimited number of drones with an unlimited number of remote operators. This enables UPS to scale operations to meet customer demand. It will start by building out its drone delivery to hospital campuses in the U.S., and then to other industries.

Bringing together the two primary forms of artificial intelligence could make the convergent result more powerful - and interesting - by merging logic with experimentation and learning. JL

Arthur Miller reports in Scientific American:

There are two main sorts of AI. Symbolic machines are programmed to reason as humans do, working through logical steps to solve specific problems. Artificial neural networks are inspired by the human brain and need less human input. Their forte is learning. With the proper software they can each do a little of the other. A new artificial neural network can learn to form relationships in raw input data and represent it logically to build flexible reasoning. Combining the two systems could lead to more intelligent solutions and also to art, literature and music more accessible to human audiences while also being experimental, challenging and fun.

This is a reflection of two new realities: anti-trust pressure from government regulators concerned with big tech companies' anti-competitive behavior which both customers and authorities find increasingly unacceptable.

And it acknowledges the new market conditions Apple's recent design, pricing and marketing decisions represent, which is that dominance based on insularity is harder to maintain as consumers become more adept at using technology to meet their own needs, not just the seller's. JL

Jon Porter reports in The Verge:

Apple will allow Siri to use third-party messaging apps by default. It will later expand to phone apps for making calls. Siri will learn which service you prefer for each contact and default to that for communications. The advantage Apple’s apps gain over competitors by being the preinstalled defaults on Apple devices is a practice starting to attract the attention of regulators.Apple is already taking steps to allow Siri to work better with third-party music apps.

Uber. WeWork. Slack. All are companies in which Softbank was a lead investor. And all are businesses as well known for their financial losses as for their disruptive potential. Vision is wonderful as long as it leads somewhere economically sustainable. JL

Peter Eavis and Michael de la Merced report in the New York Times:

SoftBank’s critics said its investments have poisoned the ecosystem for young companies by encouraging founders to take excessive risks with little regard for building businesses that can last through the ups and downs of the economy. Mr. Son has acknowledged that the businesses his company invests in need to become financially sustainable more quickly. SoftBank’s approach imposes few constraints on founders of the companies it invests in. (And) they lack discipline because he lavished hundreds of millions or billions of dollars on them before they figured out what customers wanted and how to turn a profit

Digital transformations are essential in an era when every enterprise is a tech company. But they are also disruptive, expensive and risky. Three of the most iconic US corporations - GE, Ford and Procter & Gamble - each with vast resources and expertise, attempted such initiatives and failed to achieve their goals. In every case, the CEO was forced to leave.

Leadership is more important to transformational success than is the technology itself. Planning, coordinating, collaborating and inspiring are among the intangible assets whose effective implementation is crucial to realizing digital optimization. JL

Blake Morgan reports in Forbes:

70% of digital transformations fail. Digital transformation is a complicated and risky endeavor.A failed digital transformation doesn’t spell the end of a company, but it can be costly in lost money, resources, time and credibility. A digital transformation for transformation sake only won’t be effective. It must consider all outside factors and be tightly tied to strategy. To be successful, digital transformation efforts need to be integrated with the rest of the company. (But) digital transformations are often done best with a handful of passionate people leading.

Supreme Court Says FCC Can't Bar States From Crafting Own Net Neutrality Regulations

The Court said the FCC was within its rights to eliminate net neutrality at the Federal level, but exceeded its bounds when attempting to prevent states from doing so.

Given that 22 of the 50 states wanted to keep net neutrality, this could mean that almost half the country could return to net neutrality. JL

Tony Romm reports in the Washington Post:

A federal appeals court on Tuesday affirmed that the Federal Communications Commission acted lawfully when it scrapped the U.S. government’s net neutrality rules in 2017, but the ruling opened the door for state and local governments to introduce their own regulations designed to treat all web traffic equally. It overruled an effort by the FCC to block states from adopting open-internet protections of their own.

Probably a transitional phase before AI generates the thoughts and emotions that generate communications. JL

Kyle Wiggers reports in Venture Beat:

The model generates high-fidelity speech with “naturalness” but it’s highly parallelizable, meaning it’s more easily trained across multiple machines compared with alternatives. A convolutional neural network learned to produce raw audio by training on speech with encoded phonetic, duration, and pitch data. Discriminators distinguish among real speech and synthetic speech. Researchers evaluated performance quantitatively. The best-performing model achieved comparable scores to baselines

Practical applications at the individual level, which may be more disruptive collectively at the macro level than at the institutional level. JL

Kurt Cagle reports in Forbes:

Augmentative AI can be used to do parts of a task that were high cost for little value add. Augmented AI personalizes that AI (so it) becomes a friend and confidant, not just a tool. Remembered history is a good description for how most augmented AIs work. Augmentation is likely to be the way most people will interact with artificial intelligence systems. The effects will be subtle - steadily improving the quality of the digital products, reducing the number of errors, and reducing the time to create intellectual works - art, writing, coding. They raise intriguing ethical questions, such as “if an AI is used to create new content, to what extent is that augmenting technology actually responsible for what’s created?”

And it isn't even a major election year. Yet. JL

AFP reports via Yahoo!:

Organized efforts to manipulate social media and public opinion are being carried out in at least 70 countries by government agencies and political parties, researchers said. Manipulation efforts have doubled over the past two years and are being used by both democratic and authoritarian governments. Facebook remains "the platform of choice" for social media manipulation, with evidence seen in 56 countries. "Sophisticated state actors" from seven countries are working outside their borders on global foreign influence using Facebook and Twitter, identified as China, India, Iran, Pakistan, Russia, Saudi Arabia and Venezuela.

Whether Sprint customers are as enthusiastic about this development as is management will be interesting to learn. JL

Patrick Kulp reports in Ad Week:

30% of online chats are now conducted solely by machine-learning bots, up from 4% at the beginning of 2019. “Chat use help customers resolve their questions at a good experience rate and allow human reps to handle the more complex questions.” Sprint trains the bots on data from actual conversations between customer service reps and callers, an ongoing process that allows bots to hone their skills. The company has also been using machine learning to adjust the design flow of its website through A/B tests and targeted personalization based on which types of experiences most often leads to sales for a given profile of consumer.

Manufactured 'natural' content is beginning to outweigh the amateurish personal posts for which much of social media was invented and which attracted vast numbers of followers.

The question is whether users will simply shrug it off as part of the digital media landscape - or whether it will drive them away, killing another golden media goose. JL

Georgia Wells and Jeff Horwitz report in the Wall Street Journal:

In its rise from a picture-sharing app to celebrity endorsers and influencers, Instagram has spawned a behind-the-scenes industry that brokers content and followers to sell sponsored posts. Meme factories are reshaping how Instagram content is made: More shareable content yields more followers, which makes more money. 75,000 people this summer flooded the Anaheim Convention Center in Calif., to learn about making money on social media. Done well, the marketing appears more authentic than campaigns on TV. Done poorly, it comes across as mass-produced, tarnishing the platform’s appeal and driving away users.

Massive offshoring of white collar jobs - once predicted with emphasis akin to certainty - never happened.

The same lessons may apply to automation and artificial intelligence. But if any lesson has been learned, it is that there is no certainty on any such predictions. JL

Ben Casselman reports in the New York Times:

“The companies that started the offshoring trend were based in Manhattan or the West Coast, in very high-cost places, and they realized there are a lot of other places in the U.S.." The evolving on offshoring could carry lessons for automation. Economists saw parallels between offshoring and automation: Both threaten one set of jobs but should make the overall economy more productive, creating new opportunities requiring different skills.“The lesson is, change is evolutionary, not revolutionary. There’s not mass movement of jobs anywhere. And because it’s a slow change, it gives people and companies a chance to adjust."

AI Holds Great Promise For Medical Dianoses But Research Standards Must Improve

Hype could undermine potential clinical gains. JL

Edd Gent reports in Singularity Hub:

Medicine is one of the hottest fields when it comes to applying AI to real-world problems, in particular using deep learning systems to detect disease in medical imagery. There have been promising early results, in particular work on eye disease, but there’s been skepticism around whether results would translate to the clinic. Deep learning holds huge potential in medical diagnosis, but the authors say there needs to be more standardized and rigorous approaches to testing them—otherwise that promise will be undermined by over-the-top claims from dubious studies.

Knowledge of which will help digital mapping provide better directions. JL

Diana Gitig reports in ars technica:

Navigating requires the coordination of the neural processes mediating multi-sensory integration, contextual learning, and spatial cognition, so isolating which aspect fails as we age has been difficult. This study indicates that the problem lies, not in navigational strategy, but in cue-processing. Visual cues for navigation include brick walls of a building or a hedge border. Landmarks (are) like a dead tree trunk or broken street lamp.

Media companies want to own data to capture the additional revenues available. But for clients, access to the best data. not the most convenient or bundled, is almost always the optimal solution. JL

Eric Oster and Shoshana Wodinsky report in Ad Week:

All data is not created equal. “Media agencies owning data companies seems like an outdated model that presents a conflict of interests. If your media agency holding company owns a specific data company, aren’t they also incentivized to recommend that to clients over other, potentially better-fitting solutions?” If a brand is smart, they’ll want to own their own data strategy, rather than taking an out-of-the-box solution.

But not surprisingly, before achieving that unlikely goal, he wants - more characteristically - all his customers' personal information. JL

Amir Efrati reports in The Information:

He has snapped up properties in cities around the U.S. and in London, Singapore, South Korea and other countries. He is turning the properties into kitchens he can rent out to restaurants that want to make food for delivery. He hopes to cash in on on-demand food delivered through services like DoorDash and UberEats rather than at their dine-in locations. But Kalanick’s ambitions include trying to access restaurant customer data by tying his startup’s software into the restaurants’ point of sale systems allowing the company to reach the customers directly and advertise to them.

There is more money seeking investments than there are good ideas. And the lengthy bull market has dulled the senses of those who should be evaluating risk more carefully. JL

Esha Day and colleagues report in Bloomberg:

Unprofitable companies are raising money in initial public offerings at the fastest pace since the dot-com bubble. “Today’s technology is better than anything that was around in the Pets.com day. Logistics weren’t as well developed in the early internet era; e-commerce wasn’t as efficient. There was a lack of infrastructure.” True disruptors, like Amazon or Netflix , can deliver incredible payoffs. Would-be disruptors that fail to transform their markets can wind up like a wilted tulip.(But) “You know things are ridiculous but don’t want to miss out, so suspend your disbelief. But eventually bubble investors realize the emperor has no clothes.”

Compensation, the amount of time a person has worked in an organization (the less time employed, the more likely to leave) and lack of managerial responsibility are the most significant drivers of employee turnover.

From the standpoint of behavioral economics management, the challenge is figuring out how to systemically optimize the variables that impact those decisions in order to produce desired outcomes for as many of those involved as possible. JL

Matthew Castillon reports in Price Economics:

Employee turnover is expensive. Not only does business pause when an employee leaves, but you also have to spend valuable time finding that person’s replacement. 32% of employees quit or are terminated every year. For hourly employees, turnover rates level off at the $25-per-hour wage. After that wage, additional pay has a minimal effect on turnover. Newer employees have a higher turnover risk than tenured employees, particularly within the first year. Employees who manage others have a lower turnover rate than those without direct reports.

Alexa's Real Competition Is Still Your Smartphone

Why the failure of Amazon's Firephone failure still rankles. JL

Dieter Bohn reports in The Verge:

Amazon clearly has ambitions to make Alexa the leading platform for ambient computing. But to do that, it needs more ubiquity than it can achieve right now. Amazon is probably never going to get there, because Apple won’t allow it. The phone problem is Alexa’s biggest issue. It’s the easiest explanation for why any third-party technology is bad: without deep system-level access, third-party accessories and ecosystems end up failing on phones all the time. Alexa on phones is often more capable than it gets credit for — but it’s still nowhere near as good nor as integrated as the Google Assistant or even Siri.

Could? JL

Ines San Martin reports in Crux:

Speaking to executives from some of the world’s most important tech companies, Pope Francis said that technology needs “both theoretical and practical moral principles,” and that advances in fields such as artificial intelligence (AI) cannot lead to a new “form of barbarism,” where the common good is abandoned to the rule of law of the strongest. Francis also warned against the use of AI to “circulate tendentious opinions and false data that could poison public debates and even manipulate the opinions of millions of people, to the point of endangering the very institutions that guarantee peaceful civil coexistence.”

Though co-evolution theory suggests this is not a battle that will end anytime soon. JL

Karen Hao reports in MIT Technology Review:

Once a detection method has been developed to exploit a flaw in a particular generation algorithm, the algorithm can easily be updated to correct for it. As a result, some experts are now trying to figure out detection methods that assume the perfection of synthetic images. Others argue that reining in deepfakes won’t be accomplished through technical means alone: instead, it will also require social, political, and legal solutions to change the incentives encouraging their creation.

Just as Apple learned about the price of smartphones, there are limits to what people will pay for anything, especially housing and education for their children. JL

Janet Adamy and Paul Overberg report in the Wall Street Journal:

Large U.S. cities lost tens of thousands of millennial and younger Gen X residents last year. Cities with more than a half million people collectively lost almost 27,000 residents age 25 to 39 in 2018. It was the fourth consecutive year that big cities saw this population of young adults shrink. New York, Chicago, Houston, San Francisco, Las Vegas, Washington and Portland, Ore., were among those that lost large numbers in this age group. High housing costs and poor schools are main reasons that people are leaving. Millennials are marrying, having children and often settling down in suburbs.

Oct 5, 2019

Oct 4, 2019

Oct 3, 2019

Oct 2, 2019

Oct 1, 2019

Sep 30, 2019

Sep 29, 2019

contact

Search This Blog

Blog Archive

Labels

links