omenti

sometimes you need one heck of a butterfly

In the early 60s Edward Lorenz was working on one of the first computational weather prediction models.  Many of the early explorers, new to computer programming, were getting bitten by limits of the machines and software.   Lorenz, the story goes, was running a numerical computer model to redo a weather prediction from the middle of the previous run as a shortcut. He entered the initial condition shortening it in a way he didn't think would matter given the accuracy it was measured to (say 0.504 rather than 0.504103)

 

"At one point I decided to repeat some of the computations in order to examine what was happening in greater detail. I stopped the computer, typed in a line of numbers that it had printed out a while earlier, and set it running again. I went down the hall for a cup of coffee and returned after about an hour, during which time the computer had simulated about two months of weather. The numbers being printed were nothing like the old ones. I immediately suspected a weak vacuum tube or some other computer trouble, which was not uncommon, but before calling for service I decided to see just where the mistake had occurred, knowing that this could speed up the servicing process. Instead of a sudden break, I found that the new values at first repeated the old ones, but soon afterward differed by one and then several units in the last decimal place, and then began to differ in the next to the last place and then in the place before that. In fact, the differences more or less steadily doubled in size every four days or so, until all resemblance with the original output disappeared somewhere in the second month. This was enough to tell me what had happened: the numbers that I had typed in were not the exact original numbers, but were the rounded-off values that had appeared in the original printout. The initial round-off errors were the culprits; they were steadily amplifying until they dominated the solution." ¹

 

Small changes to an input that was iterated many times created a big change in the output. He presented it at a conference causing someone in the audience to comment a single flap of a seagull's wings could change the weather forever.  Lorenz started using that in his talk changing it to the more poetic single flap of a butterfly's wings in Brazil setting off a tornado in Texas.

 

The image resonated with many existing notions about complex systems including science fiction where a time traveler steps on a bug and returns to a vastly changed future.  From there it has managed to become an embedded concept - usually used to describe chaotic and large inefficient systems.  The problem is it isn't always true. In fact it's usually false.

 

Chaotic systems are often less sensitive to change than other systems.  Run a stream of water in your sink.  The motion in the flow is extremely chaotic - no computer could keep up with the details.  Stick your finger in for a second.  The overall flow changed while your finger was present, but quickly returned to its normal chaotic level almost immediately after you took it out. 

 

The climate system is extraordinarily complex. Predicting weather in advance has made incredible progress thanks to better models, hardware, and software.  It turns out there's probably a limit to how far models can predict based on fundamental changes in the atmosphere.  They aren't butterflies.  Butterfly wing flaps are quickly dampened out by thermal motions of the surrounding air.  To make a change you need big butterfly - a really big butterfly ..  the equivalent of several massive thunderstorms taking place in areas you can't predict ahead to the necessary accuracy.  A butterfly flap equivalent of several thermonuclear explosions.²

 

Inefficient complex systems tend to heal themselves.  The Internet, the way food used to be grown and distributed .. almost any large scale process before it was made "efficient" .. tend to be much more robust than efficient systems where "wasteful" redundancy has been removed.  I worry we're witnessing some of that now as the result of a large perturbation to numerous systems.

 

People often come away with the idea that you can manipulate a large complex system if you only can figure out what breed of butterfly to use and exactly when and where to turn it loose.  It's a fool's errand.  Instead you need to study these systems to learn if it is possible to steer them and how you might do it.  The real insight may be you need to change the system itself rather than its inputs.  

 

The same is true for complex efficient systems.  A very small change can crash them.  High efficiency can be the enemy of robustness. Of course systems can be too inefficient - the trick is to find the right balance to build robust systems.

 

________

It's easy to build simple systems where butterflies are very effective.  Lorenz's weather program was simple and easily influenced and it's really easy to find simple expressions that are hair trigger sensitive.³

 

__________
 

¹ E. N. Lorenz, The Essence of Chaos, U. Washington Press, Seattle (1993), page 134

² a few times 10¹⁵ J - several hundred kilotons to one megaton.  A large thunderstorm is in the ballpark.

³ I won't spell it out, but if you're comfortable with differential equations consider y'' - y = 0  (say y(t) is position as a function of time) with the initial conditions y(0) = 1 and y'(0) = -1.  The solution is simple and predictable.  

Now imagine you have a measurement error and y'(0) = -1 ± ε  where ε is small.  Different, eh?  :-)

 

 

the great reset?

I've been on the lookout for signals in this wild storm and offer a few early comments.  They aren't terribly well thought-out, so take them with a grain of salt and in the spirit of encouraging those who are more observant of change. 

 

Society didn't screech to a complete halt when an army of workers and students went home.  Communication services on the Internet are providing connection for most of us, but perhaps the nature of what we're doing is shifting. People seem to be valuing voice or video one-to-one links over text messages and email.  We're spending more time talking to people we haven't connected with in awhile.   Synchronicity with the rich non-verbal communication that comes with vocal inflections and the visual physical mannerisms that we're wired to understand are suddenly valuable again.  It's not as good as real presence, but with the mistakes, pauses, silences,  emotion and even the electric excitement of authentic human conversation it's the best many of us have.

 

Unedited imperfect personas facilitated by a network connection.  Authentic communication rather than the heavily edited unauthentic asynchronous online personas of FaceBook and Instagram.  Apart, but not alone.  What an inefficient use of attention :-)

 

A belief I've held since I began to work with anthropologists and sociologists 25 years ago  is human relationships are necessarily inefficient.  Think about walking on on a sidewalk in the Winter covered with freezing rain and the same sidewalk in the Summer - enough friction is a feature.  The drive for efficiency in many social tools ignores the need for social friction.   We've been trained to be inauthentic online.

 

Are we witnessing a shift?  If so will it last?

 

We need to make choices and perhaps now we're getting an opportunity.  Zoom is clearly more efficient and may be preferable to normal meetings for some, but what is the balance and purpose of your connection?  We should think about that.  I worry that some will see online education as "better" - I won't go into it here, but consider that disastrous for many fields.   And then there are personal assistants - I'm in agreement with Sherry Turkle: we need to think carefully about their place and the tech community is intellectually underpowered to tackle the thorny questions.

 

One more thing.  Some of us are finding a bit more time on our hands.  Be on the lookout for serendipity and seize it appears.  I was lucky this week and will keep looking.

 

Stay well and stay safe!

local mobility

Our family physician sent a note to his patients over 40 recommending cardiovascular exercise and losing excess weight as additional defenses against COVID-19.  They won't prevent you from getting it, but you'll be in better shape to fight it.  Among other things he recommended dusting off your old bike for trips to the store.  

 

I've been an advocate of human and human-electric hybrid transport for approximately forever.  Of course transportation is a system rather than a vehicle and countries like the US and Canada much needs to be done. That said, local errands and commuting are within the range of many.  There's an enormous amount of new activity in and around this area these days. Unfortunately there's a lot of bad information and products that won't meet your needs.  You need a compass to find your way.  I'll recommend two that are gold standards.

 

Horace Dediu has been doing great work promoting micromobility - vehicles from two passenger 500kg intracity electrics down to scooters as well as the physical and information infrastructures that support them.  He has a blog,  podcast, and newsletter. If you're looking for where all of this is going and it's impact on society and business, Horace is the best person around. (I'm biased - he reads this blog:)

 

But what if you're looking for something like a cargo bike?  There's been an explosion of new models, both pedal only and pedal-electric hybrid.  Unfortunately there's a lot of garbage and sorting the wheat from the chaff is difficult task.  Most bike shops cater to people interested in recreation.  They might be great at repair and understanding carbon-fibre frames and aerodynamics, but practical cargo bike experience is hard to come by.  For this flavor of micromobility my strong recommendation is Arleigh Greenwald a.k.a. BikeShopGirl.    Her page has links to her blog, useful how-to podcast, honest cargo bike reviews and more. Arleigh has huge amounts of practical experience and as a mechanic knows how well various models will hold up. She's also into getting kids started on bikes. Just consult with and buy from her if you're close to Denver.  She offers consultations for local people .. if you're out of the area you might try and write and offer to pay something for her specialized wisdom.  

 

Of course you can just dust off that old bike for some exercise to keep from going stir-crazy.  A good bike shop can tune up anything for not that much.  

 

 

a shoe-in mistake?

Imagine Sarah and Melissa up against Alix and April at the Tokyo Olympics this Summer.¹  The score is 20-19 in the second set of the Gold Medal match.   I doubt that anyone will say but what if Sarah was using the same sunscreen Alix uses? or what if they were only wearing another brand of bikini..   On the other hand F1 auto racing and 12 meter yacht racing represent a combination of athlete and machine and in the case of F1 the team strategist may be the most important element.  You have to ask what is important in sport?  What makes for fair competition and how do you compare results among events?  Convention and rules decide what is important and fair.  Sports make distinctions between male and female athletes.  Age classes are common throughout sport and a number have weight classes. Technology has become part of many sports, but generally it's heavily regulated.  

 

except in running

 

Recently the World Athletics, the body that governs many running events, ruled on a Nike shoe with a new rule that basically allows it.  I'm not a runner, but there's a lot of interesting physics and biomechanics going on.  In endurance sports the elites are judged by their athleticism - what are their VO2 max, COT and a few other measures?   VO2 max is roughly the maximum rate of oxygen consumption - think of it as directly related to the maximum amount of power you can generate.  The cost of transit or COT is calculated for about anything that moves.  You can think about it as your fuel economy.  

 

The graph shows what the cost of transit looks like for a person.  Everyone will be different, but the curve shapes are similar for people in reasonable physical condition.  Walking has a very narrow region of high efficiency or low COT. People are very good at finding their optimal walking speed and not straying from it  You can become more inefficient and burn more Calories by increasing your walking speed - it doesn't have to be by much - and get an additional benefit of a more intense exercise.

 

Running is different. Here COT is almost flat over a large speed range until you've reached your maximum comfortable speed.  Reducing your COT will improve your times. Doping generally buys you a two to five percent lower COT.  Of course that's dangerous and banned, but some do it anyway.

 

What if a piece of apparatus lowered your COT? What woud that mean for the sport? Would someone regulate it?

 

Variations of the Nike Vaporfly shoe have been around in development form for over four years. Early on it wasn't clear how much of a difference it made.  Then about two years ago it started to become clear there was a serious advantage using it.  The almost current shoe used in the two hour marathon assault effectively lowers COT by about four percent.  It does so by reducing the amount of energy lost to the ground.  The human foot acts a bit as a spring, storing energy as it hits the ground and releasing some as it rebounds.  It's not terribly efficient, which is one of the reasons you're faster on a bike than on foot. A shoe can increase this efficiency, but until recently the increases have been under one percent - less than the natural daily variation an athlete sees.  

 

It turns out World Athletics, the body that regulates running, basically hasn't regulated running shoes in over a decade until now. Nike uses a curved carbon fibre insert that adds a small amount of efficiency, but more importantly it has a very thick layer of a lightweight foam that acts as a spring.  The amount of energy it can store is proportional to the thickness - in this case about four centimeters.²  The carbon fibre insert gives stability to what would otherwise be a very unstable foam pillar.  Most conventional running shoes have less than a two centimeter platform which doesn't allow enough room for a curved carbon fibre insert .. that and the foam depth would be much lower reducing the spring effect in the process.

 

Tests so far indicate a four to six percent COT gain - this translates into two to four percent better times for a given athlete.   The shoes gives an increase similar to doping.   You might give everyone the same shoes, but there are a couple of issues. Nike patents heavily and not everyone runs for them ..  we have an uneven playing field until other shoe manufacturers can find their way around.  it's quite possible that medals might be determined by lawsuits.  Worse are indications that not everyone has the same COT response to the shoe.  The range appears to be zero to six percent with back strikers and very short contact times doing better. It also works bette for shorter runners as it increases their effective leg length faster than taller runners.  That's probably about an additional one percent effect. A number of potentially great athletes may have been eliminated from the podiums of the world.  Will be be looking at COT, VO2 max and Vaporfly responce as fundamental characteristics of a runner?

 

There are any number of access issues.  Many of the Div-I colleges have shoe sponsorships.  What if you're sponsor is the "wrong" company.  And for non-sponsored programs and amateurs can afford a more expensive shoe.  And the people impacted most - sponsored athletes - can't speak out one way or the other.

 

I don't follow athletics (track and field), but running seems like a rather pure sport.  Of course shoes make some difference, but it seems like that component should be minimized rather than maximized so we can see what the athlete can do rather than some combination.  A physicist might say something like just what is it the sport is trying to measure?  I would like to think it's who is the best athlete. 

 

So many questions.

__________

¹ Two of the best women's beach volleyball teams in the world. Sarah and Mel are Canadian and current world champions.  April and Alix are the best America has.  

² Hooke's Law.. the force needed to compress or stretch a spring varies linearly with distance.  Twice the distance, twice the force.

 

connecting the titanic, sputnik and dealing with forgetfulness

1955, Fred Whipple had just been appointed to head the Smithsonian Astrophysical Observatory in Cambridge.  The International Geophysical Year was less than two years away and he wanted to involve citizen scientists as much as possible.  Whipple was one of the few who realized a satellite was likely and set about creating a web of radio operators to monitor and relay transmissions from satellites along with groups of amateur astronomers who could optically track their orbits.  Operation Moonwatch was formed and by Sputnik's launch over 200 teams were in place. 

 

Tracking a satellite was done with a clock and a low power telescopes that could read azimuth and declination.  The National Bureau of Standards time signal broadcast on WWV in Fort Collins, Colorado used as the reference clock and most people into amateur astronomy had a shortwave radio. Readings from stations separated by hundreds of miles were sent by amateur radio operators to a computer at MIT that gave the project a bit of its precious time.  More clusters of readings would come in from around the world and a model of the satellite's orbit generated an ephemerides - predictions of where it would be. Later specialized cameras were developed by Bell Labs to automatically generate satellite ephemerides as well as tracking ICBM tests, but Moonwatch was a success and lit the imagination of a few people.  

 

In physics it's common to turn a problem on its head to gain a bit of insight.  Operation Moonwatch used known positions on earth and an accurate clock to measure a position in orbit.  What if you put synchronized clocks in space and measured how long it took radio signals to reach a position on earth?  With enough satellites with synchronized clocks you could triangulate a position anywhere on the Earth's surface.  The technical details were difficult details, but the basic idea was simple.  

 

It turns out accurate navigation is very difficult and expensive and a limiting factor in fighting the cold war.  It was so difficult that as technology progressed ARPA decided to push ahead with a satellite based global positioning system - the inverse of Operation Moonwatch.  It was a bargain for what it gave to the military.  Finally a good enough signal was made available to civilians and navigation changed.¹ 

 

GPS doesn't work everywhere.  It would be nice to track things indoors and at a lower cost. Bluetooth and WiFi are currently used -  usually measuring signal strength from several transmitting beacons and triangulate a position.  Sometimes it's just proximity to a single beacon - a lighthouse model.   The approaches aren't very accurate - three to ten meters is common.  Good enough to track your phone as you move around, say, in a mall or subway, but doing better would be nice.

 

This is where the Titanic comes in..

 

On April 15, 1912 the Titanic struck an iceberg.  Frantic radio signals went out, but there was confusion among rescue vessels and people on land trying to coordinate the rescue effort.  Sometimes it takes a disaster to create order and an effort to regulate radio frequencies began soon after. Transmitters were assigned narrow swaths of the electromagnetic spectrum that were large enough to effectively communicate with receivers that could tune to those frequencies. Regulations changed as larger chunks of this resource were needed - AM radio required more than Morse Code, FM radio even more and television much more. Government would parcel out these increasingly scarce resources, but power was also regulated and it was possible to give much larger chunks to low power short range communications - WiFi and Bluetooth are examples.  Most of radio engineering tried to pack as much information into as little spectrum as possible and to build transmitters and receivers that could be accurately tuned to specific frequency ranges  If you waited for some technology to catch up there was another way (actually several, but I'll only mention one) way to do radio.

 

You may have heard about ultrawideband (UWB).  In the mid 2000s it was supposed to be a good way to send large amounts of information over short distances.  Cheaper and higher bandwidth than the WiFi of the day. Often existing technologies are improved with scale and an embedded base making it difficult for new players.  Sophisticated coding technologies and enormous penetration (WiFi has greater penetration than smartphones) cemented WiFi as the prime high bandwidth low range wireless connection.  In a few years UWB went from redundant to lagging. Many thought it was dead, but it had a few advantages in some use cases.   

 

Here's how it works.  Rather than sending out a precisely tuned narrow band frequency, UWB sends out narrow (in time) digital pulses that "splash" over a very wide frequency range.  The signals are usually only one or two billionths of a second long and aren't sent that often - a hundred thousand to ten million a second are common rates. The frequency range is so wide that usually at least some of it can pass through objects that block signals like GPS and WiFi.  It's also so weak that narrowband communication can't detect it.²

 

 Imagine scattering a few UWB transmitters around where you know their location with some accuracy.  Now start transmitting synchronized pulses.  A receiver notices the time of arrival of each signal and, using the fact that light travels about 30 centimeters in a billionth of a second, can triangulate it's own position.  Accuracy is about ten centimeters - about four inches.

 

A use case I'm familiar with is finding the locations of players in a beach volleyball game.  In international FIVB games a very small (less than a half ounce) UWB device goes in the back of each player's  shirt or sports bra.  Their positions are tracked within ten centimeters about a hundred times a second and accelerometers and a compass send out rotation and orientation information on the UWB signal that the announcers, coaches and sports science people use.  Four transmitters are used and all four players are easily tracked.

 

It would also be useful for augmented reality and Apple has included an UWB chip in every iPhone 11 - the greatest deployment of the technology so far.  For now it's being used for data transfer, but wait a bit.  Someone could also make very cheap ($10 or less) postage stamp devices that might be powered by available light that could be location tags.  Put them on your keys, glasses, remote control .. anything you tend to lose track of.  We went through trying to manage dementia with my mother.  I would have paid a lot for this capability.  And thousands of other uses from amateur sports to tracking boxes accurately in warehouses.

 

There are probably some bad things that can be done with it.  We need to be thinking these things through, but our track record isn't exactly good. 

__________

¹ It also gives a very good frequency standard and a timing signal.  Without this our mobile phone system, high frequency trading and a host of other core technologies we depend on would break.  This is an issue as GPS is vulnerable to attack and we've done almost nothing on backup technologies. 

² The FCC defines it as taking up at least 500 MHz of spectrum somewhere from 3.1 to 10.6 GHz.  Power is under -41.3 dBm/MHz .. or 75 nanowatts.  Since it only deposits a tiny bit of power in any narrowband frequency, it's just below the noise floor of a narrowband receiver.  Fortunately it's straightforward to build a receiver that can detect the signal. 

making sense of it

Some of you are probably using smartwatches and fitness sensors for long standing exercise routines or perhaps new routines that began with a resolution a few days ago.  Since regular exercise is considered one of the most powerful and safest medicines known, anything that encourages it can be a terrific investment. There's evidence that competition with yourself or others based on feedback from these devices is motivational for some people. 

 

But what do the readings mean?  

 

Regular exercise is great, but no one knows how much is enough or even too much as there's variation from person to person and careful studies are very difficult. Even the understanding of metabolism is general - it is important to forget the concept of a total energy budget (calories eaten - calories burned) as that requires very careful lab measurements.   What is known is a few hours a week at moderate intensity is a great start.  A heart rate monitor can give feedback on intensity, but there are other effective non-technical methods too.  As the ad says - just do it using whatever works for you.  But what if you are an amateur or professional athlete - do these technical aides give an edge?

 

For four hundred dollars a smartwatch gives you information it would have cost thousands of dollars twenty years ago. The problem is you need to understand what it's reporting in context.  For most of us that's somewhere between very difficult and impossible. An elite runner or cyclist might have access a to multi-million dollar support effort with specialists who try to understand information from a variety of sources in context for that elusive one percent performance gain, but even then they regularly fail.  Generally these tools  are most successful used in training as the complexity of understanding the information changes during competition.  In recent years telemetry from athletes in team sports became possible.  A coach could see the condition of each of their players on the field as well as track how much and how fast they moved. It became popular in football, soccer, rugby and a few other sports, but is being set aside by many as understanding context during competition difficult and it had led to some dramatic coaching errors. 

 

Besides wearables video analysis can make a big difference for amateurs.  Usually a coach or trainer who understands what they're watching and how to make changes is needed - you're not going to find great personalized information on youtube.  A few sports like baseball, golf, tennis and gymnastics are making use of motion capture.  I'm guessing we'll see that make its way to smartphones one of these days (there have been a couple of so-so efforts already).

 

I'll cut it short as probably only a few of you are interested in diving in (for those that are feel free to get in touch - I have an interest in sports science and sports tech)  For everyone else the message is use whatever it takes to find the motivation for regular exercise.  It may be a smartwatch or fitness tracker, but you're probably in the minority if those work over the long term. Find what works for you.  And if you're an amateur athlete the lowest hanging fruit may well be a few hours with a trainer working on form.  There's a tendency to use tech for that one percent where a simple change in form might buy much more.  And an added befit of a few sessions with a trainer is they might be able to spot and correct bad habits that might lead to injury. 

 

There's a message for anyone who relies on numbers for their work.  Where did they come from and what exactly do they mean?   Perfect understanding is rarely possible, but understanding what is good enough and how to find and understand context is essential.  One of the reasons why much of what is being done with big data and machine learning frightens me.  They can inspire false confidence.

 

 

apple announces its first serious augmented reality product

A few days ago Apple announced the AirPods Pro.  It is largely being described as an AirPod with a noise cancelling feature and a higher price.  It turns out to be much more than that and a couple of features give a few hints: transparency mode and how it does noise cancellation.  I haven't played with them yet. My fun money budget is at zero dollars for the rest of the year.¹  In the meantime there are trustworthy reviews from Om Malik and John Gruber. 

 

Transparency mode is what it sounds like - ambient sounds around you are mixed in with those meant to go to the earbuds.  You get a bad version with poorly sealed headphones/earbuds.  Doing it propertly is tricky and  initial reports indicate Apple has pulled it off.  While most people will use the feature to be aware of their surroundings, it can easily become an augmented reality - specifically an acoustic augmented reality.

 

A variety of visual augmented reality devices and apps have appeared in the past decade.  All have issued and none have cracked the mass market.  iPhones and iPads happy to be visual AR ready.  They mix a live image from the front pointing camera with a computed overlay. The result can be cute and even useful, but mostly it's cumbersome and something you try approximately once.  Google, of course, has had a couple incarnations of their glasses.  There are a number of issues and they appear to have found a few niches in business.  At this point they're not exactly consumer clueful devices.

 

Augmented audio is less intrusive and a natural domain.  Your AirPods Pro could be sending you turn by turn directions as you walk or bike through a city.  Hey Siri is their ready to go interface.  It would be a natural for reading recipes or a thousand other things.  Anything where you want contextual information that is an augmentation of the acoustic soundscape around you.  Then you don't have to switch modes completely and take out a device and/or give up the acoustic soundscape.  It would be dandy for seniors -  hey Siri, where did I leave my glasses?²

 

Speaking of senior citizens AirPod Pros have almost all the features needed to make a very high quality hearing aid.  There is a test that uses feedback to find the best fitting silicone earpiece adapter. It woudn't be a big deal to do it over a range of frequencies for each ear and compute the corrections to sound in transparency mode. It would become an active computational hearing aid.   They could probably even correct for tympanic movement (suggestion by Bryan William Jones).  At the very least these could be the equivalent of non-prescription reading glasses and that could be free assuming you own these.  Apple could work on what it would take to do proper testing and get them certified.  

 

AirPods Pro augmented hearing could mean serious trouble for the conventional hearing aid industry. If you've ever had to deal with hearing aids - your own or for someone you help take care of - you find they cost two to five thousand dollars and they don't work very well. These hold the promise of being able to tune to new acoustic spaces in real time.. something regular hearing aids never have pulled off.  

 

Everyone has been expecting Apple's AR glasses.  They may well be coming, but in the meantime acoustic augmented reality can happen now and Apple could easily extend it into owning a new health market while they're doing it.  There are some other computational tricks to enhance sound for folks with normal hearing, but for now I think these are the two big areas.

 

And finally - to answer some emails from frustrated hearing aid users - they did not announce a hearing aid.  It's just the pieces are there for them.  It wouldn't be much of a stretch and there is frustration over price and quality among hearing aid users. 

___________

¹ When we got married we decided to pool our income except for individual fun money accounts. They aren't big, but the agreement is the other spouse can't complain about what they're used for. One of the best things we've ever done.  Most of my funds support time tithing activities.

² Apple has the tech that allows good location (like 10 cm)  using ultawide band beacons that could be very inexpensive and taped onto a pair of glasses for example.  If you've ever had to help a friend or relative with dementia, this is a killer app.

 

happy bob from finland

Several years ago I met Jutta Haaramo though a bit of serendipity.  We share some interests and it's been great fun talking with her over the years.  A bit over a year ago I learned about her efforts to make a dent in the Universe and do something about juvenile diabetes management.  It's a tough problem that consumes an enormous amount of time and worry on the part of the children and their parents.  It also happens to be an area where technology can be used to help out.  

Jutta is amazing - she put together a team and an iPhone app that uses a bit of machine learning and gamification was born.  This is really great stuff and is just emerging into something that can be used.   So if you have diabetes or know someone who does - or if you're interested in serious medical aids that can run on iPhones and Apple Watches read on as Jutta continues with a guest post.

__________

 

My son was diagnosed with type 1 diabetes five years ago when he was just six years old.  In type 1 diabetes (T1D) your pancreas stops producing insulin, so you need to dose it manually to keep blood sugar at a normal level. Keeping blood glucose stable is hard, as everything affects it: insulin, carbohydrate consumption, exercise, stress, other hormones etc.

We started to use continuous glucose monitor (CGM) to follow his sugar level every five minutes. For insulin delivery we use an insulin pump that delivers insulin in small doses throughout the day. Before every meal we calculate the carbs as the insulin is dosed according the amount of carbs eaten

 

 

Diabetes management is really complex. There is a risk of short term complications (severe hypoglycemia i.e. too low blood glucose and hyperglycemia, too high blood glucose), which can both be fatal. But the long term complications, which are caused by elevated blood glucose over time, are as scary. Heart disease, nerve damage, kidney failure, blindness, premature death can all be caused by diabetes. Diabetes also takes a huge amount of time: counting every carbohydrate you eat, calculating insulin doses, changing infusion sites and sensors just needs to be done, no matter how tired or overwhelmed you might feel.

No wonder there are many who know the mental burden of diabetes far too well. But being stressed about your condition does not help you to take any better care of yourself.  Rather, we noticed that humour and positivity did help. Noticing the small wins (“oh, your blood glucose was really good at school today - well done!”) made a big difference.  So we wanted to turn that approach into a new way of diabetes management. One that gives insights to future glucose levels and rewards you when you're succeeding in managing on your own.

 

This idea turned into Happy Bob, an app that uses gamification and machine learning to turn health data into rewarding and engaging experiences.
Users can follow their own blood glucose data and there are stars which can be collected when blood sugar is kept within target levels. The star counts are displayed and a user can achieve daily star targets and share scores with others. The app also uses personal blood glucose history data with machine learning to predict the user’s blood glucose up to two hours ahead. The prediction helps to foresee high and low blood sugar before they happen.

The interface is simple and playful.  Happy Bob is also available in the Snarky Bob mode for those with a more sarcastic sense of humour.

 

The app is currently available on iPhone and is used by both children and adults. Users tell us Happy Bob makes diabetes management more fun and helps them to be more aware of own data. Everyone enjoys collecting stars and getting positive feedback. It’s still early but the app retention is really good.

Next we plan to add more health data in the experience, to provide a more holistic health management experience and more ways to be rewarded. Our goal is to help users foresee periods when they should pay extra attention to their blood sugar and health to prevent the risk of complications.  We also want to add ways to use the stars earned, as that’s something many users have been asking for. We believe that by creating a daily, fun habit of good glucose management with Happy Bob, many people with diabetes can have a better life and health.

 

Happy Bob user feedback

 

First check out this video from Happy Bob user T1Swimmer .. she's great!  It should come up for those of you who aren't FB users 

https://www.facebook.com/T1Swimmer/videos/441338936474643/

 

If you are interested in testing out the Happy Bob App, have an iPhone and wear a CGM (continuous glucose monitor), you can download the app in the App Store https://apps.apple.com/app/happy-bob/id1444877516

For any information, please get in touch with me at jutta@happybob.app. You can follow us on Instagram https://www.instagram.com/happybob_app/ or read more at our website https://happybob.app/

 

My three boys. Finland has the world’s highest incidence of type 1 diabetes. Diabetes is on the rise across the globe. With good glucose control most of the complications can be prevented.

 

 

mad science

It seems Apple released a new smartphone yesterday. This one has several reasons to make me want to upgrade, but the most drool-worthy is its power as a computational camera.  I'll leave it to other to write the reviews and compare it to Google's efforts - a new iPhone isn't in my budget this year.  But perhaps a few words on computational photography are relevant.

 

Computational photography has been around for over sixty years.¹  In particle physics you slam particles into each other and carefully study the wreckage that emerges from the collision. About sixty years ago bubble chambers came into use.  The idea was to record the wreckage in three dimensions.  A chamber was filled with a transparent liquid like liquid hydrogen, kept just below its boiling point. Just before a fast moving particle entered the chamber a piston at the bottom of the chamber was dropped suddenly causing  the liquid to superheat. The incoming particle collides with a particle in the liquid - usually a proton if the liquid is hydrogen - and charged particles from the collision leave a trail of tiny bubbles in the sensitized superheated liquid.   Three cameras at right angles to each other simultaneously captured three views of the bubble tracks and the piston is moved back into position waiting for the next collision.  

 

A large magnet surrounds the chamber causing the particles to bend one way or another depending on their charge.  The slower a charged particle moves, the more it curls.  The photographic sheets are developed and "girls" - it was considered women's work - would trace the tracks by hand with digitizers creating lists of coordinates for each track that a computer program could use to compute the fundamentals of each collision.²  This type of computing was the most demanding of the day.   In the late sixties and beyond recording the collisions directly in digital form using a variety of "counters" took over and the scale of experiments and necessary computation exploded.

 

By the late 70s people were beginning to categorize medical x-rays and ultrasounds computationally and resonance imaging began to emerge.  The computational requirements were stiff in the day, but your phone could easily handle the computing required for your MRI.  It has been argued that this form of computational photography has had the largest impact on human health of any form of computation.

 

Playing with photographic images from telescopes and microscopes was  necessary in science, but the techniques quickly migrated to computer science labs.  Just as the Utah Teapot was the universal test for 3d imaging, a photo known as Lenna was the ubiquitous test for imaging algorithm research in male dominated CS departments during the 70s and 80s.  It wasn't unusual to see her image appear in several papers in a single journal issue.  Journal images today aren't as sexist.

 

Some of what went into academic image processing made its way into Photoshop in the late eighties.  Photoshop went on to be a preferred platform for creative artists to do computational post-processing of their images.  Take a photo on your camera and play with it on your computer.

 

But what if the camera and computer could be combined into something that could redefine imaging?  I'd argue that took place in science some time ago as the imaging part of telescopes, microscopes and particle detectors had began to require built-in computation.  It's been in the past few years we've seen the emergence of a new type of photography in handheld cameras.  Smartphones, with vast development resources, are leading the way and we're on the verge of an explosion of new techniques and ideas.

 

At this point it would make sense to talk a bit about the basics of 2d fourier transforms to get a sense of how that's done.  Here's a short non-technical introduction that gives a sense of a basic function.

 

 

The computational power of the latest smartphones is staggering and, at least in Apple's case, a good deal of it can be devoted to computational photography.  Currently much of it is used to improve snapshots that would otherwise be unusable.  Noise can be reduced while still keeping important detail - often using psychological models of human perception, the dynamic range of an image can be extended by taking a number of shots and bringing out detail in dark areas and suppressing burned out overexposed regions, realtime video can be overlayed with other information creating a computation synesthesia of sorts, machine learning is becoming useful in many ways,  .. the list goes on and on.  

 

I'll offer one that could appear in about a year with the next generation of smartphones.  Time of flight image sensing is now sort of practical for a smartphone and will probably be a feature in high volume phones like the iPhone next year. Basically a pulse of infrared laser light is sent out to an object and the time it takes to go out and return is used to calculate the distance.  This is done for a somewhat smaller image than your camera takes - a few hundred thousand to a million pixels. But these pixels now have distance information in addition to enough image information that it can be used to create a distance map for the full image from the main sensor.   This could be very useful in augmented reality, but here's one I'd like to see.  Take two or three phones on tripods and point them at a dancer, athlete or someone getting physical therapy.  A very accurate motion capture video could be quickly made and you wouldn't have to spend $50k and up on specialized equipment.  Coaches and physical therapists would be delighted. 

 

A thousand other uses will emerge we haven't thought about. Independent of their platforms there are strong reasons why Apple needs a great computational camera.  What will separate winners from losers will be usefulness, playfulness, and great user interfaces and experiences.  That's the hard part. 

 

For the best photos a good photographer is still required and the tiny lenses and sensors on smartphones are limited by basic physics.  But always having a camera with you and one that can correct basic mistakes is a serious plus.  That and there are emerging capabilities that will take photography in new directions.  

__________

¹ digital computational photography.  I won't go into details here, but optical computation has been around even longer.  If you're interested in the subject realize a lens is an optical computer capable of creating a Fourier transform.

² I like this example as a neutrino slams into a protron bangnig the vacuum hard enough that a D⁺ (d plus meson) appears.  It is composed of two quarks:  charm and anti-down.  I have a fondness for charm quarks:-)  Also this was made during the twilight of bubble chambers in the late 70s. 

 

 

 

giving time a hand

Hold out your palm and point it skyward.  Every second about one muon will strike it.  Muons are the most common remnant of cosmic ray collisions with the upper atmosphere that make it to sea level.  They're responsible for some genetic mutations and even cancers, but the levels are low enough that we really don't have to think or worry about them much.

 

When I was about thirteen I came across a cosmic ray experiment near the summit of Sulphur Mountain in the Canadian Rockies.  I've described the experience before - two young physicists had the patience to explain what they were doing to a kid. When comic rays collide with molecules in the upper atmosphere a lot of debris is produced - mostly in the form of subatomic particles.  Some have very short lifetimes and don't make it far.  Muons live long enough to make it down to sea level.  But there's a twist.  The half life of a muon is about two point two millionths of a second.  Even if it traveled at the speed of light it would have traveled about six hundred and sixty meters.  Most of these collisions take place a few tens of thousands of meters up. The trick is when we look at something moving very fast, time for it appears to run slowly.  The number of muons making it to your hand is a manifestation that time is relative. 

 

My mind was blown

People put up Stonehenge type markers in the prehistory to note seasons and other special dates.  Moving closer to historical times a variety of sundials were developed.  Clouds and night made them problematic and, depending on design and the skill of the observer, they were frequently only good to an hour or so.  Even so the some Romans had no trouble complaining about them forcing a structure on life.  The Romans moved a bit further  timing devices in the form of water clocks and hour glasses.  Water clocks were used to limit debate in the Roman Senate as well as arguments in legal proceedings.  But time was still mostly for special occasions and localized.  Most people relied on an approximate natural rhythm from cues from the sun and stars.

 

The thirteenth century saw the invention most associated with modern time. The mechanical escapement is a bit of cleverness that converts an oscillation into a ratcheted advancement in one direction.  In a clock a pendulum's swing causes a gear to turn one notch at a time.  The first use was in the church with pendulum driven bells to call the monks to prayer.  Several decades went by and dials appeared allowing the visual estimation of time and a key driver of the Industrial Revolution had appeared a few hundred years early.

 

Since then mechanical time managed to penetrate deeper and deeper into society.  The Industrial Revolution demanded workers at factories at certain hours.  Home clocks were affordable to some households, but reliable alarm clocks were still expensive enough that the profession of the knocker-upper - someone you would hire to wake you up by a certain time - persisted in England until the twentieth century.

 

Now we have accurate portable time and increasingly an expectation to make use of our time. And now if clocks were limited to an accuracy of a second every century, much of our modern technology would no longer work.  Sometimes it seems as if we're to be optimizing the time we have.  We're admonished - or we admonish ourselves - against "wasting" time.  For many being bored or having idle thoughts is somehow wrong.

 

Our sense time in our brain is very non-linear and irregular and an active area of study.  Almost certainly there is no master clock.  We seem to have individual interactions.   I find I require a fair amount of idle time to daydream, think and make associations that don't seem to happen when I'm filling my time "slots."  I can only be creative if I'm bored at some level.  

 

When I began to work out of the home I made it a point to take a walk in the woods across the street every day I was around.  You begin to notice the cycle of the seasons, and after a few years other patterns emerge.  It's been about fifteen years now and the experience grows richer with time. Some of this Summer seems to echo with one eleven years ago and other parts seem unique.   There are cycles for the trees, flowers, weeds, deer, wild turkeys, coyotes, bats, weasels and so much more.  You experience some visually, some with sound and some at time and audio scales we normally don't notice (I often take along a microscope or a box that allows me to listen to ultrasound.)

 

Hold your arms out side to side.  Imagine the distance from the furthest tips to be the age of the Earth - something like four point five billion years.  Now take an emery file and take a few swipes from the fingernail that is touching the present.   Now look at that fingernail.   You've completely eliminated the period of human civilization. 

 

I worry that humanity is failing the marshmallow test -  where you give a little kid a marshmallow and tell him there will be two if he can save the one he has for wait for your return.  Many key countries can't visualize the generation or two ahead to confront global warming.  Perhaps we'd be better off if more of us moved a bit away from the forced regulation of our clocks and allowed more opportunity to consider the broader world.