The Geek’s Reading List – Week of July 3rd 2015
I have been part of the technology industry for a third of a century now. For 13 years I was an electronics designer and software developer: I designed early generation PCs, mobile phones (including cell phones) and a number of embedded systems which are still in use today. I then became a sell-side research analyst for the next 20 years, where I was ranked the #1 tech analyst in Canada for six consecutive years, named one of the best in the world, and won a number of awards for stock-picking and estimating.
I started writing the Geek’s Reading List about 12 years ago. In addition to the company specific research notes I was publishing almost every day, it was a weekly list of articles I found interesting – usually provocative, new, and counter-consensus. The sorts of things I wasn’t seeing being written anywhere else.
They were not intended, at the time, to be taken as investment advice, nor should they today. That being said, investors need to understand crucial trends and developments in the industries in which they invest. Therefore, I believe these comments may actually help investors with a longer time horizon. Not to mention they might come in handy for consumers, CEOs, IT managers … or just about anybody, come to think of it. Technology isn’t just a niche area of interest to geeks these days: it impacts almost every part of our economy. I guess, in a way, we are all geeks now. Or at least need to act like it some of the time!
Please feel free to pass this newsletter on. Of course, if you find any articles you think should be included please send them on to me. Or feel free to email me to discuss any of these topics in more depth: the sentence or two I write before each topic is usually only a fraction of my highly opinionated views on the subject!
This edition of the Geeks List, and all back issues, can be found at www.thegeeksreadinglist.com.
1) Cancer reproducibility effort faces backlash
Strange things have happened to science over the past few decades. “Publish or perish” has resulted in a massive increase in peer reviewed studies, many of which have been shown to be deeply flawed or even fraudulent, often many years after publication. Journals regularly publish paper with import things (like the data and/or methods) retained as proprietary, or only available by permission. It is very difficult to have negative results published, as a consequence, little effort goes into actually checking whether previously published studies were reproducible. Over the long term, of course, reality always wins. The problem is that other scientists often base their work and analysis off unsubstantiated and in many cases non-repeatable work. Given the funding applied to cancer research, and the stakes involved, actually verifying whether “important findings” are actually true should be a priority. Sadly, as this article shows, “leaders” in the field explicitly do not want their result research questioned. That is a deeply perverse interpretation of the scientific method.
“Either way, the project seems a waste of time, Young says. “I am a huge fan of reproducibility. But this mechanism is not the way to test it.” That is a typical reaction from investigators whose work is being scrutinized by the cancer reproducibility project, an ambitious, open-science effort to test whether key findings in Science, Nature, Cell, and other top journals can be reproduced by independent labs. Almost every scientist targeted by the project who spoke with Science agrees that studies in cancer biology, as in many other fields, too often turn out to be irreproducible, for reasons such as problematic reagents and the fickleness of biological systems. But few feel comfortable with this particular effort, which plans to announce its findings in coming months. Their reactions range from annoyance to anxiety to outrage. “It’s an admirable, ambitious effort. I like the concept,” says cancer geneticist Todd Golub of the Broad Institute in Cambridge, who has a paper on the group’s list. But he is “concerned about a single group using scientists without deep expertise to reproduce decades of complicated, nuanced experiments.””
2) Tech giant plans for robots, defends labor record
Foxconn has been discussing increasing automation for some time now. Of course, a significant amount of electronic assembly is currently automated: it would be almost impossible to hand assemble a smartphone circuit board without advanced robotics systems. Replacing people with capital has been a trend since the industrial revolution and it isn’t going to change any time soon. Of course, some of this might be bluster in order to keep workers in line.
“Chairman Terry Gou said Foxconn Technology Group will be able to replace 30% of its workers on the production lines with robots in five years’ time. Gou said he wants to cut the number of staff in the repetitive, monotonous work of putting together mobile gadgets including Apple’s iPhones, and he plans to shift that 30% of workers to sales and software design positions. “I am not replacing jobs with robots,” Gou said during the annual shareholders meeting on June 25. “I am having workers do work that require thinking.” Gou’s statement came as the tech giant has been developing robotics technology for automated production, an area the executive has identified to be the company’s core businesses in the future.”
3) Robot brickie: Perth engineer invents world’s first robotic bricklayer
A robotic bricklaying machine has long been a dream for me. Bricklaying requires a lot of skill, the work proceeds quite slowly, and it is very easy to screw up. This machine appears to place structural bricks of the type common in house construction outside of North America, and that is probably a simpler task albeit one which is still rather labour intensive. Unfortunately, like many things call inventions nowadays, it is unclear whether the machine actually works. I have not been able to find a video of it in operation, just this animation https://youtu.be/Rebqcsb61gY.
““We’re at a technological nexus where a few different technologies have got to the level where it’s now possible to do it, and that’s what we’ve done.” “Hadrian” the robot — named after the famous Roman defensive wall of antiquity — will be commercialised first in WA, then nationally and then globally. Laying 1000 bricks per hour, it can work day and night, with the potential to erect 150 homes a year. It works by creating a 3D computer-aided design (CAD) laying program of a house or structure, then calculates the location of every brick and creates a program that is used to cut and lay the bricks in sequence from a single, fixed location. A 28m articulated telescopic boom goes to work and mortar or adhesive is delivered under pressure to the robotic laying head and applied to the brick which is then laid in the correct sequence as per the program. The robot de-hacks, measures, scans for quality and cuts to length the bricks and routs for electrical and other services.”
4) Detectives Investigating Drone Crash at Pride Parade
I remain pretty skeptical about the widespread use of drones for things like parcel delivery, however, there are clearly valid uses for the things. This incident was due to a drone operated by an amateur but that is not the point: getting struck in the head can kill or seriously injure somebody, and flying machines are going to occasionally drop from the sky. You can imagine the damage if the drone had been large enough to carry cargo, and was carrying cargo when it fell.
“A 25-year-old woman was knocked unconscious Sunday after she was struck by a small drone aircraft while attending the Pride parade in downtown Seattle. The woman was in the crowd at the parade near 4th Avenue and Madison when the 18″-by-18″ drone reportedly crashed into a building and fell into the crowd, striking the woman in the head. The woman’s boyfriend caught her as she fell to the ground. An off-duty firefighter helped treat the woman and called for police. At the scene, one the victim’s friends handed over the damaged drone—which retails for about $1200 and weighs about 2 pounds—and provided photographs of a man, who may have been piloting the aircraft.”
5) Maine crews use drone to rescue 2 boys from raging river
This is an example of a legitimate application of a drone, and this one was the personal property of one of the rescuers. Lifeguards are testing drones to deliver life jackets to swimmers in distress (http://www.roboticstrends.com/article/drone_vs_lifeguard_race_who_will_save_swimmer_faster) and the systems are being used to survey disaster areas. The difference between these and frivolous applications like delivering parcels or taking pictures at a parade is that there is a clear risk reward benefit and the people using the machines know what they are doing. It is hard to imagine an autonomous delivery drone would be able to avoid putting people at risk unless it is delivering to an empty lot.
“Officials in Maine have used a drone to deliver a lifejacket to a boy stranded on a rock in the middle of a raging river. Two boys needed rescuing Tuesday afternoon after their tube overturned in the Little Androscoggin River. Only one was wearing a lifejacket. Auburn Fire Chief Frank Roma told CBS affiliate WGME that before attempting a rescue, crews wanted to get a lifejacket to the boy without one. Roma said they used a drone to get a line to the boys so they could pull the lifejacket to them. “We wanted to make sure we got a life jacket on that second child so that if they did fall in the water we could catch them downstream,” Roma told WGME.”
6) California Adds Income-Based Caps for Clean Vehicle Rebates on EVs and Plug-In Hybrids
As we have noted in the past, EV subsidies provide minimal bang for the buck environmentally and there is considerable evidence the system is gamed. California is a peculiar place where environmental policies appear based on “truthiness” rather than sound science, so I doubt much will change. The implementation of an income based cap on a small portion of EV subsidies may be indicative of a trend, but it is doubtful this move alone would impact the industry. After all, the average income of a Tesla owner is just under $300K (http://insideevs.com/strategic-vision-says-testosterone-is-what-sells-the-tesla-model-s-to-wealthy-americans/) so a minor reduction in subsidy would not impact a purchase decision. Of course, if political actually starts being applied to the use of taxpayer money to pay for rich peoples’ cars to no significant environmental benefit, that would be the end of the industry.
“The second change is the complete elimination of the incentive for those with incomes over $250,000 for EVs and PHEVs. If you think high-income earners weren’t applying for incentives, stats compiled by the Center for Sustainable Energy say otherwise. Survey data from the second quarter of 2015 indicates over 26% of recipients had an income over $200,000. That’s nearly equal to the 27% of those earning less than $99,000. Interestingly, nearly three quarters of rebates so far this year went to those with incomes over $99,000.”
7) Heads-up displays in cars can hinder driver safety
There was a fair bit of excitement over Google Glass a couple years ago. I remain unconvinced that people can handle a constant stream of visual information as our brains did not evolve within the context of multiple concurrent activities. Even our eyes have acute vision in a very narrow field of view. Fighter pilots seem to manage, so perhaps a properly designed display can be managed, given enough training. Although I have never been in a car with a Heads-up Display (HUD) I would have figured the ability to keep your eyes on the road would be a major advantage. Apparently not.
“Heads-up displays (HUDs) in cars were once a rare thing. More and more, new cars now come with HUDs as standard, and you can even buy aftermarket HUDs. HUDs project useful information like the car’s current speed and navigation directions into the driver’s field of view, saving them from having to look down at an instrument panel or display, the idea being to reduce distractions and keep a driver’s eyes on the road. But a study from the University of Toronto led by Ian Spence suggests that HUDs might actually have the opposite effect and can even be a threat to safety. According to the study, published last month in PLoS ONE, the question of how our brains deal with dividing our visual attention between spatially commingled information isn’t currently well understood. Rather, most studies have looked at how divided attention works when performing a single task that requires us to get visual information from two distinct spatial locations (i.e., looking down at an infotainment display and then at the road). The researchers wanted to get a better idea of how commingled division of visual attention works in practice, using a simulation of an augmented reality HUD to do so.”
8) Not interested! Jaguar Land Rover will never make a self-driving car
I have to admit it is hard to imagine there will be much of a market for driverless Ferraris or Lamborghinis, but I don’t place Jaguars or Land Rovers in the same category. According to earlier articles, including at least one still on their website, (http://www.jaguarlandrover.com/gl/en/innovation/news/2014/04/16/jlr_auto_research_tech_160414/) the company has been working on a host of self-driving features and their cars. Land Rovers in particular, seem loaded with systems designed to decouple the skill of driver when off road though it is hard to imagine the Land Rovers I have seen ever driving the places I regularly take my pickup truck, which has no such systems. Regardless, in the future some degree of autonomous system, such as auto-braking, is bound to be a required safety feature.
“Google self-driving cars are already on the streets of Mountain View, California as part of tests to understand how the tech works in midst of manual vehicles. Nissan and Renault said they plan to introduce vehicles that can navigate without driver intervention in nearly all situations. Tesla, Mercedes-Benz, Audi, Volkswagen etc all have their own take on automated driving. So we can safely say almost all the big names in the automotive industry are all for it except Jaguar Land Rover. The British automaker has vowed that it will never build a driverless car because the company doesn’t consider its customers cargoes.”
9) Why Semi-Autonomous Vehicles are Still Relevant
This is pretty much an advertisement for ZF TRW’s current control systems but it does make a couple of good points. Semi-autonomous vehicles don’t require anywhere near the amount of electronics, and software as fully autonomous vehicles and yet they can accomplish much of the same objective. By decoupling the system at low speeds, as cruise controls do, you remove many of the most difficult tasks which need to be mastered for full autonomy. I figure these sorts of vehicles will be serve as a stepping stone to full AVs. Don’t bother with the video: it is just the guy narrating the article with a couple of pictures thrown in for good measure.
“In an age of Google self-driving cars, what’s the big the deal about semi-autonomous driving? Well, the ZF TRW system uses sensor and actuator technology already present in millions of cars on the road today – costing a fraction of the price of fully automated systems. ACC systems are currently used with radar and laser measurement systems and the self-park feature currently offered in popular models uses steering actuators as an offshoot of electric power-steering technology. The 25mph minimum speed limit requires drivers to take over the wheel completely at city intersections and with point-to-point navigation – two of the most difficult technical problems faced by today’s automated driving systems. The minimum speed limit also means drivers still need to get out of driveways, navigate through neighbourhoods and move on to secondary roads or freeways themselves. However, as the driver has minimal input into the vehicle’s movements for much of the ride, it will be easier to become distracted when the car is ready to relinquish control.”
10) Tesla to open seven new service centers in Japan by end of 2015
Another example of “Muskematics”. Akhita Japan gets about 3 kWh/day of sunshine on average (see http://pveducation.org/pvcdrom/properties-of-sunlight/average-solar-radiation#). At about 15% efficiency (which is high), and 2 axis tracking, they would need about 142 square meters of solar cells per charge (assuming they charge an 80 kWh battery at 20% remaining). My local filling station fills about 20 cars per hour (I live in the country) all day. Let’s say Tesla is only a bit successful and they do 20 per day. That’s just under 3,000 square meters of real estate, 60% of a football field, which I assume is very cheap in Japan.
“Musk also said that Tesla, which began selling the Model S electric car in Japan last year, will build more “supercharger” stations across the country. The company currently has six charging stations: three in Tokyo and one each in Yokohama, Osaka and Kobe. Musk emphasized that the charging stations will operate independently of Japan’s electric grid, which has seen prices rise as nuclear power plants in the country remain offline following the 2011 Fukushima nuclear accident. “Our proposal is to use solar panels to power charging stations, and then there’s no impact on the grid,” the CEO said. “That way, the cost of electricity is much less and the electricity is coming from pure sunlight, which is therefore very clean.””
11) Google’s New Project Is So Insanely Advanced It Will Blow You Away
Ignore the clickbait title: it seems everybody wants to be the next Business Insider. This article, and the interesting video, are about Google’s Project Soli, which uses an ultra low power, short range, radar system as a user interface. By using radar they capture 3D information associated with your movements, which you have to admit it pretty amazing. Despite the video, it is not clear how useful such a thing is – like voice recognition, misinterpretation can render a user interface impractical. Nevertheless, the fact a radar system can be put in such a small volume is pretty remarkable.
“If Google has its way, our future will be nothing less than a sci-fi movie. After creeping us out with a robotic cheetah and the Google ‘Glass’, Google is all set to bring forth something really amazing. Google’s Project Soli has invented a new interaction sensor using radar technology that can capture motions of your fingers at up to 10,000 frames per second. And that is something that has never ever been done before. Simply put, this technology is so bafflingly accurate that you could operate any device (fitted with this) without having to even touch it.”
12) One of the Most Important Tools in Science Now Fits Inside Your Phone
Here is another example of interesting design, however, it is not clear to me how useful the device will be. For the record, I doubt self-diagnosing skin cancer will be a killer app.
“By using tiny amounts of strange, light-sensitive inks, Bao and his colleague Moungi Bawendi—a chemist at MIT—have designed a working spectrometer that’s small enough to fit on your smartphone. Because of the tool’s simple design and its need for only an incredibly small amount of the inks, Bao says, his spectrometer only requires a few dollars worth of materials to make. They report the research today in the journal Nature.”
13) ‘Cloud tax’ upsets Chicago tech community: ‘Life just got 9 percent harder’
The headlines around this brilliant stroke focus on its impact on consumers by taxing Netflix and other streaming services, but lawyers have determined that it actually includes all kinds of cloud based services, including SaaS (software as a service). Many business applications have moved to SaaS so this would increase costs of Chicagoan businesses and create an accounting nightmare for service providers. Since it is doubtful cloud service providers would implement the necessary accounting changes to serve such a small market, and since, in many cases, they have no idea where a customer is located, it is hard to believe this will produce much in the way of income for the city. It is bound to cause tech companies to reconsider locating there, however.
“Chicago’s new 9 percent tax on streaming and cloud services appears to have the local technology community agitated and, more than anything, confused. Reports on Wednesday of the “cloud tax” took many Chicagoans by surprise, leaving providers and consumers of streaming and cloud services scrambling to understand the implications. Technology companies, among the heaviest users of cloud services, are likely to be taxed for the services they use as well as those they provide. The cloud tax extends ordinances governing two types of taxes — the city amusement tax and the city personal property lease transaction tax. The taxes cover many products streamed to businesses and residents. They also cover use of various online databases that could especially affect businesses.”
14) BBC to lay off 1,000 people as Britons cut the cord and TV licences decline
Cord cutting generally refers to disconnecting from cable or satellite services, which is clearly not the case here. What seems to be going on is that people are simply not buying TVs, choosing to stream content to various forms of computers instead. This is part of a global trend, in particular amount the youth, and it will result in significant challenges for traditional broadcast business models down the road. This need not continue to be a problem for the BBC. Provided they have political support, they’ll just ask for a tax to be placed on broadband services to make up the gap.
“The British Broadcasting Corporation will cut more than 1,000 jobs to cover a 150-million-pound ($294 million Cdn) gap in licence fee income next financial year as millions of viewers turn off their televisions and watch programmes on tablets and mobile phones. … The BBC’s Head of News, James Harding, last month predicted that by 2025, most people in the United Kingdom would probably get their television programmes over the Internet. “The Internet has ripped a hole in the business model of many great news organisations,” said Harding.Just 69 per cent of viewing by British adults is now through live TV and among 16- to 24-year-olds only 50 per cent of viewing was done through live TV, the country’s telecoms regulator said.”
15) Apple’s Beats 1 launch adds momentum to radio’s big digital shift
I guess something hasn’t been invented until Apple copies it, but the article does raise number of good points. As I predicted long ago (1997) the Internet jeopardizes traditional broadcast models (see article 14). Interest in broadcast radio among the youth (a core demographic for advertisers) is on a long slide to oblivion, and in many markets the response has been to lower quality even further. It is worth noting that people still listen to AM radio, or so I am told, so it will take a very long time before FM dies completely. It will happen, however.
“Radio was the original social network. It was the first format that allowed people in different areas to listen to music at the same time, in real time, and react. Driving this were DJs — men and women who obsessively sought out new music and new artists so they could deliver new music experiences. And consumers loved it. They were a part of something bigger than just themselves, and it gave birth to the modern day music business. But the costs of operating a radio network — based on either traditional FM stations or satellites — are very high. The consolidation that ended the era of local radio led to mega-corporate airwaves that controlled what music you could hear as a few companies came in and bought up almost every station in the country. That’s how we got to the cluster model operated today. With this, radio programming shifted from tastemaker DJs to radio consultants and MBAs driven by ratings and revenues.”
16) What’s Blocking The $11 Trillion Internet Of Things Opportunity
Despite the hype, the article covers a number of important issues associated with slowing the adoption of Internet of Things (IoT) technology. Open standards are extremely important, as without them the technology will never take off. Of course, companies are promoting their own standards because control of standards lead to the development of tremendous wealth in the tech industry. Another factor with in not addressed in the article is the fact most IoT systems use a cloud based infrastructure which ceases to function once the IoT vendor bankrupts or loses interest in the product. This may not be an important consideration today, but it will be after enough consumers and businesses are left with mission critical non-function IoT gadgets. Don’t believe the hype – $4 to $11 trillion is likely overstated by a factor of 1,000.
“The Internet of Things (IoT) is a goldmine waiting to happen, says a new report from management consulting firm McKinsey & Co.. However, according to its findings, we may be waiting a long time. The tech industry seems to have reached full lather over this trend, with companies big and small rushing in to connect all manner of gadgets, home appliances, even cars and other technologies, to the each other and the Internet. The reason is obvious: McKinsey points to $4 trillion to $11 trillion of positive economic impact each year by 2025. But its report also highlights roadblocks that stubbornly refuse to go away. Increasingly, the world is going to need to turn to open source to get the standards “unstuck.””
17) New Device Provides Secure and Anonymous Wi-Fi With an Incredible 2.5-mile Range
This article shows the power of the maker movement. You might recall some time ago the FBI arrested the operator of the Silk Road (illegal) drug market by tracking him down in a library and grabbing his computer before he could disable it. This open source device will allow miscreants to be physically distant from their Internet connection, making such an arrest very difficult. Of course, nobody has a lot of sympathy for drug dealers, but the same technology could be used by people organizing against a repressive regime.
“While a range of technologies (such as ToR) can provide some level of anonymity, a fundamental flaw still exists: a direct relationship between IP address and physical location. If your true IP is ever uncovered, it’s game over – a significant threat when your adversary owns the infrastructure. To resolve this issue, I present ProxyHam, a hardware device which utilizes both WiFi and the 900Mhz band to act as a hardware proxy, routing local traffic through a far-off wireless network – and significantly increasing the difficulty in identifying the true source of the traffic. In addition to a demonstration of the device itself, full hardware schematics and code will be made freely available.”
18) SpaceX Rocket Failure Threatens Support for Commercial Spaceflight
The spectacular failure of the SpaceX rocket last week was fascinating, but not so much because of the explosive, which was very impressive, but the need many commentators had to explain it away – after all Elon Musk is the greatest human who ever lived (I have actually seen him compared to Einstein), so you have to cut the guy some slack. The same websites which reacted gleefully at the destruction of a Russian spacecraft were sombre and cautious with respect to the loss of the SpaceX one. This article looks at the bigger picture, namely the loss of three ISS supply ships in relatively short order. At this time it appears the launch of a Russian spacecraft today went off without a hitch, so at least the astronauts won’t have to abandon the space station.
“The accident also casts a shadow on NASA’s plans to use commercial spacecraft to carry astronauts to the space station starting in 2017, to avoid continuing to rely on Russia. The same Falcon 9 rocket and Dragon capsule that failed over the weekend are the basis of the vehicle that SpaceX plans to use to fulfill its $2.6-billion contract under NASA’s Commercial Crew Program. Space industry veteran Boeing has a separate $4.2-billion deal to transport astronauts, but despite its history, the company had been seen as lagging behind front-runner SpaceX in the Commercial Crew race. “Now the bloom goes a little bit off the SpaceX rose,” Handberg says.”
19) Airplane Coatings Help Recoup Fuel Efficiency Lost To Bug Splatter
A while back I read something about super hydrophobic coatings being applied to aircraft as a permanent solution to icing. Rather than waiting while your aircraft is hosed down with antifreeze, a coating would prevent any ice from forming (http://www.theengineer.co.uk/channels/design-engineering/news/hydrophobic-coating-prevents-the-build-up-of-ice-on-aircraft/1014681.article). I don’t know what happened to that, but scientists have moved on to another problem: bug guts. Its hard to believe bug guts have a 5% impact on fuel economy, but, hey, who is going to argue with somebody who studies bug guts at NASA? I figure they should team up with these guys http://scitechdaily.com/liquiglide-nonstick-coating-coming-to-consumer-goods/ so the bug guts will slide right off.
“NASA scientists are now developing coatings that help aircraft shed or repel bug guts during flight. After screening nearly 200 different coating formulations, the NASA researchers recently flight-tested a handful of promising candidates on a Boeing ecoDemonstrator 757 aircraft in Shreveport, La. The team explored different combinations of polymer chemistry and surface structure and reports that it has created a coating that could reduce the amount of insect insides stuck to the wings by up to 40%. With further optimization, such coatings could allow planes to use 5% less fuel, Siochi says. Although that may not sound like much, it adds up. “That could be millions of dollars in fuel savings,” Siochi explains. The bump in fuel efficiency would also curb the amount of greenhouse gases emitted by planes, she says.”
20) Google apologises for Photos app’s racist blunder
A few weeks ago my son sent me a link to a Google Photo classification which had classified our cats as dogs. I did not immediately conclude the app was racist (speciesist?), even though Google EXPLICITLY declares itself a dog company (https://investor.google.com/corporate/code-of-conduct.html#toc-dogs) and tries to exclude cats from its offices. No – saner minds prevailed and I realized the algorithm needed some work. Indeed, if Google Photo had correctly classified Donald Trump as orangutan (https://thegeeksreadinglist.files.wordpress.com/2015/07/24ce7-trum-orangutan.png) nobody would have said a word. Its not racist: it just software.
“Its product automatically tags uploaded pictures using its own artificial intelligence software. The error was brought to its attention by a New York-based software developer who was one of the people pictured in the photos involved. Google was later criticised on social media because of the label’s racist connotations. “This is 100% not OK,” acknowledged Google executive Yonatan Zunger after being contacted by Jacky Alcine via Twitter. “[It was] high on my list of bugs you ‘never’ want to see happen.” Mr Zunger said Google had already taken steps to avoid others experiencing a similar mistake.”