An article about a new battery is making the rounds and I am getting a lot of questions about it – Ex-Navy officer turned inventor signs a multi-million deal to produce his electric car battery that will take drivers 1,500 miles without needing to charge. As stated, that sounds like a significant advance, about a 5 fold improvement over the current lithium-ion batteries powering a Tesla, for example.

That would certainly be a huge advantage and give the electric car industry a significant boost. Increased range would alleviate “range anxiety” and also mean the recharging could happen once a week rather than every night. It would also make electric vehicles easier to use on long trips. Further, increased range is the same as smaller batteries. Instead of a range of 1,500 miles, you cold have a battery with a 300 mile range that weighs one-fifth as much. (I am assuming that when they state the range they are comparing batteries of the same size.) That would make the vehicle more efficient and potentially cheaper.

But as always, the devil is in the details. What exactly are they talking about? There are lots of red flags in this article, starting with the fact that it is in the Daily Mail, which doesn’t exactly have a good reputation for high quality journalism. Also it makes it seem like this is the invention of one guy, rather than a lab, company, or even industry. That’s not realistic. There is also this:

Few will have heard of Jackson’s extraordinary invention. The reason, he says, is that since he and his company Metalectrique Ltd came up with a prototype a decade ago, he has faced determined opposition from the automobile industry establishment.

Sorry, but this conspiracy theory does not pass the smell test. New battery tech would not threaten the automotive industry, it would be a new option.

Continue Reading »

Like this post? Share it!

Evaluating the risks and benefits of nuclear fission is a bit of a moving target as the technology develops. Even with established nuclear power plant designs and management technology, I think the benefits outweigh the risks when you compare it to the alternatives and factor global warming into the mix. (I discussed this before and won’t go over all the points again here.)

However, we are not stuck with the current nuclear technology. We are on the brink of developing so-called Generation IV reactors that have a number of advantages. They are safer, smaller, more efficient, and generate significantly less waste (used nuclear fuel – UNF). Scientists, however, have now reported an advance that can potentially significantly reduce UNF, with or without Gen IV reactors.

Closing the Nuclear Fuel Cycle with a Simplified Minor Actinide Lanthanide Separation Process (ALSEP) and Additive Manufacturing.

This is a process for separating various components of the UNF, specifically removing the actinide lanthanide elements. The targets are long-lived isotopes of americium (Am) and curium (Cm) and also neptunium (Np). These are called minor actinides (MA) in the paper. Why is it important to separate out these elements from the UNF? There are two potential reasons.

First, these are the highly toxic and very long-lived radioactive elements in UNF. If you separate them out, the rest of the nuclear waste can be stored much easier and the time it would take for the half-life to decrease the radioactivity down to the level of uranium ore would be reduced from hundreds of thousands of years to just hundreds of years. It’s also easier to store because it will not get as hot.

Continue Reading »

Like this post? Share it!

Let’s consider the following scenario – the Earth is at risk for a disruptive event. This event has, conservatively, about a 0.2% chance of happening on any given year. But that is the most conservative estimate, at the high end it could be more like 12% over the next decade. Either way the chance of this type of event happening in the 21st century is quite high, and no matter what it is inevitable.

The result will likely be taking out power grids, possibly world wide in a worst-case scenario. Reasonable recovery will take about a year, with full recovery taking about a decade. Just imagine what would happen if we lost our power grid for a year. No digital banking, no internet, no household power. The most conservative estimate of how much such an event would cost is $2 trillion dollars, but experts are increasingly leaning toward $20 trillion as being a closer estimate (and this figure will only go up in the future).

So here’s my question – what do you think we should spend now to avoid a high probability of civilization collapse over the next century costing tens of trillions of dollars and growing? I am not talking about global warming, or environmental degradation, the death of the bees, an asteroid strike, or massive crop failure. I am talking about a coronal mass ejection (CME) – a solar storm.

A CME is actually the greatest threat to civilization that we face, in terms of probability and effect. In fact I think we are underestimating the chaos that a worst-case scenario would cause. Imagine going without power for a year. I know, there are people around the world who live without power, and the residents of Peurto Rico recently experienced something similar. But if this happened on a global scale, there’s no one coming with aid. Global infrastructures on which we all depend would collapse. How many people would starve or freeze? How much wood would be burned to keep warm or cook until the power comes back on? There are so many downstream effects that we cannot anticipate.

Continue Reading »

Like this post? Share it!

We are beginning to experience some growing pain with widespread adoption of wind and solar energy. Solar in particular is causing utility companies some heartache because of its distributed design and intermittent energy production. None of the issues raised are fatal problems, but we do need to address them head on.

The basic problem is that we cannot simply look at each piece of energy production in isolation. It’s tempting to think that if you install solar roof panels, that is a pure environmental good because clean energy will be replacing more dirty energy. Initially this may have been largely the case, with very low penetration of distributed solar. However, as the amount of installed solar increases, tensions with utility companies and the complexity of integrating into the power grid are rising.

In 2018 solar was producing 1.6% of total electricity generation in the US (total renewable is 17.1%), but this is projected to rise considerably over the next decade. Solar and wind cause challenges for the grid because they are intermittent sources. For the individual installing solar will likely result in decreased electricity cost. However, if our concern is the overall efficiency and environmental impact, a more complicated evaluation is necessary.

A recent study by Duke Energy highlights this complexity. This is not an objective source – some utility companies are taking a hostile attitude toward distributed energy because of the problems it causes. Here is a quote from Dan Kish, distinguished senior fellow at the Institute for Energy Research, to give you an idea:

“Renewable energy sounds good, but it performs terribly. If you want electricity available when you need it, you don’t want intermittent, unreliable, renewable energy,” Kish said. “It’s like a cancer on an efficient grid, with its ups-and-downs forcing other sources to pick up the slack in the most inefficient ways, which, in some cases, are more polluting.”

Continue Reading »

Like this post? Share it!

The next NASA rover to Mars, the Mars 2020 Rover (final name to be determined), launches next July. It will arrive at Mars in February 2021. This is the next iteration of rover design and has some interesting new features, include a drone that can fly around to survey more of the Martian surface.

But perhaps the feature that is getting the most attention is the drill. For the first time a Mars rover will have the ability to drill down into the rock and dirt. Why is this so important? If Mars ever contained life, then it is likely the remnants of that life can be found down in the rocks, rather than on the surface. This is the first rover specifically designed to look for signs of life.

There is even the remote possibility of finding signs of recent, or even current life. The mission will be looking for life signatures, such as certain forms of carbon, or signs of sustained water presence in the past. Once the rover lands and is operational, it should only take a few months for answers to start beaming to life. We may know by the middle of 2021 if life ever existed on Mars.

Finding signs of life on Mars will have profound scientific implications. However, CNN, when reporting about this, chose to go with this headline: “When — or if — NASA finds life on Mars, the world may not be ready for the discovery, the agency chief says.” They quotes NASA chief scientist, Jim Green:

“It will be revolutionary,” Green told the Telegraph. “It will start a whole new line of thinking. I don’t think we’re prepared for the results. We’re not.”

Continue Reading »

Like this post? Share it!

A recent systematic review and meta-analysis of studies comparing humans to artificial intelligence (AI) in diagnosing radiographic images found:

Analysis of data from 14 studies comparing the performance of deep learning with humans in the same sample found that at best, deep learning algorithms can correctly detect disease in 87% of cases, compared to 86% achieved by health-care professionals.

This is the first time we have evidence that the performance of AI has ticked over that of humans. The authors also state, as is often the case, that we need more studies, we need real world validation, and we need more direct comparisons. But at the very least AI is in the ball park of human performance. If our experience with chess and Go are any guide, the AI algorithms will only get better from here. In fact, given the lag in research, publication, and then review time, it probably already is.

I think AI is poised to overtake humans in diagnosis more broadly, because this particular task is right in the sweet spot of deep learning algorithms. Also, it is very challenging for humans, who fall prey to a host of cognitive biases and heuristics that hamper optimal diagnosis. A lot of medical education is focused on correcting these biases, and replacing them with clinical decision-making that works better. But no clinician is perfect or without blind-spots. Even the most experience clinician also has to contend with an overwhelming amount of information.

There are a couple ways to approach diagnosis. The one in which human excel is the gestalt approach, which is essentially based on pattern recognition. With experience clinicians learn to recognize the overall pattern of a disease. This pattern may include signs or symptoms that are particularly predictive. Eventually the pieces just click into place automatically.

Continue Reading »

Like this post? Share it!

The current world population is 7.7 Billion. World population will approach 10 billion by 2050. The primary limiting factor on human population is the availability of food. We are already currently using essentially all the available practical arable land. Expanding farmland further at this point would involve using less productive land, cutting down forests, or displacing populations. Converting ecosystems into farmland also has a huge impact on the environment and species diversity.

So how are we going to feed 10 billion people in 30 years? So far agricultural development has kept up with demand. It’s tempting to assume that such development will continue to keep up with our needs. This is the endless “peak whatever” debate – if we extrapolate any finite resource into the future, it always seems like it will run out. But historically scientific development and simple ingenuity has generally changed the game, pushing off any resource crash into the future. This has lead to two schools of thought – those who argue that history will continue to repeat itself and we will indefinitely push our resources as needed, and those who argue that no matter how clever we are, finite is finite and will eventually run out.

When we are talking about food, the result of a crash in resources means mass starvation. That is how nature solves the limited food problem; animals starve and reduce the population until a natural equilibrium is reached. And you cannot say that mass starvation has never happened. In the 1960s at least 36 million people in China starved to death due to mismanagement of their agricultural system. We may be setting ourselves up for repeats of this situation. If we push our agricultural system to its limits in order to feed a growing population, does that system become increasingly vulnerable? What if a blight wipes out a staple crop, or bad weather significantly reduces yield? We may have less and less reserve or buffer in the system to handle these kinds of events.

We can talk about population control, and that is a valid approach. I mostly favor population control through lifting people out of poverty and gender equality, which are the most significant causes of overpopulation. Regardless, there is no reasonable expectation that we will stop the human population from reaching 10 billion and it’s not clear where we will level off. We need a plan to feed this population. I reject the radical suggestion that we should keep food production where it is, or even reduce it, and let that be a natural check on population growth. This solution is mostly suggested by those will little chance of starving as a result, a burden that will mostly fall on the poor and oppressed.

Continue Reading »

Like this post? Share it!

Sometimes technology is developed for a specific function. At other times, however, technology is developed simply because it is possible, and then uses are found for the technology later. Most often, however, it seems that technology is developed with some specific application in mind, but once it exists out in the world many new functions are developed. In fact the “killer app” may have nothing to do with the original purpose. This is partly why it is so difficult to predict future technology, because it is impossible to replicate the effects of a massive marketplace.

This is what I was thinking when I read about a somewhat new technology – indoor solar cells. This is actually not totally new, it’s basically organic solar cells. However, the solar cells have been optimized for the spectrum of light typical in indoor environments. There are, in fact, already indoor solar cell products on the market, but they are basically just regular solar cells sized for indoor applications. What’s innovative about these new indoor solar cells is their greater efficiency in indoor environments – the researchers claim a 26.1% efficiency. Further:

“The organic solar cell delivered a high voltage of above 1 V for more than 1000 hours in ambient light that varied between 200 and 1000 lux. The larger solar cell still maintained an energy efficiency of 23%.”

That’s pretty interesting, but not surprising. We are still on the steep part of the curve when it comes to solar cell development. I reviewed the technology recently – there are basically three types. Silicon-based cells are the ones currently dominating the market. Thin film, like perovskite, are being developed and promise to be better and cheaper, but there are still some technical hurdles to overcome. They are not yet stable for long term use, for example.

Continue Reading »

Like this post? Share it!

Our president is a global warming denier, is anti-vaccine, and is a conspiracy theorist. Regardless of where you are on the political spectrum, being anti-science is never a good thing. When those in positions of power are ignorant of science and hostile to the institutions of science and the methods that those institutions espouse, that is a recipe for disaster.

But even a stopped clock is correct twice a day. And even though there appears to be a significant asymmetry in the degree to which our two major political parties take anti-scientific positions, on some issues the political left has it wrong for their own ideological reasons. The two big anti-science issues popular on the left are anti-GMO stances and anti-nuclear energy. The latter was recently brought into sharp relief when Trump signed a, “Memorandum on the Effect of Uranium Imports on the National Security and Establishment of the United States Nuclear Fuel Working Group.” 

I doubt that Trump, who has demonstrated profound anti-intellectualism and even an unwillingness to read, has a deep knowledge of the scientific issues surrounding nuclear power, but he is a conduit for those who do, unfettered by political opposition (which remains on the left). Meanwhile, Bernie Sanders, in his version of the green new deal, states, “To get to our goal of 100 percent sustainable energy, we will not rely on any false solutions like nuclear, geoengineering, carbon capture and sequestration, or trash incinerators.” He plans to completely and quickly phase out all nuclear power in the US.

I have to point out for completeness that more moderate Democratic candidates, like Joe Biden, do include nuclear power in their energy infrastructure plans to combat global warming.

Also, attitudes toward nuclear power have been moving toward more favorable in recent years. This seems to be due to a few factors. The more people know about nuclear power, the more favorable they are towards it. Fears about global warming have caused some to moderate their views on nuclear energy. And newer reactors designs are moving toward smaller and safer designs.

There is still an asymmetry politically, however. Only 31% of Democrats say that nuclear power is essential or helpful, while 34% say it would be harmful. For Republicans the numbers are 50% and 17% respectively.

Continue Reading »

Like this post? Share it!

A recent commentary in Nature Communications echoes, I think, a key understanding of animal intelligence, and therefore provides an important lesson for artificial intelligence (AI). The author, Anthony Zador, extends what has been an important paradigm shift in our approach to AI.

Early concepts of AI, as reflected in science fiction at least (which I know does not necessarily track with actual developments in the industry) was that the ultimate goal was to develop a general AI that could master tasks from the top down through abstract understanding – like humans. Actual developers of AI, however, quickly learned that this might not be the best approach, and in any case is decades away at least. I remember reading in the 1980s about approaching AI more from the ground up.

The first analogy I recall is that of walking – how do we program a robot to walk? We don’t need a human cortex to do this. Insects can walk. Also, much of the processing required to walk is in the deeper more primitive parts of our brain, not the more complex cortex. So maybe we should create the technology for a robot to walk by starting with the most basic algorithms similar to those used by the simplest creatures, and then build up from there.

My memory, at least, is that this completely flipped my concept of how we were approaching AI. Don’t build a robot with general intelligence who can do anything and then teach it to walk. You don’t even build algorithms that can walk. You break walking down into its component parts, and then build algorithms that can master and combine each of those parts. This was reinforced by my later study of neuroscience. Yeah – that is exactly how our brains work. We have modules and networks that do very specific things, and they combine together to produce more and more sophisticated behavior.

Continue Reading »

Like this post? Share it!