omenti

your partner in efficiency

As an undergrad I developed a serious bicycling habit.  Our club would take longish sixty to one hundred mile rides on weekends. These were tours rather than anything approaching racing so we talked along the way.   One of the members was obsessed with the idea of human powered flight.  His spirt was contagious and several of us volunteered labor and calculations helping to build the first prototype in quest of the Kremer Prize.  It was one of those rare challenges where the goal was just barely possible. 

 

My involvement wasn't that deep, but I quickly learned to appreciate the efficiencies required. Questions arose.  What sort of athlete would be required?  How would power move from muscles to where it could be useful?  What kind of aerodynamic design would be most efficient and what was the flight envelope?    

We knew bicycles were efficient.  A good chain drive was almost 99% efficient and you could match the power output curve of the pilot to the requirements of a propeller. That's when I learned only the pilot's legs would be used... using arms and legs wouldn't work for a couple of reasons.  And what about the nature of the leg motion?  Lots of interesting issues and answers.

 

A bit earlier an article on the cost of transportation appeared in Scientific American.  This is the famous piece that Steve Jobs loved to refer to when he pointed out a person by themselves isn't incredibly efficient, but on a bike they're spectacular.  But why?

 

There are a number of ways to approach the problem, but I'll ask a slightly different question and hopefully cover a bit more ground.  

"Why can you go faster on a bicycle than you can run?"

We have two basic models of bipedal motion: walking and running.  When you walk you straighten your knee which lifts you just a bit as your heel comes up shifting weight to the front of the foot. The other leg swings forward, staying close to the ground most of the time.  One foot doesn't lift completely until the other has landed. You don't move up and down much and most of the work is done by muscles in the upper part of your leg.

 

Running is very different.  Your body lifts further and lower leg muscles become important as the motion makes you rise up on the forward part of your foot.  Runners will go into great detail on what's happening, but basically you're launching yourself with quite a bit of force using more muscles and force than when you walk.  When a foot lands some of the energy used to break the fall is stored in tendons for release during the next launch -  a mechanical battery. Optimizing efficiency is the work of training and genetics. [ An odd thing about the efficiency of running and walking is that walking is most efficient over a very narrow speed range, while running has a much flatter efficiency curve.¹ ]

 

You can go faster running by increasing the length of your strides (or leaps), but there are limits. Your leg is a kind of double pendulum.   It takes energy to accelerate it forward and some energy to brake that motion as it starts to reach its forward limit.  It's worse if your leg is heavy.²

 

Rather than swinging legs a cyclist spins a crank.  The crank's motion is a circle about a third of a meter in diameter. A much smaller distance than walking or running strides.   Circular motion means the cyclist isn't having to waste the kinetic energy of their moving lower leg by slowing it (it does waste a some energy, but not nearly as much). The knees and thighs still need to be stopped and started, but the motion is much smaller.  The end result is less wasted energy from leg motion.³

 

Like running there's an upper limit to how fast a cyclist can move their legs and still generate their greatest power.  That's where gears come into play.  The gears and large rear wheel allow the bike to move a longer distance forward than the distance your feet travel in one revolution of the crank.  And different gear ratios allow you tailor your effort to the conditions - speed, head wind, hills ...

 

On a smooth flat road your speed is usually limited by wind resistance and the power your muscles can develop.  It's amazingly efficient.  With very little effort you can sustain a pace that no marathon runner can keep pace with.   

 

note:  I've skipped over a lot.  Human motion is a rich and not completely understood science, but hopefully this is good enough.

__________

¹ without going into great detail I show a chart here.

² To think about this consider the leg's moment of inertia and the fact that all parts of the leg aren't moving at the same speed. The upper leg moves slower - this is why fast running animals have their muscle concentrated in the upper leg.  Carbon fiber lower legs can make partial amputees faster than an otherwise similar runner with an intact leg.

Note the power your muscles generate increases as cross sectional area, while the weight increases as the volume.  Larger muscles increase in power more slowly than their weight increases.  This is why most competitive distance runners are of average height and very lean. 

³ The cyclist's center of gravity isn't moving as much either and that saves energy.

 

your local replicator

minipost

I thought we were past the hype stage.  3d printing has been around for about three decades and has undergone a couple of "gee-whiz - this is the next PC .. the next Internet" hype stages.  Recently a friend mentioned a futurist was pushing it and a few other technologies as a way out of many of our problems.  It has it's place, but in the grand scheme of things it's still very much a niche player.  Once you get past prototyping or technically difficult shapes you run into the fact that it offers a limited palate of materials, is usually much more expensive than other techniques in volume and often comes with environmental and health issues. 

 

I've been playing with it as a hobbyist for about fifteen years.  I like making things and some shapes are too complex to use conventional tools.  It's fun to play with so give it a whirl if you like making.  Rather than buy a printer go with a service bureau.  Good printers are probably too expensive to justify and you don't want the noise and sometimes toxic fumes in your home. If you have some money to burn spend it on a computer controlled laser cutter or CNC milling machine...

 

The reasons why it's unlikely you'll be printing things at home rather than buying them in the next decade or three would fill a book.  Instead here's a book recommendation that will give you background and is one of those fun references almost no one has heard about.  Making It, Third Edition by Chris Lefteri.  It's aimed at designers who want to learn about manufacturing processes, but is highly readable detailed over 100 production processes at a high level.  It gives a sense of the importance of materials and dealing with scale and budget constraints.  

 

The practical openings hanging automated making are equally interesting.  Clothing is an area that could be huge.  I've written about it a few times and good progress is being made.  It also turns out to be interesting as it has great socio-economic issues.

 

Real innovation is rare.  A fun way to spend a day over the holidays is go to your library and go through Popular Mechanics and Popular Science magazines from the 50s and 60s.  These are excellent illustrations of technologies being hyped. You'll see a couple of things that worked out and hundreds that didn't.  At the same time they give a sense of how popular, and perhaps necessary, building skills were around the home then.

 

two innovations for charging into the future

Electric cars are on people's radar.  Three friends bought new cars in the last month.  Without prompting all three gave reasons why they didn't go with an electric car. It's interesting that we're at that point.  Without going into details a conventionally sized electric vehicle is a poor environmental choice if the metric is lowering carbon emissions per dollar.  Of course there may be other good reasons for going electric and the breed is improving. Battery technology is central to conventional EVs like Teslas and emerging unconventional and disruptive vehicles and change is in the air.

 

It seems every few months there's a news release announcing a dramatic breakthrough in battery development.  The problem is Lithium-ion is a huge family of battery chemistries rathe rather than a specific type of battery.  Hundreds of varieties exist all with different characteristics.  For vehicles you want low cost, high energy density, high power density (how quickly energy can be delivered), fast charging, long lifetime (thousands of charge/discharge cycles), tolerance to temperature extremes, safety and a few other things.  These translate into longer range, faster charging, extended lifetime, safety - all at a reasonable cost.  The press releases usually only address one characteristic - often at the expense of others.  Improving the breed has been a long road and fairly steady road, but two innovations are on the near-term horizon:  solid state electrolytes and lithium metal anodes.  They could change the industry.  

 

The anode is the negative terminal of a battery.  Currently most Li-Ion batteries use graphite anodes. The cathode or positive terminal is where the exotic material science is usually invoked along, but a lithium metal anode can increase the energy density (energy stored per kilogram of battery) by as much as fifty to one hundred percent.  Suddenly the weight penalty of the battery is lowered. It probably won't happen tomorrow, but seems likely on the five to ten year horizon.

 

You've probably seen videos of flaming Teslas and cellphones.  Thermal runaway causes many current Li-ion batteries to burn if they manage to get too hot (usually around 150°C) A lot of careful engineering can minimize this but, like the gasoline in a tank, it's a risk.  Solid state electrolytes are non-flammable. There are a variety of types being investigated including some that charge quickly.  Unfortunately no one has a handle on exactly how to pull it off. Toyota announced they'd have something ready by the early 2020s.  That seems a tad aggressive.  Then again it's possible a few breakthroughs could happen in the near term and they'll be on track.  It raises an interesting question.. how proprietary will the technologies be?  Will there be a class of solutions that allow many players or are we going to see the emergence of one or two very powerful players? 

 

The magic number cited for cost parity of EVs and conventional cars is around $100 per kilowatt-hour of energy storage. Exact internal costs are difficult to come by and there is more than a bit hype, but most educated guesses are in the $170 to $200 range.  This seems to be improving incrementally, but solid state electrolytes and lithium metal anodes in the next decade would disrupt the curve.  An EV with 400 to 500 mile range between charges, a safer and faster charging battery than the current crop and for the same price as a similar car with a gasoline engine. 

 

Many issues that are poorly addressed.  Recycling a vehicle batteries is going to be necessary, but is currently basically non-existent and a difficult problem.  Then there's beefing up the grid - particularly the low voltage local distribution component as electricity demand will rise significantly and could double in areas with high electric vehicle density.  Issues that will take time and money to address, but nothing insurmountable.  

 

In the meantime use whatever vehicle seems most appropriate and drive it less.  Better yet use a bike or ebike to offset some of your normal driving.  There's a short list of other environmentally meaningful steps you can take from flying less, to eating less meat (particularly beef), to voting global warming denying politicians from office. 

 

hacking creativity and collaboration

A recent Bell Labs reunion celebrating the 50^th anniversary of Unix gave me a chance to visit with old friends I hadn't seen in a long time.  Some of the people had moved into industry (Google was a popular landing place) and others found university posts.  I found myself thinking about the differences between the evaluation process of major research universities and the old Bell Laboratories.

 

In a run that lasted about five decades Bell Laboratories - call it Bell Labs Classic - was arguably the most applied research organization in the world.  There were certainly warts and missed opportunities and it effectively came to an end about five years after the mid 80s breakup of AT&T, but it was an astonishing place while it lasted.  I'm continually been struck by differences between that institution and almost everything else I've run into.

 

In universities hiring and promotion (tenure and beyond) is strongly connected to ten or so recommendation letters from leaders of the field in different universities. The authors are almost always experts in the same subspecialty as the junior researcher.  Bell Labs was different. There was an internal ranking that extended throughout the research organization. A department head would rank their people and the results were merged with the rankings from the other department heads in the same center. These rankings would then go through the same process for the next two levels up until a ranking of the entire institution existed. An expert in a field might receive fantastic marks inside their department, but do poorly further up in the process as others might be unaware of them. Interdisciplinary work would stand out and the highest ranked people tended to have one solid speciality along with work in centers far removed from their own.

 

I entered the Labs as an experimental particle physicist. There were a number of physics groups, but no particle physics. I was initially assigned to an applied physics group and found myself working with a number of other groups as I scrambled to find my way. In the process I made connections inside and outside the center and found some encouragement to trade help with others.   All researchers were given the title of  MTS  - member of the technical staff.  It resulted in a certain equality that made approaching others easy.  

 

Most researchers weren't interested in management, but there was one small step up.   If you were in the top ten percent or so for a number of years you became a distinguished member of the technical staff and given more freedom. In my case I was initially given a day a week and resources to do anything I wanted.  Those projects were all collaborations far from my home organization. 

 

Guaranteed funding for pure and applied research went away about five years after AT&T's breakup and folks had to attach their work to business units. It’s what killed Bell Labs, but realistically it was the only logical path for the company at the time. Bell Labs Classic could only exist as part of a regulated monopoly, but that’s another story.  

 

Some universities are playing around with this kind of ranking to stimulate interdisciplinary work. I don’t think it works in industrial applied research.. at least not at scale, but it might be worth considering.  Of course there are different mechanisms in other organizations - places like Pixar - but the difference between Bell Labs Classic and research universities is quite possibly the result of collaboration encouraged by a simple ranking process.

 

woodie

an Omenti mini-post

I was shocked and saddened to learn of Woodie Flowers' passing.

 

There's an interesting question you can ask about a school.  Who is the most important person?  Of course it depends on your point of view and what you think a school should be doing.  It might be an administer who has managed to turn the place around,  a remarkable alumni member who does the same, a Nobel Laurette perhaps? and for some it might be a coach who somehow energizes the place.  Is there someone who makes the place special?  Universities mean a lot of things to different people.  Ask me about MIT and I'll suggest Woodie Flowers.  He managed to light *that* spark and inspire more young engineers than you can imagine.  In light of the current ethical problems ripping apart a school on campus, he was a beacon of ethics for decades.  

 

I was lucky enough to have spent time with him on a few occasions.  He was even more optimistic and even evangelistic than what comes over in this video:

 

 

Woodie was like this in many areas.  He was responsible for developing and promoting the concept of gracious professionalism..  from an email a few years ago:

 

The election has been painful. It reminds me that we have a responsibility to help students become thoughtful citizens. Rational thought blended with empathy seems too rare everywhere.

Personal experience leads me to believe our students would be receptive to hearing us address their obligations to society. Dean Kamen founded FIRST (For Inspiration and Recognition of Science and Technology) in 1989. A couple of years later, Dean and I co-founded the FIRST Robotics Competition (FRC) patterned after the creative design exercises I had been running for many years in the Mechanical Engineering design courses. The “2.70 Contest” was anchored by “gracious professionalism,” a term I coined to celebrate the wonderful behavior prevalent among students in that course. They competed like crazy, but treated one another with respect and took pride in helping and teaching one another.

At the first FRC Kickoff, I used one slide with the term “Gracious Professionalism.” Encouraged by Dean, at the second FRC Kickoff, I used six slides featuring the phrase. Since then, Gracious Professionalism™ has become a powerful part of the ethos of the whole FIRST community. FIRST competitions have grown and have millions of alumni. Scattered over 80 countries, young people have embraced the notion that competition and kindness are compatible. At MIT, 10% of our freshman classes are FIRST alumni.

As we make the transition to digitally-enhanced learning and machine-assisted professions, I believe “uniquely-human” will be a powerful differentiator. Professionals who understand the laws of the universe and also have an emphatic response to others will be the leaders. Those who think science is a la carte will have aspirations that Mother Nature will not tolerate.

MIT students will have an obvious advantage because of their understanding of what is possible. We need to make sure they are also well equipped to manifest creativity, leadership, good judgment, and ethical behavior.

I offer Gracious Professionalism as a convenient label. It blends rigorous adherence to the laws of the universe with the human qualities we hope to see as those laws are applied by our alumni. Giving back with conscience. Blending hard knowledge with soft feelings. Facts, feeling, and fairness. The FIRST community, which includes all ages, have embraced “GP” and actually compete to “out GP” one another.

How might Gracious Professionals have voted?

Woodie Flowers

 

I'm not an engineer, but am amazed what the good ones do.  This is a big loss for MIT, engineering and the world

 

canada's beautiful new $16 million camera

My Grandmother King gave me my first camera for my eighth birthday - a Kodak Brownie.  My parents had a camera, but I wasn't allowed to use it.  This was my camera.  I also received two rolls of black and white film and a voucher for developin.  Of course I did what most eight year olds would do - I took pictures of the sky, the side of the house, my sister, the dog - twelve exposures in all.  A week later the prints came back and you can guess how bad they were.  The second roll was a bit more better, but I never developed the artistic skill.  But I did I did become fascinated with devices that capture images and have designed and built a few over the years.  

 

Some of you are excellent photographers.  My friend Om happens to be just such an artist and I believe he is writing a book on the subject.  My sister Corinne is an artist who uses photography and computers to offer quirky commentary.  I'll be content with my simple snapshots and videos enjoying the art of others.  And I'll think about the process and tools.

 

The term camera conjures up something that focuses light on film or an electronic image sensor where an image is captured.  Still or moving images, but almost always visible light.  Currently the definition is being extended to computational photography - the processing that goes on in your iPhone for example.  I stretch the definition a bit.

 

For me a camera is something that captures some event or series of events in time.  It doesn't have to be visible light, but any form of electromagnetic radiation and or something a bit more exotic - like charged particles or the shaking of space-time.

 

I first came to this way of thinking on Sulphur Mountain near Banff, Alberta.  I was almost thirteen and had gone over to a curious little shack on one of the mountains twin peaks while my parents had lunch at an observation deck a half kilometer away.   The two young men in the shack turned out to be cosmic ray physicists.  They patiently explained in kid language how they captured muons.  What they were really doing is capturing and recording part of a particle shower that came from cosmic rays striking the atmosphere twenty or so kilometers up.  They'd run these recorded events through some calculations and  would be able to say something about the shower and from that perhaps something about the Universe.  After listening I said something like "you've built a cosmic ray camera.  That tape is your film and you turn it into something like a picture"  They nodded , smiled and showed me how an image could be abstract and still give a lot of information. That day would change my life. 

 

Since then I've been around my share of particle physics detectors - essentially cameras that capture the aftermath of violent particle collisions at places like CERN.  I've also used optical and radio telescopes that, typically after a lot of computation, produce real or abstract images of places and events back in time.  It still astounds me to look at parts of the Universe as it was twelve or thirteen billion years ago.  And for fun I  even designed a homemade digital camera back in the 80s. 

 

Cameras capture images that can be meaningful.  Images that can be artistic, historical or even offer evidence of Nature's workings.   Some of the techniques are familiar and straightforwards, others are sometimes deeply computational. And if you consider the eye-brain combination a camera, as many do, there are surprises  We usually think of an image falling on an imaging surface and being turned into a more of less faithful representation of the image.   In fact there is a lot of processing that goes on between the front of the retina and the brain.  What eventually turns into an image has very little in common with what fell on the retina.  Our brain stitches together information, often out of order and spread over several seconds, to construct an approximation of reality we think of as the real thing and a sense of "now" that is equally artificial.  The comic strip is a nice example of something that is temporally impossible that happens in your brain all the time.  It turns out temporally out of order processing often happens in computational photography.    And then there are many instruments you might not think of as a camera that are much closer to the classical definition.. things like MRI, ultrasound imaging, radio telescopes and the like.  

 

Back to this $16 million Canadian camera...  

 

CHIME is a radio telescope at the Dominion Radio Astrophysical Observatory near Caleden, British Columbia.  It consists of four 100 meter by 20 meter upward facing semi-cylinders that are sort of like snowboarding half-pipes.  Sprinkled up and down the foci the semi-cylinders are 1024 radio receivers that operate between 400 and 800 MHz .. covering a frequency range similar to that of UHF television - a frequency range was chosen to detect a signal hydrogen emits.¹  That's where the name  comes from - the Canadian Hydrogen Intensity Mapping Experiment.

 

An enormous amount of computation takes place in several steps after the radios receive the signal.  The first step sets the instrument apart.  You may have a wifi router that does beam forming.  Basically you use a small array of antennas and the properties of radio waves to construct a "beam" to the source.²   CHIME creates 1024 beams listening to over 16,000 frequencies (think of them as colors) all imaged 1000 times a second.  While beam-forming adds to the computational load it dramatically decreases construction costs over a mechanically aimed antenna and for this application is a very clever strategy.  The computing just behind this is equally clever.  Sixteen million dollars to get this kind of data over what will be several decades of life is a bargain. The electronic and computer designs are elegant and if you're into that sort of thing you should look into the details.

 

The hope was it could detect an unusual recently discovered event called a fast radio burster.   At first some suggested the short radio bursts  of a thousandth of a second or so might be signals from extraterrestrials as no one had a compelling physical explanation.  They're very rare and transient.  Before CHIME only one burster called FRB121102 has been seen to repeat sending out about 100 irregularly spaced bursts since its discover in 2012.   A few hypothesis have been made, but studying a single event can lead you astray.   Then CHIME switched on for three weeks last Summer and thirteen new events were detected even though it wasn't fully functional completely.  This is far and away the most sensitive camera in the world for studying these events.  It is also well suited for mapping - making images of - hydrogen clouds in a much younger Universe, studying a special class of neutron stars called magnetars, and answering a number of other astrophysical and cosmological questions.  There's the added bonus of looking at the sky with a new type of camera.  You tend to discover things you hadn't imagined.  

 

 Human beings have become remarkably adroit at extending and even going beyond their senses as we try to understand the world.  CHIME is easily one of the most beautiful cameras ever made.

__________

¹ The 21 centimeter line of hydrogen is a great signal for studying the distribution of hydrogen atoms in the Universe.  It would be far too short to be detected by the telescope, but distant hydrogen is red shifted down to this frequency range..  Since the emission line is very narrow and well known, you can calculate the redshirt and therefore the distance of the gas.

² Beam forming is done by changing the signals phase at each antenna in an array in such a way that constructive and destructive interference create a pointed beam. It works well for wavelengths not much different from the antenna size.  Wifi base stations aren't very accurate, but they don't have to be. You can also do it with sound leading to some interesting pranks.

 

baseball and machine learning

A few weeks ago two of us were talking about the perception of "now".  A fusion of sensory inputs, responses and thoughts takes place and a highly edited version is assigned in what we perceive as a smooth and linear flow of time.  Consider vision.  It turns out it takes us around three tenths of a second to put together a meaningful image from the time light hits the retina. First we note color, a bit later motion and finally form.  No big deal unless you have to react very quickly.  For most of us that isn't a big deal.  For athletes it is.

 

Baseball is useful as the time it takes for the ball to travel from pitcher to bat isn't much longer than the time it takes to process an image. But baseball "works."  To examine vision in baseball consider the great women's fastpitch softball player Jenny Finch.  A few years ago she broke what the best Major League Baseball hitters are capable of. 

 

Their problem is they don't know how to "read" her.  Beginning as little kids  they learned how to hit a ball getting progressively better over time.  Early on the time of flight for the ball was so slow that human vision and reflex speeds weren't the major issue.  With practice they began to learn how to read the pitcher.  Posture, arm movements and so on.  The shards of image they formed gave them information on the physics of the ball's flight path.  They were learning all of the patterns even though they thought  that they really sensed when the ball was released and when it hit their bat - after all, that's what their mind told them.  But what their mind put together wasn't accurate.  In reality they were linking reads and other early information with ball trajectories.

 

Jenny comes along with a pitch that is over ten meters per second slower than what they're used to, but the information they receive doesn't link reliably with ball trajectories.  They either try to rely on reflexes missing the ball entirely as it flies past before they can swing or they use a mistaken trajectory from their mistaken read and strike out. They could learn... give them a lot of time working with women's fastpitch experts and they'd finally build the appropriate library.  Ultimately some of them might even play the game well.

 

There's a striking similarity with machine learning.  You show a program a huge amount of data trying to be clever and you get a program that can quickly do great pattern recognition.  At least within a restricted domain.   The richness of training datasets is often poor.  Facial recognition programs might have respectable results for white people, but completely fail with people of color.    Such biases of omission are very common, but many other types of bias exist.  (again a rich and deep subject well beyond the scope of an hour). 

 

Machine learning can be useful but is only robust when the limits of where you can use it and the types of errors it makes are well-understood.  There are many areas where it's sort of good enough even though these limits aren't well established. Other areas are more critical.  We're a long way from it being generally useful in areas the tech press considers nearly solved -  self driving vehicles come to mind.

 

Back to the athlete.  Sports are played at many levels.  At an amateur level we play because it's fun.  Beyond that most of us are watching others.  In direct competition we usually prefer watching serious pros.  Beyond the beauty of watching human motion being executed so well there are those magical times where something extraordinary happens.  Athletes at the highest level develop a deep sense of the game and, in addition to being able to react faster than their vision and reflexes would suggest, are tracking and processing other channels of information.  In some games players develop a "field sense" .. they know where the other players are and how they're moving.   You see this dramatically in beach volleyball when played at the highest level and it is common in soccer and basketball.  This gives them the ability to do the impossible. These are very beautiful human moments beyond machine learning.   And it goes beyond sport ..  think of the human experts that just out of the ordinary when the time comes - like Captain Sully Sullenberger landed in the Hudson. 

 

We tend to think about athletes as having some physical gifts and a lot of training.  Certainly those are necessary, but there's an impressive mental component.  We don't say someone is good  at foreign languages because they have a dexterous tongue.

 

machine learning without a net .. from sports to smart toasters

And I don't mean Cylons.

 

Every now and again you notice a few trends coming together even though they hardly seem related at first glance.  Someone asked a question today that turned out to be something I've thought a lot about over the years - a collision between tiny machine learning, privacy and an ocean of tappable ambient information.

 

Twenty years ago some of us were playing with sound. After spending a lot of time analyzing it we wondered what areas existed beyond voice recognition and music. There was a small "name that tune" excursion where we tried to match sung, hummed or whistled bars of music to - well - name the tune.  We were trying to turn the sound into note patterns.  It sort of worked, but the the problem of building a notation library was too difficult at the time.  The easy way out would be a hammer and tongs approach - namely matching played music with a library of played music.  We filed a patent on it, but never pushed hard as we didn't see the use case. Oh well...

 

Next I tried to recognize misbehaving car sounds..  Auto mechanics have great stories of people trying to make broken car sounds.  Unfortunately the problem was not small enough and I shifted gears when a ventilation fan in my office started squealing.  

 

Big heating, ventilation, and air conditions systems mostly run with boring industrial sounds, but a good HVAC mechanic can recognize when something's about to fail.   They also tune these systems for efficiency by ear - for some systems the sweet spot isn't far from the point where problems happen and every installation behaves differently.   With my trust Sony DAT recorder I made recordings of these systems in various stages of failure.  Then some fancy pants signal processing and, lo and beware, you can detect and analyze failure modes.   There was probably a lot of interesting work that could be done, but my organization evaporated around that time. 

 

I was fascinated with low power computing a bit earlier than the sound work.  How little energy does it take to do a computation?  (you can make some interesting entropy arguments and get down to basic physics.)  More practically what could be done with very low powered sensors that didn't need batteries or wires?   The interest took several directions - all good as you slowly begin to learn.

 

Then about five years ago it struck me that the HVAC problem could be done more easily with machine learning.  I've expressed reservations about the misuse of machine learning and regard some of it to be not very different from phrenology, but potentially very useful where you can understand it and its use.¹  I supervised a senior thesis project that attacked the HVAC project using machine learning and a microphone.  It showed some potential, but for me the learning was finally realizing how computationally cheap machine learning can be at the low end.  I also ended up learning a fair amount about machine learning as the best way to learn is to do it yourself and then teach.

 

Given Bitcoin mining efficiency and the massive machine requirements at FaceBook and Google it isn't all together clear, but think about unconnected machine learning on your iPhone and perhaps your Apple Watch.  In conventional computing the CPU deals with a large number of instructions compared to the type of processor used in machine learning.  ML processing chips have a large number of tiny processors that work in parallel. Most ML is just matrix multiplication.  While there can be a lot of steps, each is really simple .. generally just addition and multiplication.  A properly designed ML element holds the numbers in the matrices its working memory cache.  It's much more efficient in time and energy use to keep what you're computing nearby rather than having to constantly read and write it to memory.  You should be able to do a specialized bit of silicon with enough ML horsepower to do some rather surprising tasks while using very little power.  And you can do a lot on a smartphone if some of it's silicon was designed to support ML.

 

I think Apple understands this at a rather deep level.  They're very careful about how to efficiently deal with graphics processors which is shorthand for saying they have optimized how to do huge volumes of very very simple computations.  The chips in the current iPhones are optimized for the task.  They've build machine learning machines you can hold in your hand and, in theory, run ML applications that don't connect to the net and divulge information.  The new measuring app on iOS 12 is an excellent example of a local machine learning deployment. I don't know if it is being done on the watch, but there's no reason why they couldn't. Perhaps the biggest design advantage Apple has these days is silicon. 

 

machine learning in your hand rather than the cloud.

 

I can think of a number of applications that use the sensors on the phone or watch -- from health, to motion analysis (think sports).. why you could imagine wearing a watch and using a couple of iPhones on tripods or held by friends to analyze your volleyball, gymnastic, or whatever moves.  A personal coach or a serious tool is you already have a coach.

 

Exciting stuff, but I suspect it goes much deeper.

 

A back of the envelope calculation suggests you can do useful ML with micropower.  I'm sometimes asked to come up with senior projects in engineering classes (dangerous as I'm not an engineer).  I think you could do the HVAC work with an accelerometer or microphone and simulate the type of compute power that, if done in silicon, could run for months on a watch battery.  Maybe some very simple chemical analysis.   Or recognize a very small vocabulary -- perhaps five words - to control your toaster,  thermostat or anything else.  You could use a key word to make it listen or shine a laser pointer on it.  You could also build a very simple image recognizer that would detect someone's gaze.  Look at your toaster and say medium please.  

 

Way safer than saying medium please to a Cylon.

 

A need for privacy may drive this even deeper and away from the Net.  You can sit down and go through a few dozen applications in an afternoon.  It seems very rich and very latent.    Apple knows how to do silicon and low power.

 

So many possibilities.. perhaps even my dream "big data" project of listening to the woods. 

__________

¹ I've written and talked a fair amount on problems.  It can get technical and I won't spend more time on them here, but it's a serious problem as surveillance systems transition to judgement systems.

 

raindrops keep fallin' ...

A few people sent links to the announcement of a "revolutionary" advance for solar cells - namely to let these large areas that would be unproductive in a rainstorm, be productive.   Numbers weren't mentioned, but it's worthwhile thinking about the problem.

 

The Sun heats evaporates water which rises up into the atmosphere and can condense and precipitate. A question is how much of the energy that went into evaporating the water can be recovered?  If rain could fall without resistance it would be a lot.. also walking outside in a storm could be a lethal activity.  The simple physics is the moisture in the sky has a considerable amount of potential energy relative to the ground it will fall on ..  as it falls the potential energy turns into kinetic energy .. you can work out some numbers if you like.¹  But that's not what happens.  First think about it.. a sanity check ..  are you feeling the force of rain when there's no wind?  What does that say about power density?

 

Raindrops run into air resistance and their terminal velocity is low..  fast ones are traveling about 8 meters per second - about 18 miles per hour.  You can throw a ball a lot faster than a raindrop.  So now the problem is simple.  You don't have to know how far it falls.   So here's a sample calculation I did in my head when I heard about this.  The kinetic energy for one centimeter of rain falling on a one square meter area is 320 joules.²  That isn't much ..  about 0.09 watt-hours per square meter.  One way to think about it.  If you had a 100 square meter array (large for a house!) you'd average 9 watts during a one hour storm that put down a centimeter of rain .. an inch of rain would get you into the 25 watt range ..  less as the converter won't be close to 100 percent efficient, but maybe enough to run a small LED lamp during a strong rainstorm.

 

 On my rowing machine I regularly convert power from my muscles into mechanical power at a flywheel averaging about 150 watts or so.   A good generator could convert about 90% of that into electricity.   A photovoltaic in cloudy Germany averages about 40 watts per square meter across the year (including night) - that includes the conversion efficiency.  Problem solved..  this is impractical for solving the general energy problem... no way would it average out periods where other forms of renewable power aren't working well.

 

So now for the more interesting question.  Why were people excited?  I think there may be several things going on.  

 

First it appears a group did make an advance in turning mechanical energy into electrical energy with piezoelectrics.  Traditionally this is very inefficient - one percent or so.  They've made an advance and it could be very useful for niche applications like low power electronics located where it is impractical to run power cables or replace batteries.  Aces for the team.

 

These announcements often fall into the hands of university or corporate PR people who don't understand what the invention or discovery is and don't ask questions or get sign off. This turns out to be a major problem even at name universities and often worse in the corporate world.  I don't blame the PR people as getting this right is a non-trivial problem.  The announcement is picked up by a tech writer who lacks a clear idea of the announcement, but thinks they understand.   They write something right out of the telephone game.   Meanwhile the latest version is picked up by social media or others who are too lazy to check things out.  The reader needs to learn what sources are trust-worthy and, if the work is important to them, figure out how to drill down.

 

There's a lot of interesting physics thinking about rain and even liquid water,  And water for energy storage can be inexpensive and enormously practical if you have the right location (a mountain lake for example).  Denmark cleverly uses Norway as a storage battery for wind energy to even out the load.  This may become less practical for the Danes as Norway is seeing a larger demand for electricity at night recharging electric cars so it will come down to economics.

 

__________

 ¹ gravitational potential energy mgh becomes kinetic energy mv²/2

² E = 1/2 mv²   where the mass of the water is density times volume.  The density is very close to 1000 kg per cubic meter.

 

the next step in home audio?

 Still cold from the UPS truck the Apple HomePod woke up when I plugged it in and asked my iPhone for a dance.  They spun about and chatted in their local network for a minute or two before Siri introduced herself.  It was completely unexpected surprise from a great friend who  shares the love of music.  A day into the journey I have a few thoughts.

 

I'm a poor musician, but find it an important part of my life.  Grad school was at Stony Brook.  Besides an excellent physics department one of my reasons for doing grad school at Stony Brook was its proximity to music in  New York City..  That wasn't quite enough, so two of us founded a concert series at the university that thrives to this day.  At Bell Labs it became apparent that music might be a significant driver for Internet use and I found myself walking several paths.  One was sound field reconstruction.

 

Sound field reconstruction is just the fancy way of  recreating the sound where it was recorded or engineered in your home.  In theory you record the acoustic signatures of the recording studio and your living room and create a mathematical operation that makes your living room sound like the recording space.  In practice it's approximately hard.  Sound waves are fairly large and get reflected, absorbed and transmitted by and through everything in your room.  Making matters worse the sound field at different spots in the room is usually very different and things in the room move between and during listening sessions.  We made serious progress, but it took exotic microphone and speaker arrays as well as state of the art (for the time) computing to make it sort of work.,

 

One detail is centrally important.  Imagine sitting before a simple speaker cone.  Most of the sound comes straight at you with the volume falling off as you get away from the speaker's central axis.  Speakers have very characteristic patterns showing how loud they are depending on the angle you're sitting at as well as the frequency of the sound they're producing.  If you have two stereo speakers there is usually a sweet spot where the sound reproduction is optimal.  In this area the sound field can be fairly good and you can close your eyes and visualize where the musicians are.

 

Imagine an array of speakers, each with its own amplifier, all controlled by a computer, you can control the phase and amplitude (loudness) of each speaker.  The waves they send out interfere with each other as they expand outwards.   By controlling the phase and loudness (amplitude) of each speaker you can make them interfere in such a way that they create a new wave that can be aimed.  The little animation shows an array of four speakers in a row, but it can be any number and they don't have to be in a line.  The HomePod uses seven small speakers arranged in a ring.  

(you may have to click on the animation to start it)

 

The HomePod also has a ring shaped array of six microphones.  They listen to the room (sample it) at a fairly high rate and decide what parts of the music, from its stereo signature, need to go where and how it needs be modified to interact with the room and everything in it.  An Apple A8 processor, the same two billion transistor processor used in the iPhone 6 and 6 Plus, handles the computational chores.  This is much more power than we had back in the day.  Although I don't know exactly what they're working on, several people I know who are real experts at this have been at Apple for years.

 

 

The HomePod also has a computer controlled single woofer with its own amplifier mounted vertically to handle the low notes.  

 

I'll leave the specsmanship, mechanical and most of the UX discussion to others and just talk about what it sounded like.  I'm not golden eared, but rather a serious listener.   I focus on the music and let it take me where it pleases. 

 

I listened to works across several genres with the HomePod placed at about ear height in a small acoustically complex room.¹   For very small groups - soloists and piano quartets for example - the sound stage was excellent ..  about as good as my stereo pair of conventional speakers (about $3,000 in speakers).  The HomePod didn't match the reference speakers when I was sitting at the sweet spot, but was better when I moved around the room.  That is astounding. 

 

It was creating stereo -a  term has nothing to do with a pair of anything, but rather three dimensional.  Moving  beams of audio around it was painting a sound field in my living room. 

 

As I listened to larger ensembles I found a less accurate sound field.  That said it was still comparable to my reference pair when I was away from  my very small sweet spot - its effective sweet spot was larger.   Note for some pieces of music a good $500 pair of speakers would do a better job if properly installed, but  would love to see what a pair of HomePods placed about two meters apart would do.  Apple has promised the feature will be available later this year.  I've already earmarked the funds for immediately buying a second unit then.  I suspect a pair of HomePods will blow my reference pair away when it comes to sound stage placement.   That will be even more true for those who haven't carefully adjusted their current speakers with test recordings and a spectrum analyzer. 

 

You can control it from an iPhone, iPad or Mac, but Siri is the intended interface.  I've had poor luck with Siri on my iPhone, but here it works much better.  It's also much more robust to interference.  Microphones beamform too and it's clearly listening to you doing its own cocktail party trick.   That said it, and every other interface I've seen are just wrong when it comes to classical music.

 

Comparing it to the networked speakers from Amazon and Google is silly.  Their speakers are terrible and useful only as acoustic wallpaper you listen to in the background.  They're really microphones in your home listening to you assuming you agree to Google's  or Amazon's value proposition (I don't).  Privacy is vastly different - I'll go with Apple's differential privacy rather than Google's "you are the product" flavor.   But if music is secondary and you want a digital assistant with a speech interface, Google and Amazon are the ticket.  Apple is going after the other audio companies along with making its platform stickier in the process

 

I suspect the speaker will be like the Apple TV - fairly stable hardware with the bulk of improvements coming through software updates.   My list of UI and UX issues is growing, but I suspect this will become part of my life. 

 

And a final caveat - speakers involve personal taste.  YMMV

 __________

¹ I'm very familiar with these pieces.  With one exception  I've heard them live and I've even been involved in the recording process on two. 

Ecstacy  Anonymous 4 and Mike Marshall

Us  Regina Spektor 

Seven Swans Al Petteway

The Heart Asks for Pleasure First  Ahn Trio

Black Horse and the Cherry Tree   K.T. Turnstall

The Littlest Birds  The Be Good Tanyas

Submarines The Lumineers

Pipeline (electric and acoustic versions) The Duo Tones

An Spailpin Fanach Connie Dover

The Heroic Conditions of the Universe Pt 7 After the Storm Alexandre Desplat

Turn it Around Lucius

Jolene Mindy Smith's version

Vespro della Beate Vergine: Domine ad Adiuvandum a 6 Monteverdi

Shchedrik Kitka

Our Town Iris DeMent

Sun Will Set Zöe Keating

Don Giovanni Madamina, IL Catalogo and Questo Mozart  (sort of a sign of politics)

Canon a due Violincelli in D Major Gabrielli

Cello Sonata in D Major Op 69 Beethoven

Suite No 6 in D Major  JS Bach (Yo-Yo Ma)