omenti

the universe has it in for computers

a minipost

Cosmic rays have fascinated me since I was 13 and found a detector with a couple of patient physicists on the summit of Sulphur Mountain in Banff.   That was my first exposure to relativity.  I written a few posts on cosmics, but just say the word if you ever want to get me started on particles that can have the energy of a fastball and require detectors spread over thousands of square kilometers to properly detect.   Smashing into molecules in the upper atmosphere they produce large showers of elementary particles.  These can make it to the ground - or to a memory chip in a computer.  A direct hit on a memory cell can flip a bit from 1 to 0 without damaging the memory.  It's why you don't locate computer animation companies in Albuquerque or Denver.  It's why you use ECC memory in computers that run programs for more than a few hours and why many spacecraft have multiple radiation-hardened computers.   In addition to computer problems they're responsible for a large number of biological mutations.  They're one of many reasons why I think it's nutty to send humans into space beyond near Earth orbit. 

 

It's a very rich topic and this video by Derek Muller is an excellent introduction with historical examples of problems.  They've even interfered with an election. 

very light narrative structures

I first heard about Norwegian Slow TV a decade after it began.  I watch a half hour of one of the rail trips. It didn't do anything for me.  Other than hearing about other hits like chopping wood and knitting, I almost forgot about it.  Two months into the pandemic I found my watching a semi-live train trip across Norway.  Much of the  scenery was similar to where I grew up and I found myself watching.   It wasn't long until I found myself daydreaming, creating my own narrative to match the flowing images.  Two hour later I stopped a dream-like story in my head.  

 

About four in the morning the next day I was jolted out of bed with an interesting approach to some physics that I had abandoned months before due to lack of process.    Having your mind work on something when you're not working on it is well known.  I suspect the neural connects I had was making in the narrative I was writing in my imagination had a good deal to do with it.   

 

This morning I caught an episode of Invisibilia that dives into the subject.  Recommended if you have an interest in narrative in any form.  Slow TV is an hours or days long TV show without plot, characters, or tension.  How Norwegians responded as a shared experience is the most fascinating part

I've done a bit of work in remote audio ambiances.   A few created serendipity.  Here's an excerpt from an old post on one:

 

5:25 am - excellent!

 

I brush the last of the snow from my coat and switch on the lights before taking my seat.   few minutes of silence and then a door opens and closes reverberating for about three seconds.  The clicking of shoes echoes through Warner Hall as she makes her way to the bench.  A minute of preparation and then the Praeludium in G by Bach fills her space and mine.  I sit back and shut all but the Bach from my mind.

 

She's doing an excellent job until - rats.  Bach derails and I sit up in my chair looking around.  A female shout fills the space reverberating for about three seconds.

 

sh*tttttttttttttttttttttttttttttttttttttttt!

 

She goes back a few bars and dances through the problem smoothly.  Now a piece I don't recognize, but it is still clearly Bach until about 6:55.   Her shoes and coat go on and steps click to the door in the distance.

 

And a little magic as a female voice is softly singing Bach.

 

A door opens and closes.  The reverberation is about three seconds and then silence. Time to switch off the amplifiers and make some hot chocolate before getting to work.

She was an organ student in Oberlin and I was in a special room in New Jersey where we were working on the reconstruction sound fields.  A special seven microphone array had been installed in the hall with a computer handling compression mixing and compression in realtime.  The bitstream made traveled eastward to our room where a reasonably convincing illusion of the hall's sound space was created.  We had created a virtual acoustic reality.

 

repairability

Growing up our family always had Fords.  For most of grade school through junior high that meant a blue Ford Falcon.  My Dad's reasoning was simple - a neighbor had good mechanical experience with Fords, particularly the straight-6 engine in the Falcon.  He also had a nice set of tools. Much of it was simple enough that I was involved in part of the routine maintenance even before I had my driver's license.  Simple things like changing the oil every 1,500 miles, gapping and replacing the spark plugs, making sure the battery levels were ok, draining the radiator, testing and replacing the antifreeze, and taking off the air cleaner to spray ether in when cold-starting it when it was below -20°F.   

 

By the time I was driving legally  (I drove a pickup on a wheat farm during harvest the year before I had my license. I suspect that was common in Montana.  Learning how to drive a manual transmission amounted to me trying to not stall the pickup while my boss watched laughing for an hour.)  By high school I graduated to more sophisticated work.  Replacing the head gasket on my own and even replacing the rings under the watchful eye of my Dad's friend.  Everything was exposed and simple - most of it was just sorting out fuel, air and spark.   Sometimes, driving to King's Hill on a hot Summer day, the engine would quit.  That meant pouring some cold water on the fuel line near the engine and waiting a few minutes to break the vapor lock.  Each car had a set of peculiar tricks, but makes had personalities.  Changing to a Chevy or Dodge would have a learning curve.

 

While the Falcon was simple and to work on, almost everything was poorly made.  Perhaps not to the eye, but tolerances in the engine and transmission were sloppy.  A careful break-in period was necessary if you wanted an engine and transmission to last.  It was inefficient even though it didn't weigh much by today's standards.  Although it was a compact, it only got 18 miles per gallon on the road and 13 or 14 in town.  Engine and transmission oil had to be replaced regularly.  Tires went about 10,000 miles if you were lucky, brakes needed adjustment every few thousand miles, the chassis needed lubrication regularly and so on.  The body and chassis designs were primitive and non-crashworthy by today's standards.  Cars often remained fairly intact even in fatal collisions.  

 

The seventies brought more efficient designs along with increasing safety regulation.  More sophisticated tooling that produced dramatically improved engine and transmission tolerances that allowed more efficient designs.   They lasted longer, but it was becoming difficult to work on them.   Repair was moving towards replace.  The failure of a wiring harness on the mechanism for an electric window might be a fifty cent part, but the replacement module could cost hundreds and require several hours of work to replace.

 

Home electronics followed a similar path.  Our first TV was a 19" black and white Magnavox from about 1961.  It had a few dozen tubes, a lot of point to point wiring and hand-soldering, and mechanical switches.  If something went wrong the first step was to take off the back of the set and see if any of the tubes weren't glowing.  You'd pull the suspect out and test it at the local drugstore - most had a tube testing kiosk with a stock of replacement tubes.  That happened a few times before we got our first color TV in 1969 - a Motorola "Quasar" which only had two tubes including the picture tube.  It's main selling point was reliability. By the 80s integrated circuits began to invade home electronics and home repairs became next to impossible.  It didn't matter as we perceived our devices were improving faster than their failure rate.  Prices dropped so rapidly that throwing them away became popular.

 

This replace rather than fix mindset applies to many consumer goods, and many aren't engineered for a long life.  During the pandemic our eight year old refrigerator failed.  I was able to get it working again in about a day, but the repair was anything but easy.  The dishwasher also failed.  A gear on the motor, but the replacement module was $260 and special tools are required. It was eight years old so the leap to a new unit was an easy choice.  I probably wouldn't have tried to repair these things if I didn't have the background of repairing things that broke at home.

 

Ten years ago a mechanical engineer friend thought there might be a market for simple good design and quality manufactured items - mostly home appliances.  They would cost more, but last twenty or thirty years and be easily repairable.   He built a few prototypes, but was laughed at.   It turns out we're mostly ok with disposable.

 

The design integration in most smartphones and watches makes repairs difficult.  If compact, rugged and waterproof are goals, replaceable batteries are an issue even though you may need to replace them every few years.  A cynic might say it gives the user a push to get a new device.  

 

More important is the level of software integration across the system.  Apple's walled garden is problematic for many, but the value of more security and privacy (assuming you don't live in China), and mostly well-behaved apps is worth it to many.  Trading price and flexibility for stability and privacy.   

 

This raises the question of what should we be using?  Cars are much safer and efficient, although more expensive to operate, consumer electronics are mostly better and much cheaper.  Thinking of different ways to do things probably makes sense.  Getting rid of a second car and adding an electric cargo bike if your roads are safe enough is an option.  Three very sophisticated friends have traded their smartphones for simple "dumb" phones.  Well-made and designed clothing that lasts a decade.  Not everyone has multiple choices, but the pandemic seems to have people questioning many things. 

 

I'll end with a recommendation that applies almost universally..  staying away from IoT (Internet of Things) devices makes a lot of sense unless you know exactly what the security and privacy issues are in each case.  And a fundamental flaw in many - they have software flaws and the only way to update is to throw them away and buy something new.  And while it's beyond the main message, some of these IoT devices send information back to organizations that market it.  They turn you and your behavior into a product.

 

 

objects and friction

Randall Munroe's xkcd is a necessary visit for anyone doing science, math, or engineering.   Artistically simple, he often sums things up with an almost poetic simplicity. Some have suggested topic ideas - he's taken two of mine and created beautifully distilled views of the underlying concepts that make me envious of his talent.  And he does it about three times a week.   Every now and again he creates something that resonates across a community.    It happened again last week:

 

 

Anyone who reads a lot of scientific papers recognizes this. Dozens of variations emerged overnight.  Some were very specialized - at least three relevant to astrophysics alone and I only know that because I know the community.    Here's one Abeba Birhane posted.¹

 

 

Serious truth!  A good deal of the community is this clueless.

It made me think of friction and objects.

 

One of the most remarkable features of physics is its simplicity.  While it may be non-intuitive and sometimes requires a bit of tricky math, you can remove complexity and discover underlying bedrock.  Once you understand the simple basics you can restore the complicating bits a piece at a time and describe a vast set of physical processes.  The fact that you can deconstruct and reconstruct so effortlessly is unique among the sciences.  It's why a physicist will tell you physics is the simplest science.  

 

Beyond physics complexity and emergent properties become important.  It turns out that asking "what is life?" is one of the most difficult questions. Working backwards from biology to biochemistry to chemistry to physics breaks down along the way.

But back to simplicity 

Students learning about motion and collisions start with a statement like "assume a frictionless surface."  Concepts become clear and it's easy to describe interactions where everything is understood.  Two blocks colliding elastically for example.  Later on you can restore friction.  Mechanical engineers, working in the real world,  are forced to deal with it.  In some of their systems - say a chain and gear bicycle transmission - it's possible to have efficiencies that exceed 99%.  Frictionless is the ideal here.  On the other hand you don't want frictionless tires or brakes - you'd never start and, if somehow put in motion, you couldn't stop.  The bicycle and road system is complicated enough that friction is good in places.  Fortunately good engineers know how to deal with such things.

 

The concept of "frictionless" has extended to interactions between computational objects and people.  In fact people are often treated as objects that contain a long series of properties.  Constructions like these lack the meaningful emergence that social creatures have. Serendipity and insight are often the result of social "frictions" in our interactions with other people.   We can have simple low friction interactions.  Sometimes they're great as we're not looking for much, but they're generally shallow.  A worry is large scale machine learning can create a faux sense that interactions with machines are deeper than they really are.  And many of these interactions are with other parties who make decisions about and for us. 

 

A few who build large scale social platforms have deep backgrounds in the social sciences and the humanities and think deeply about underlying social issues, but much of it is done without much clue by coders who build things quickly and then iterate to make them better - often more frictionless.  

 

Separately I had two wonderful reports from people at a very creative company.  Fully vaccinated people are back at the office with some precaution and maskless outside.  One reported she's had more insight and exciting ideas in the last week than in the last six months - largely through random talks with coworkers where they could "read" each other.  I hope all of you find the light at the end of the tunnel soon.

__________

¹ She's a cognitive scientist from Ethiopia who specializes in critical race theory - a rather important intersection these days.

 

 

 

the third wave

Electric cars are beginning to take off.  If you need to replace your old car they may be a good choice as maintanence is potentially much less expensive.  They're usually a bit more environmentally friendly than a conventional car of the same size, but not by much. There are much more cost effective ways to lower your carbon footprint. One is replacing local commutes and errands with a bicycle. 

 

We may be at the beginning of a third wave of bicycling in the US.  This time it's a bit different with pedestrian, human powered, and human-electric hybrid bicycles in a category that's called active transportation.  I'm not so sure it will take hold in the US this time - after all, our second wave fizzled in the 70s.

 

The first wave came with the invention of the safety bicycle in the 1880s.  Suddenly cheap transport that could extend your walking range by a factor of three or four appeared and bicycle fever swept the US and much of Europe.  I've written about it a few times and won't say much other than it was an important component of women's suffrage in England. It also lead to widespread mass production and engineering leading directly to the automobile and airplane.  

 

The second wave began in the late 1960s - first in the Netherlands, followed by the US, and slightly later by Denmark.  None shared a common trigger.  The Netherlands was a reaction against the beginning of affluence that came in the mid 60s that flooded the streets with cars.  The narrow streets, children playing and cars were a nasty combination and people reacted strongly closing streets and going back to the bicycles they had used since the war.  The US boom was fueled by the green movement - particularly in university towns.  Denmark's was a reaction to the oil crisis.  The Netherlands and Denmark stayed the course building physical and social infrastructure that favored cycling in cities and small towns. Progress was slow, but by the early 90s the bicycle had become an important piece of transportation.

 

The US started much more aggressively, but fizzled for a number of reasons shortly after the mid 70s. Carlton Reid wrote a piece on the US boom that failed - a short recommended read!

 

Now something has changed - the ebike.

 

A first ride is astonishing.  They're not motorbikes - you pedal and an electric motor adds power to your muscle power.  You can set the level of added power to levels better than any Olympic bike racer. Hills flatten and the distance you're willing to ride doubles on average.  The efficiency of a person on a regular bike is amazingly high - an ebike typically improves it by a factor of two. You're still get exercise and it's much more practical than a Peloton if exercise is the goal.  Currently there are more customers than bicycles..  

 

So they make sense for you?  Probably the most important consideration is safety. Women and children are usually seen as the two indicator species - when at least half of the adult riders are women and most of the children walk or bike to school..   Women and children already dominate bike riding in the Netherlands and Denmark.  Bicycle racks were removed from schools in my New Jersey town as bike riding was considered too dangerous by the school board.  They were afraid of lawsuits if they encouraged bicycle use.  Some of you live in areas with very low traffic densities and safe conditions. For others the question is how to build a safe transportation infrastructure. 

 

You don't want to share the road with drivers who aren't aware of you, don't know what it's like to be on a bike, and may have more legal protections than you.  A variety of approaches can create safe cycling - certainly European counties have taken differing approaches.  A commonality is traffic calming.. speed limits should be 20 mph or less if bikes and cars share a lane.  Bikes on more heavily traveled streets should have physically separated lanes.  Paris has shown this isn't difficult or expensive.  The most cost effective way to come up with real estate for separate bike lanes is to eliminate car parking and the experiment has been very popular to the point where the changes are becoming permanent.  The city is on its way to removing about 70 percent of parking.  

 

Assuming the risk of cycling is low enough for you, it's important to find a good ride as there are a lot of poorly built bikes out there.  A good ebike that should last a decade will set you back at least $2.5k.  These usually have a similar design to a conventional bike  They're great for commuting and just riding around, but consider a cargo bike if you're serious about replacing a car on local trips.  The past few years has seen a revolution in properly engineered  and practical electric cargo bikes.  Good bikes in this class tend to be spendy starting around $4k.  I have a favorite, but these are individual choice and spening time at a good bike shop is important.  

 

a promotion video of a very good cargo bike illustrating some use cases - you can easily carry two hundred pounds of cargo without worrying about balance, power or overloading.

 

Back to that third wave - will it continue or just fade?  Many regions in Europe already see a lot of cycling and the trend will only increase as automobile use is minimized in the cities.   North America is more difficult as the infrastructure and expectation supporting cars and city layouts are somewhat rigid.   Emerging cities, mostly outside of North America and Europe,  have a choice of city layout and transportation infrastructure.  It will be fascinating to see how and where active transportation fits in.  And there are other practical but non-conventional transportation modes - and I'm not talking about self driving vehicles...  

 

 

looking at the algorithmic economy

minipost

If you think about how society and the Internet, you probably know about Joan Donovan of Harvard's Shorenstein Center.  Among other things she's known for rejecting the notion of the "attention economy" as being somewhat gas lighting.  Instead she uses the phrase "algorithmic economy."

Andy Revkin of Columbia interviewed her this past Friday as part of his Sustain What series.  She talks about the possibility of a path to a public interest Internet while safeguarding speech.  Also how disinformation memes develop and need to be watched and how the current control of interests on the Internet fail.   You may or may not agree with her, but she's one of the deeper thinkers in this area and this is a particularly clear introduction to what she's about.  Andy does a great interview job, pushing back here and there to go deeper.  An excellent way to spend an hour.

power different

Recently a friend sent a piece on an energy harvesting device that coverts some of your body's waste heat into electricity.  The description was sparse, but it mentioned the average person produces about one hundred watts of waste heat and went on to say that is enough to power a laptop.  

 

umm..  Both true, but there are some problems ...

 

The average adult metabolizes about 100 watts on average to stay alive.  That's about the same as eating 2000 calories of food a day.   The average adult has about 1.7 square meters of skin area.¹  This means we need to get rid of about 0.006 watts per square centimeter of skin area.  Six milliwatts of power per square centimeter.  The photo of the device showed a ring like affair with some fins - maybe two square centimeters of radiator area.  Let's round that to about ten milliwatts for the device - hardly enough to deal with a laptop, which are on the order of 20 watts or even a smartphone which are a few watts.  That assumes all of the waste heat can be harvested with one hundred percent efficiency.  Here things begin to go down hill quickly.

 

A thermoelectric converter on human skin has a theoretical efficiency of about 5% at room temperature.²  This reduces the six milliwatts per square centimeter down to 0.3 mw/cm².  Even if you covered your entire body, you'd only harvest five watts and you'd be shivering.   And that's unrealistically optimistic as these devices rarely have more than single digit efficiencies on top of their thermal efficiency in this temperature range.  Even optimistically we're down to about 0.03 mw/cm² or 30 microwatts/cm².  Something like an armband might have 100 cm² giving us 3 milliwatts.

 

This isn't reason to abandon hope.  There are hardware devices with sub-milliwatt power requirements. Small energy harvesting devices that harvest mechanical or solar energy also exist and make it practical to put tiny sensors with tiny transmitters nearly anywhere as long as your receiver is nearby. The micro part of the Internet of Things (which has its own issues).  It turns out this is a rich area for college engineering projects.

 

But what about larger power demands where it's inconvenient to change batteries?  Spacecraft that travel past the orbit of Mars generally rely on radioisotope thermoelectric generators (RTGs) because the Sun is too dim to make solar power practical.  An RTG puts a thermoelectric converter on something hot - a radioactive material that decays rapidly.  Ideally it's an alpha emitter so you don't need heavy shielding.  There are a number of candidates, but Plutonium-238 (Pu-238) is the most common American choice with a half life of about 88 years. Not only do these generators produce electricity, the extra heat is used to warm the spacecraft's warm.   The current generation of Martian rover is RTG powered. 

 

There are a number of RTG stories.³  I don't want to go on forever, but the nuclear powered pacemaker and artificial heart are among the more interesting.  

 

A problem with early pacemakers was longevity. Batteries had to be replaced every eighteen months or so necessitating expensive and risky surgery.  In the late 60s and early 70s two types of tiny radioisotope generators were manufactured for use by several pacemaker manufacturers. They had enough shielding to be very safe and would last longer than even very young patients.  They were also very cost effective and a few thousand patients received them. Then, in the mid 70s, the practice suddenly halted.   The worry was devices might not be removed before cremation from patients who had died.  This would destroy the shielding and turn an area in and around the crematorium into a radioactive hot zone requiring evacuation and a very expensive cleanup.  

 

The radioactive heart implant never made it into a patient.  To make enough power to run the heart, the RTG generated about 50 watts of waste heat adding to a normal person's 100 watts.  It wouldn't be bad if it could be spread over a skin sized area, but that would be impractical.  It would be like holding an incandescent light bulb.   One of those interesting ideas that you wonder about afterwards...

_________

¹ The approximation medical researchers usually use is from Mosteller:  (height[cm] * weight[kg]/3600)^0.5  a

² The theoretical maximum efficiency of a heat engine is 1 - Tc/Th  where Te is the temperature of the environment - the air around the device in this case - and Th is the temperature of the radiator - the skin temperature.  We have to use an absolute temperate scale so Th = 310°K and Tc ~ 295°K.  So about 4.8% efficient.. let's call it 5, although it is probably half that with poor coupling to the skin and air. 

³ One could go on for a long time .. the CIA's RTG in the Himalayas is particularly interesting.  And then there's the Apollo 13 RTG that required last minute maneuvers as the limping spacecraft approached the Earth.   The very hot RTG is somewhere on the floor of the Pacific.

 

you're here because dinosaurs didn't have good telescopes

Sixty six million years ago - the end of the Cretaceous period  - an asteroid or comet struck the Earth near what is now Chicxulub, Mexico creating a crater 150 kilometers in diameter.  Huge tsunamis that rose hundreds of feet as they approached coastlines.  Worse, firestorms swept the planet and the oceans underwent a sudden and deep acidification.  Food chains collapsed on land and in the oceans and with them three quarters of plant and animal species vanished.  

Arguably it was a bad day.

If the non-avian dinosaurs only had evolved larger brains perhaps they would have made it to the point where they had good astronomy and a good enough space program.   But nature keeps moving on and the massive change allowed other lifeforms to evolve including us.  In retrospect it was a great day for the human race.

 

Could it happen again?  That one's easy -  100 percent probability.  Fortunately we're at the point where we have astronomy and a space program so it makes sense to think about risk and mitigation efforts.   

 

In 1994 a comet called Shoemaker-Levy 9 broke into 21 major pieces and struck Jupiter.  It had been captured by Jupiter's gravity and had been orbiting the planet for about three decades it put on a show that would have devastated life on Earth.

 

An Earth strike would be different and Jupiter's large size makes it a more likely target it did ring alarms and people started thinking about the probability of something big enough to do serious damage hitting the Earth as well as what could be done about it.  After all - we had telescopes to figure out likelihoods and rockets to do something about it. 

 

On the 1^st of July, 1770 we came very close to a major extinction event - one that probably would have included humanity among the erased.  Lexell's comet came with six times the distance between the Earth and the Moon. Cosmically speaking that's a near miss.  New comets tend to appear anywhere between a few months to a year before they could be dangerous.  But there's much more than just comets out there. 

 

NEOs - Near Earth Objects - are comets or asteroids that come close to or pass through Earth's orbit.   They range in size from a few meters to several kilometers.  Small ones are common and tend to fragment or disintegrate leaving only small pieces for collectors and scientists.  Moving up the size scale you pass through city killers to objects capable of global devastation.  Fortunately the big ones are fairly easy to spot, but over ninety percent of the estimated city killers have yet to be cataloged.

 

A program to funding the construction and operation of new survey telescopes along with retrofitting existing observatories was kicked off about ten years ago with the US and EU taking part.  It's relatively inexpensive and has made some progress, but the Trump Administration cut funding.  

 

So what can you do?  With enough advance warning you can evacuate regions for objects in the city killer size range.  Moving up the size scale means using rockets to intercept the threatening NEO.  What to do with it is non-trivial and depends a lot on its composition and size.  That's an interesting subject for a future post if there's interest (hint: you probably don't want to nuke it) , but an important step would be expanding missions to asteroids of various sizes to learn about their makeup.  In addition to insurance there are technical and scientific benefits that are probably justification enough.  

 

Perhaps it would help us think about preparation and mitigation - the next pandemic and global warming will both make their presence known with much higher probabilities in the next few decades.  

 

 

 

 

frogs, pots of water and the slow application of heat

A few weeks ago Om posted a piece on Apple's new laptop based on their M1 chip.  It's one of those huge transitions that gives Apple a much higher starting point  for desktop and laptop machines - machines that are still very important to many of us. They've been working on this for at least ten years as part of their smartphone silicon effort. People like Horace Dediu and Om (and even a non-technologist like me) have noted the progression of Apple silicon and wondered if it would be deadly to Intel.  For me the biggest surprise has been the software effort - writing was all over the place for on the silicon side. 

 

Zoom's emergence represents a pot that had been slowly warming for decades coming to a sudden boil with pandemic-fueled heat.  It's frustrating with a terrible UX and UI, can be iffy on highly asymmetrical Internet connections that are slow upstream (cable modems for example), causes enormous frustration for many, but it's better than other tools and works well enough.  Last week I described it as the modern equivalent of the <blink> markup in html during the early days of web browsers like Mosaic.  People think it's cool at the time, but it will be laughably primitive down the road.   

 

Remote video tools will get better and may change some habits. At the same time others are recognizing the importance of face to face. One university I work with just completed a survey that found 0% of 1600 students said they'd prefer in-class instruction to remote leaning when the pandemic quiets down.  Already there's a lot of experimentation on campuses and in some businesses.  An animation company saw productivity crater and is working on a better than Zoom suite of tools for internal use. The university I mentioned has found messaging apps like Discord excel in spaces that Zoom handles poorly.  And perhaps we'll get better Internet connectivity.  Price to performance in the US is abysmal, 5G is a largely marketing joke, cable modems are too asymmetrical (upstream is slow) to support serious video conferencing, and the digital divide has widened the already huge education gap.  There's a lot to fix.  It's like cycling - you need to sort out infrastructure and culture along with adding bikes.  

 

I'm confident tool improvements will come but serious leaps need to be taken and I don't have great confidence in the major players.  What I'm more worried about, at least in the US and Canada, is infrastructure. A friend is moving from Colorado to North Carolina and needs a very solid Internet connection.  The only advice I could give was to find cities that offered good fiber connections that we're from telcos.  That mostly narrowed it down to Ting, which left only a few cities. The movement of people from cities to rural and suburban areas is constrained by many forces - this is one. 

 

On to something else..

 

I've argued that virtual reality wasn't very important and augmented reality meant much more than projecting information in your field of view - and that Apple had just introduced their first AR product. 

 

In a way glasses are AR devices - they help you see more than you normally would. The same for microscopes, telescopes, stethoscopes, I have built bat detectors that let me augment my world by adding a bit of theirs, ... the list is long.  But back to my comment about Apple.   I have to admit I was wrong.  Apple's first AR product was the Apple watch, although they didn't know it at the time.  At first it was more of a fashion statement with a weird assortment of apps.  After a few years they learned what it was for.  People saw it as something for improving their health and fitness. Exercise, heart monitoring, and even the state of your body. Jutta has an app that interfaces with other hardware for diabetes management.  We're talking about life altering AR.   The area is growing rapidly and we won't even think of it as AR because we're used to its Hollywood manifestation.  Of course Alexa, Siri and other voice driven interfaces are AR.  They augment our local information space.  Apple's EarPods are just a portal you wear.  There's still a place to visual AR, but the pot's already boiling.  AR is here now.

 

Of course we should think about how these things fit into our lives. That's a serious area I'm not terribly confident about - at least if tech is making the design choices.

 

 

green flight

Two years ago the Environmental Protection Agency reported aviation accounted for about three percent of the nation's CO₂ emissions, or about twelve percent of transportation.  The international figure is about two point five percent.  These may not seem huge, but they've been growing steadily and, post pandemic, should continue their growth.  Making matters worse is the global warming contribution from the cloud-like contrails produced by the jet engines at altitude.   It's difficult to study, but indications are the contribution may double the global warming impact. 

 

A number of approaches have been suggested and are in various stages of testing.  Moving to bio-fuels isn't the greatest solution as growing and producing biofuels has issues and you're still producing contrails.  In the short term some airlines will move to it, but that may be part a PR and carbon credits game as much as anything. 

 

Electric airplanes exist and there are even a few (very) sort haul commercial flights.  The problem is the specific energy of the battery - how heavy is a battery per unit of energy storage.   Turbine fuel stores about eighty times as much energy as a safe lithium-ion battery of the same weight.  Airplanes frequently carry large amounts of fuel - the weight of fuel burned on a cross country trip is greater than that of the the passengers and crew.  The range of a battery powered airplane is severely limited even after accounting for the great efficiency of an electric motor.   We could get a lot closer with exotic batteries that aren't anywhere close to reality, but even then you're talking about a propeller driven (slower) aircraft. 

 

Hydrogen often comes up as the perfect clean fuel - the combustion product is just water.   Although it's the most common element in the universe and accounts for most of your body's atoms, we need hydrogen gas and that's extremely rare on Earth.  Hydrogen gas (H₂) must to be produced and it takes a lot of energy to create it - more than you can get out of it.  Most of it is produced from a petrochemical process that involves carbon emissions.  This is called black or grey hydrogen and, if you manage to capture most of the carbon emissions it's called blue hydrogen.  What you want is green hydrogen - hydrogen produced using electrolysis with electricity from wind, solar or nuclear power.  That seems like a nice use of excess intermittent power. 

 

It's important to think of hydrogen as an energy storage medium - a non-rechargeable battery of sorts.  It has a very low - essentially unusable energy density at atmospheric pressure.  You'd need impossibly large tanks (think about Hindenburg sized tanks) or you can compress it or liquify it and shrink your tanks to something more manageable.  In gas form at a very high but probably safe pressure it needs about five to ten times the volume of  jet fuel. Liquid hydrogen gets you down to about three and a half times the volume.  

 

Engineering is all about finding an optimal solution to several problems that generally interact.  You could convert a turboprop or turbofan to burn hydrogen and see where that takes you.  Turboprops seem to be workable on short haul routes (1,000 km or less) with small passenger loads (under 100 passengers).  They'd be slow and fly at lower altitudes than today's jets.  Turbofans present a number of problems - at altitude they're making contrails and the range and payload isn't great.  One needs to think a bit out of the box.   

 

You might use a hybrid jet fuel/hydrogen engine that climbs to cruising altitude on jet fuel as fuel consumption is much greater in that phase. Then the switch to hydrogen fuel is made when the engines are throttled back at cruise altitude.  You could also design a lifting body aircraft that has a lot of internal volume for liquid hydrogen fuel tanks without being too large for conventional airports.  Tricks might get you across the US with a few hundred passengers, but flights across the Atlantic or Pacific would require something more. 

 

The prop-driven airplanes present other opportunities. A fuel cell converts hydrogen to electricity and electric motors are more efficient that jet engines.  There are a number of little problems that can be overcome, but a fuel cell propeller airliner may just be a short to medium range solution.

 

I didn't go into the NOx problem.  The high heat of combustion in a jet engine creates nitrous oxides when the nitrogen in the air is heated to high temperatures.  That's a problem - particularly at the high altitudes jets fly at and burning hydrogen won't fix it.  The way around it is to use fuel cells and electric motors, but that's for lower altitudes and slower speeds.  There isn't an ideal solution - you just hope to reduce as much as you can.  

 

Airbus is working on the problem - or at least they were before the pandemic.  I can believe the target date for the propeller aircraft, the lifting body is probably more than fifteen years out.

There may be a near term niche for hydrogen powered flying vehicles.  Agricultural and inspection drones are much larger than hobbyist drones.  They're limited to half hour flight times with batteries, but new fuel cell drones with small high pressure hydrogen tanks have appeared in Korea.  They can fly for hours and that may be the first practical opening for aerial hydrogen.