2015 July 29 Wednesday
Long Telomeres Associated With Higher Lung Cancer Risk

Longer caps at the ends of chromosomes are associated with a greater risk of lung cancer.

A large-scale genetic study of the links between telomere length and risk for five common cancers finds that long telomeres are associated with an increased risk of lung adenocarcinoma.

This relationship did not hold for other cancer, except prostate cancer. The lack of relationship with other cancers is what surprises me. Telomere shortening when cells divide is probably an evolved response to reduce the risk of death from cancer. Cancers need to mutate to turn on telomerase to make telomere caps longer. This enables them to keep dividing.

We really really need a cure for cancer. If we could cure cancer then lengthening telomeres would provide benefits (better body repair and less aging) without any downsides from greater cancer risk.

By Randall Parker 2015 July 29 08:44 PM 
Entry Permalink | Comments(1)
2015 July 24 Friday
Robotic Cooks And Your Own Preferred Recipes

In the future you will get food automatically prepared for you wherever you go exactly the way you want it. The diversity and quality of prepared food available in a small city will far exceed that available in the biggest and most diverse city of today.

Each person will gradually build up a personalized set of computer-encoded very detailed recipes for favorite dishes along with a profile specific aversions and preferences you might have about various fruits, vegetables, fish, seasoning and other food ingredients. Friends (and random people on the internet) will give you access to their own favorite recipes. A computer somewhere in the cloud will store all this information. It will also store the information on preferred dishes of people all over the world. Hundreds of millions of people will put their preferred recipes in the public domain. Travel to go eat exotic food will no longer be necessary. An automated restaurant will deliver the world's cuisines to any place big enough to support the capital investment for a first class robotic cooking system. The cost of such a system will decline with time.

The intelligent dining recipe computer system will also have extensive knowledge about each restaurant including automated cooking devices, food ingredients on hand, and waits to expect at each restaurant. A database of information about various models of robotic cooking equipment will be very detailed with regard to their food preparation capabilities. Another database will supply detailed information about which restaurants have which kinds of automated cooking equipment. The computer system will also know the inventories of currently stocked ingredients at each restaurant. The computer system will even know projected wait times and food prep times for each dish you are considering.

So what happens when you want to go out to eat? Something akin to match making and bidding. The software will report to you which restaurants can prepare which of your preferred meals for your preferred eating time, at what price, and will report quality considerations due to ingredient quality and equipment quality. It will report expected wait times too. You'll be able to select your preferred food and restaurant and have the food cooked while you are en route.

Existing fast food restaurant chains try to deliver very consistent meals. Not great. But usually knowable from previous experience at other restaurants in the same chain. But fully robotic food prep using robots that can precisely prepare many dishes will raise the game far above the current fast food chains. Variety will be enormous because the better robotic cooks will be able to cook a very wide range of dishes, especially when they are in restaurants that have a large assortment of ingredients. Upper end robotic cooks will offer great precision, faster speed, and a much wider skill set than the average human cook.

You'll be able to record your satisfaction (or lack thereof) afterward and the computer will learn what you ended up liking or disliking and why. Its machine learning models will get better at predicting which which meals at which restaurants you'll like and it will provide probabilities for satisfaction the next time you want to eat out.

Granted, some people will be up for random eating experiences with food made by human cooks and they'll still be able to do that. But if you are on a business trip and don't want a challenge and disappointment at dinner time the ability of many restaurants to prepare it exactly as you like it will give you a much more consistent and higher quality dining experience. It will cost less, use less time, and give you a better eating experience.

Also, if you really want a somewhat semi-random eating experience you won't need a human cook. You can just try recipes recommended by others or, even better, recipes recommended by competing machine learning models. You will be able to eat a much greater variety of dishes because an advanced automated commercial kitchen that serves a lot of customers will be able to stock lots of ingredients and will feature cooking devices that can prepare these ingredients in a great many ways.

By Randall Parker 2015 July 24 07:48 PM 
Entry Permalink | Comments(8)
2015 July 21 Tuesday
National Security Will Drive Genetically Enhanced Offspring

Eugene Volokh responds to the recent poll that found most people in the United States are opposed to raising offspring IQs. Volokh thinks the US will feel pressure to allow offspring IQ boosting or fall behind other nations. It becomes a national security issue if the American population becomes 20 IQ points lower than China, India, and Russia. So expect the national wing of a nation's elites to promote offspring genetic engineering.

Volokh does not think people will go so far as to alter their offspring to have an extra pair of arms. But the internet has made it easier for people with outlier preferences to find each other and create their own echo chambers. Therefore I fully expect some religious cults or ideological cults to advocate radical alterations in human anatomy and sensory systems in order to right social wrongs, prepare for the apocalypse, or please a deity figure so the diety will bestow the faithful with huge blessings. But outlier genetic engineering doesn't even require an echo chamber. Some loners will choose attributes for their offspring just to spite the rest of us. Those kids have my sympathy in advance.

In spite of the outliers I expect offspring genetic engineering will lead to brighter, better looking, higher performing, and healthy children and adults. What I'm less sure about is personality traits and moral reasoning circuitry. What chooses will people make about how conscientious and benevolent and considerate their kids will be? Will psychopaths choose genetic variants to create offspring who are even higher functioning psychopaths? Will skills in deception be desired by some? The average performance of predator humans may well rise.

Some time in the next 10 years we will have enough info about genetic variants that contribute to intelligence that embryo selection for higher IQ will become possible. This will lead to a big market in donor eggs and donor sperm with lots of embryos created to sort thru and choose the ones with the most desired combinations of attributes. But embryo selection will only be the first generation tech for making better babies. Genetic editing with Crispr-CAS9 and similar biotechnologies will allow prospective parents to give their offspring genetic sequences they themselves do not possess. Then the rate of human evolution will skyrocket. We live in the final decades of human wild type. Some day wild type humans will be the minority. In some countries that will happen in this century.

By Randall Parker 2015 July 21 09:02 PM 
Entry Permalink | Comments(12)
2015 July 19 Sunday
Jeep Hacked And Shut Down From Miles Away

Cars have sofware complex enough to be hacked.

Can car companies do a better job than, say, PC operation system makers and build computer systems in cars that can't be taken over remotely? I sure hope so. Otherwise when cars become capable of driving themselves a hacked car could be instructed to drive anywhere or at an unsafe speed or to suddenly veer into incoming traffic.

Autonomous vehicles going to some day communicate with each other to avoid accidents. But imagine a vehicle driving along spoofing other vehicles. One vehicle could pretend to be a few other vehicles in order to make another vehicle do something very dangerous or to force it to stop. Spoofing seems like a big potential problem.

By Randall Parker 2015 July 19 01:22 PM 
Entry Permalink | Comments(0)
Beijing Area 130 Million Super City Planned

Jing-Jin-Ji will unite Beijing (Jing), the port of Tianjin (Jin), and Hebei province (known as Ji) into one large urban area united by expanded high speed rail service. The Chinese have decided that since high speed rail is so much faster than cars they can create a larger city area with commutes from bedroom communities located further away from city centers.

The development in Jing-Jin-Ji is quite unbalanced due to peculiarities of Chinese tax law. The suburbs can't build up many obvious pieces you'd expect them to have.

What I find most interesting about this report is the impact of high speed rail on official thinking. I suspect they are making a mistake in that autonomous vehicles are quickly approach on the horizon. Autonomous vehicles are going to enable much more extensive small and large bus service in many directions and the directional flexibility of autonomous vehicles will compete with the speed of rail. Go to a few places quickly? Or to many places easily and cheaply?

By Randall Parker 2015 July 19 01:04 PM 
Entry Permalink | Comments(1)
2015 July 11 Saturday
Rate Of New Business Formation On Steady Decline, Even In Tech

Company size growing, fewer new competitors forming, number of small businesses shrinking. Start-up failure rate is increasing too. Happening in tech too. This isn't just the death of small town stores and pharmacies and shift to big chains. Read the whole thing.

I suspect that computer and communications tech are behind this trend. Why I think this is happening:

  • It is easier for a single company to sell into many markets because cheap communications, cheap shipping, lower trade barriers. Best company beats lower ranked companies that used to not compete directly.
  • Large IT infrastructure with large amounts of developed source code base by established players serve as barrier to entry. Initial software development costs higher. Though the opposite argument can be made with more reusable software and cloud computing services.
  • Software reduces the need to hire and train-up a big middle management layer to manage expansion into new markets. Middle management has shrunk because its information aggregation functions and communication functions have been automated.
  • The best brains are co-locating more to work together to creates best-of-breed companies. Fewer of the best talent are being wasted propping up also-ran lower tier companies.
  • Online review sites, news sites, and forums allow news of best-of-breed products and services to spread more rapidly.
  • Venture capitalists actively look for the best new thing and cause the best new thing to take off with more talent and money. We do not get rivals slowly emerging around the world. The fastest out of the gate get a lot of money and quickly rise to the top.

The article above also suggests growing government regulations as a barrier to entry. Too much knowledge of regs and time-consuming approval process reduce the number of new businesses. Seems likely. But not the whole story.

20 years ago it seemed possible that the internet would allow more people to work remotely and new tech companies to form in little towns and virtually across rural areas. Instead we have a large surge of start-ups in San Francisco and a big migration of tech talent into urban areas. Technology seems to have made geolocation more important, not less.

By Randall Parker 2015 July 11 06:45 PM 
Entry Permalink | Comments(10)
2015 July 06 Monday
Biomarkers Show Some People Aging 3 Times As Fast As Others

How fast are you aging? Using people from a long running longitudinal study in Dunedin New Zealand and an assortment of biomarkers scientists find that some people are aging 3 years for every chronological year:

Based on a subset of these biomarkers, the research team set a "biological age" for each participant, which ranged from under 30 to nearly 60 in the 38-year-olds.

The researchers then went back into the archival data for each subject and looked at 18 biomarkers that were measured when the participants were age 26, and again when they were 32 and 38. From this, they drew a slope for each variable, and then the 18 slopes were added for each study subject to determine that individual's pace of aging.

Most participants clustered around an aging rate of one year per year, but others were found to be aging as fast as three years per chronological year. Many were aging at zero years per year, in effect staying younger than their age.

As the team expected, those who were biologically older at age 38 also appeared to have been aging at a faster pace. A biological age of 40, for example, meant that person was aging at a rate of 1.2 years per year over the 12 years the study examined.

Here is the PNAS paper by Daniel Belsky et. al.: Quantification of biological aging in young adults:

We studied aging in 954 young humans, the Dunedin Study birth cohort, tracking multiple biomarkers across three time points spanning their third and fourth decades of life. We developed and validated two methods by which aging can be measured in young adults, one cross-sectional and one longitudinal. Our longitudinal measure allows quantification of the pace of coordinated physiological deterioration across multiple organ systems (e.g., pulmonary, periodontal, cardiovascular, renal, hepatic, and immune function). We applied these methods to assess biological aging in young humans who had not yet developed age-related diseases. Young individuals of the same chronological age varied in their “biological aging” (declining integrity of multiple organ systems). Already, before midlife, individuals who were aging more rapidly were less physically able, showed cognitive decline and brain aging, self-reported worse health, and looked older. Measured biological aging in young adults can be used to identify causes of aging and evaluate rejuvenation therapies.

Great aging biomarker metrics make it possible to measure the effects of anti-aging treatments in time spans small enough to make iterative adjustments to find better drugs.

By Randall Parker 2015 July 06 10:49 PM 
Entry Permalink | Comments(1)
2015 July 05 Sunday
Robotic Brick Layers And Other Construction Robots

A company called Construction Robotics has a real working brick laying robot.

They claim 900 bricks laid by a robot and a mason (human) in 6 hours. How does that compare to human brick layers working alone? How many bricks in a typical brick house? Here is another video of their robot at work building walls for a high school in Laramie Wyoming. And another one.

An Australian company, Fast Brick Robotics, claims their Hadrian robot will be able to build a brick house in 2 days. They do not seem to have that capability yet (at least as measured by a Youtube video showing it in action).They say they will be able to lay 1000 bricks per hour.If they can deliver brick housing will gain a competitive advantage over houses built with other materials.

Another way to do it: flying robots that lay bricks. Another approach: printing a house.

Several years ago I read that half the cost of housing construction is on-site labor. Well, that portion of housing costs is going to drop a lot. Construction materials that are easier to handle robotically will get used more. Will ease of automation favor wood, aluminum, steel, bricks, plastics, composites, or other materials?

Some of the more dangerous jobs that ought to be automated: lumberjacks, fishermen, roofers. Here's a 2015 list of most dangerous jobs.

By Randall Parker 2015 July 05 03:35 PM 
Entry Permalink | Comments(1)
2015 July 03 Friday
Rapid Price Declines Driving Fast Solar PV Growth

First Solar CEO Jim Hughes says By 2017 solar photovoltaics will cost $1 per watt installed.

"We have a technology roadmap -- by 2017, we'll be under $1.00 per watt fully installed on a tracker in the western United States."

To put that in context, solar cost $4 per watt about a dozen years ago.

First Solar has been boosting the conversion efficiency of their CdTe thin film. They've recently found a way to add another 2.3% to conversion efficiency.

Ramez Naam points out that while the world now gets 1% of electricity from solar by 2020 we will get 3% of electric power from solar. These price drops are making solar more competitive in more places.

Solar's price decline has beat forecasts. At this point the industry is large enough its costs are dropping fast enough that it is quickly losing its dependence on government subsidies. I expect it will become the biggest non-fossil fuel energy source, at least in regions with a lot of sunlight.

What we still need: much cheaper batteries. Make the batteries cheap enough and then most transportation can switch to electric power and solar electric power at that.

I'm becoming more optimistic about the migration away from oil because fracking has delayed Peak Oil. This has given us time for development of more efficient vehicles and cheaper solar and batteries. The biggest problem now is battery cost. The rest of it is coming together very nicely.

By Randall Parker 2015 July 03 12:53 PM 
Entry Permalink | Comments(59)
2015 July 02 Thursday
Opposition To Smarter Babies In United States

A Pew survey finds quite high rates of opposition in the United States to choosing genes that will make for smarter babies and smarter adults: 83% opposed to genetic editing in favor of smartness.

Strong majorities of both men and women are opposed to modifications aimed at increasing a baby’s intelligence, although opinion is more negative among women (87%) than it is among men (78%).

This result hardly differs by educational level or political party affiliation.

By contrast, I think boosting offspring intelligence will be a boon. The power of higher intelligence is enormous. Children will do better in school, grow up to be much more productive at work, earn more money, commit less crime, and live longer healthier lives. Boosting the intelligence of children will boost many desired measures of societal health.

I expect opinions will shift once prospective couples gain the means to choose embryos that will result in smarter babies. It is one thing to answer a survey about a matter in the abstract. It is quite another to be faced with a real decision with consequences that will affect you and yours for decades to come. The elites especially will do what it takes to make smarter offspring who are more capable of maintaining levels of family achievement across generations. If regulatory regimes block genetic editing (with CRISPR-Cas9) or embryo selection for better cognitive characteristics then I expect the upper classes will just travel to countries where such biotechnologies are allowed.

In China and developed countries in East Asia I expect both the broad public and the elites in government to eagerly embrace biotechnologies for making smarter babies. So East Asia may pull well ahead of the West with much smarter societies. Fear of this result might spur Western elites to promote the use of these biotechnologies.

By Randall Parker 2015 July 02 02:45 PM 
Entry Permalink | Comments(23)
Site Traffic Info
Site Copyright
The contents of this site are copyright ©