omenti

solar powered hurricanes and power densities

I was walking in our woods over the lunch hour thinking about the superstorm almost exactly a year ago.  There is still a considerable amount of damage including our part of the grid which is patched together rather than robustly upgraded.   It is bright and sunny - enough to make you think about how solar energy drives our weather.   It is a good excuse to chat a bit about energy and power.

 

For a quick exercise I worked out the power associated with the hurricane - at least a quick back-of-the-envelope approximation as more accurate work requires much more information than I have along with sometime more than the simple models I came up with. 

 

There are several ways to look at the problem.  Most people associate hurricanes with their winds, so you can calculate the kinetic energy of the air mass.  I'm omit the details here, but a category 3 storm is in the vicinity of 2 x 10¹² watts - two terawatts.¹  The total electricity generating capacity of the United States is a bit over a terawatt and we use much less than our capacity.  An enormous amount of power, but how do you harvest it?

 

It turns out the kinetic energy of the wind is much smaller than the energy released by rain and cloud formation.  A quick estimate of the power released by the average category 3 hurricane is 6 x 10¹⁴ watts - 600 hundred terawatts and several hundred times greater than the total electricity generation capacity of humanity..  

 

Huge gee-whiz numbers and we need some way to make sense of them.

 

A few years ago I was looking at ways to express some of these concepts and scales in a way that made sense and could easily be compared with personal experiences.  I came to the conclusion that people think in terms of power (the rate of energy transfer) rather than energy.  We buy energy - calories of food, kilowatt hours of electricity, gallons of gasoline - but we are usually more aware of how quickly it is used.  The average person gets about 100 watts of power from their food (and gives off most of it as heat).³  You might say a person really does light up the room. 

 

To compare power from different sources we use a power density and define it as power divided by area.⁴  The power density of a person's metabolism would be about 50 watts/m².

 

Congratulations for hanging in there.  Now we're all set to make a few comparisons and even find some natural limits.

 

The power density of the wind in our hurricane is 2 x 10¹² watts/(60km)² or about 550 W/m² for the core of the storm.

 

The Sun powers supplies most of the energy on earth (geothermal and nuclear energy excepted).  Just outside the Earth's atmosphere it delivers an average of just under 1350 W/m².  If you correct for the Earth's rotation, shape and absorption and reflection, an average of about 170 W/m² makes it to the surface. About three times that of a person and a third that of the wind energy of the core of a hurricane.

 

Green plants are the most important mechanism for collecting sunlight and turn it into stored energy - at least as far as plants and animals who eat them are concerned.  It turns out most plants are only 0.1 to 0.5% efficient - the 170 W/m² of solar energy is turned into a few tenths of a W/meter² of stored plant energy.⁵   A family getting all of their energy from plants would need about 10,000 m² of crop area to stay alive - a hectare.  Figure several times as much to allow for crop failure, the energy required to harvest and so on.  Eating animals is a very inefficient way to get your solar energy as animals only turn 5 to 20% of the energy they take in as food into food a carnivore can eat - so multiply the amount of land required by a factor of about ten to account for the animal part of your diet.

 

We can look at electricity from solar power.  Skipping the details current photovoltaics average about 5 W/m² in cloudy places like Germany and about four times as much in the deserts of the American Southwest.  The average home in San Diego uses 1000 watts of electricity, so you are looking at a minimum of 50 square meters to switch to solar power.⁶

 

Fossil fuels represent millennia of stored layers of solar energy - a slow concentration of a relatively diffuse source.  The process is extremely inefficient (well under a thousandth of a percent in most cases), but that doesn't matter to us as we're just mining what has been built up over the years.  Of course it gets increasingly expensive to mine as we exhaust the easy sources, but it is a very convenient fuel.  

 

Moving away from alternate energy, power densities can be very high.  Since it is the anniversary of the big storm my mailbox has been filled with ads for fossil fuel powered electric generators.  A one kilowatt generator can comfortably fit on a square meter, so it has a power density of over 1,000 W/m² larger power plants are more impressive.  You can make estimates for any type of power production and use.  

 

And now for a beautiful chart.  This is from David MacKay - an English physicist.  He plots something like power vs population density (changed around a bit to match some units he likes).  The diagonal lines are power density.  He has labeled some of the more important generation methods (fortunately some of my back of the envelopes are close enough to his:-)  You can see what forms of generation might be practical in various locations, the wide variation in power use for different countries and so on.  One striking feature that you can work out is the average power consumption per unit of land area for the planet is currently about 0.1 W/m² (we are currently using about 17 terawatts of power, divide by land area).

 

 Wind power provides about 1 to 3 W/m² of land area.  You can see that some parts of the world can't use wind power alone ..  In theory the best biofuels are about 2 W/m² (current fuels are far from that) - telling you a better use of farming is probably to make food.  A bit of staring at the chart and using production numbers from elsewhere is illuminating.

 

I'm a big fan of solar energy - sunlight is a good thing.  Ultimately we will probably be using it directly or through other alternate energies (wind and hydro for example). It turns out there is more than enough, but you have to be careful about where and how you do it.  We've been spoiled by the thin layer of fossil fuels that have slowly been building near the Earth's surface.

 

All of this is very high level - if there is interest I can go into greater depth in areas of interest. 

 

__________

¹ The dissipation of kinetic energy due to drag per unit area will be (air density) x (drag coefficient) x (windspeed)³.  You then integrate over the radius of the storm.  I don't information of sufficient detail, so I make an approximation.  Most of the punch will be in the inner radius.  I'll consider a rotating cylinder that is about 10% the radius of the entire storm and neglect the eye since this is just an approximation anyway.  The 10% is the part of the storm where you get hurricane force winds and above on an average sized hurricane.  The rest can be neglected as the windspeed term is cubed.  45 m/s winds (about 101mph) and air density and a drag coefficient approximation (from a journal article in Nature) gives the answer.

² This one is a more straightforward.  Just use the latent heat of condensation and calculate the volume of rainfall.  A source on hurricanes suggests the average category 3 storm has a 660 km radius and produces 1.5 cm of rain/day (very light for most of the hurricane and much much more in the inner core).  

³ Energy can be measured using many scales and you can convert from one to another.  It turns out the 2,000 or so nutritional calories an average person eats is close to 2,400 watt-hours, so the power from metabolizing that food obtained by dividing by 24 hours in a day giving us 100 watts.

⁴ The terminology gets a bit tricky.  Energy density is easy - energy/mass or energy/volume.  It is customary in physics to talk about power density in terms of energy flux - the amount of energy indecent per area on a surface per unit time - or power/area on a surface. (historically heat flux).  It is usually measured in watts/area where the area is the collection area.  We make a more general use to look at the power/area that can be produced by some area on the planet.  So the area under a power plant, of a forest, or even a city.  To compare it to a person we consider the surface area of the person, which is close to 2 m²

⁵ Some plants are much more efficient - up to four or five percent, but most of those are not suited for human consumption.  Area is taken to be the most efficient planting area rather than the surface area of plant leaves.

⁶ Lots of asterisks here, but this is just a back of the envelope.  You need to store solar energy somehow and that has varying degrees of inefficiency depending on technique.  There are also additional losses such as transmission.

 

 

for new readers

For those who have come from the Asymcar podcast (#6) - welcome.  

Omenti is published roughly once a week and represents an hour's writing on something that strikes me as interesting to myself and potentially a general audience.  It is not carefully edited, but tries to be more conversational - comments are very welcome!  Subjects vary so use the category section in the right column.

For transportation related pieces try these for a start

° moving our personal atoms around

° moving our personal atoms around 2

A bit more about me here.

 

oreos, speed and internet ground fogs

It is very difficult to communicate science and technology clearly.

 

A few weeks ago I was describing the remarkable ability of a friend to connect with people in technical and non-technical fields and allow communications to flow in both directions.  Trust me - this is very rare.   I said she was a "great impedance match" illustrating my own inability to connect.¹

 

Public science and technology communication is very important and getting more important all the time.  Sadly most members of Congress are scientifically illiterate and the same can be said for many segments of the population where choices need to be made.  A few people work at communicating with varying degrees of success - among them are public relations people in higher education.  I know a few - some are terrifically good at what they do but others put out problematic information.

 

Recently I've run across several posts on a study that apparently said Oreos are addictive as cocaine.  A rather strong statement to say the least.  A bit of searching turned up the source - a press release of a piece of unpublished non-peer reviewed study from a small college in Connecticut.   A bit of searching shows it is all over the web as well as in the mainstream media - hundreds of references.

 

That people like Oreos and will eat more than a few shouldn't come as a surprise.  It takes a great deal of energy to be human - to support our big brains, the way we reproduce and take care of our young, our muscle structure, and the fact that we've managed to live through some incredibly challenging times.  For thousands of years our diets supplied very little fat and sugar.  Both are high energy foods and we tend to crave them to pack on a bit of extra weight to take us through bad times.  

 

Humans tend towards packing fat.  Even a supermodel has much more body fat than any of our ape relatives.

 

We are still evolving, but regular access to steady supply of food has only been the norm for, at best,  a few hundred years.²  It turns out we don't have a mechanism to use less energy during times when we have a surplus of food so we put on weight, but don't shed it.  A rachet of sorts.

 

Fat and sugar are tempting - we're wired to love them and it made great sense for us to scarf them up during those rare times when we had the opportunity.   This isn't an addiction.  Perhaps a  more reasonable question is why does cocaine light up the same part of the brain that turns on when we're getting pleasure from food?

 

Even if the study was good, the PR communication failed as the comparison suggested is wrong.  Worse is the fact that most of the reporting I've seen failed to note that what the PR describes is wrong.

 

long sigh

 

You have to take press releases and their re-interpretation by media outlets with a grain of salt - particularly if the news is outside your area of expertise.   Even excellent science and technogly work is often miscommunicated.

 

Speaking of communication, several people wrote about a David Pogue piece on Nova on speed.  I didn't see it, but it seems he visited a lab that studies running.  He's a slow sprinter, but was given a few tips that gave him a dramatic improvement.

 

Human performance in sport as well as normal walking is an area of interest.  I'm mostly interested in efficiency in walking and running, but also jumping and hitting in volleyball.³  You might think that something as simple as walking or running is a well understood and solved problem, but that isn't true.  There has been a good deal of theoretical work that attempts to model observed performance, but until recently very detailed observations have been difficult.⁴  

 

Running is currently understood at a deeper level than walking.  Among other things the time it takes to reposition your leg is probably about the same as Usain Bolt requires.  Where he excels is his ability to strike the ground with an enormous amount of force - a key to high speed.  Some good models are being developed and we're probably within a few years of relationships that are highly predictive.  At this time it appears that very little input information may be required to drive the models.  

 

You may be able to say something about how people run - and perhaps how they walk - based on a few very simple measurements from cheap sensors.  Sensors that could go into your shoe and talk to your smartphone.  Current instrumented running shoes are not measuring the right parameters and we don't have the right relationships yet, but there is hope we will.  

 

Ah - the Internet of Things^TM:-)

 

In this case I think it will mostly be an internet ground fog rather than a connection to the cloud.  Much can be done locally, although connecting to others with serious expertise may be useful at times. In addition to helping out amateur and professional runners there is hope, assuming walking can be cracked, that people can be helped as they age.  There is an incredible amount of potential upside here and you can get a sense by looking at the sponsors of this sort of research.

 

Much more complex sports like soccer and volleyball are going to be much more difficult, but I'm confident progress for quantifying human motion and understanding it at a much deeper level will be made.

__________

¹ An engineer or a physicist sees an impedance match as something that maximizes the power transfers between two devices.  It can allow an unimpeded flow of signal.  Anyone else, it was pointed out, may see it in terms of the definition of impede and just get confused.

² Not to mention the billion people who don't have enough food now - even though we raise more than enough on the planet to feed everyone.  I'm nearly finished with a great book on human diseases that are largely artifacts of the mismatch between how we live and the fact that we haven't evolved fast enough to minimize them.  Obesity, Type-II diabetes, back problems, breast cancer and so on.  Highly recommended and not a diet book.

The Story of the Human Body: Evolution, Health, and Disease by Daniel Liberman

³ A good friend was a pro beach volleyball player - one thing leads to another.  I spent some time looking at the physics of volleyball flight and the kinesiology of the sport.  At first just to learn a bit more about what she did and I could do, but then a few really interesting problems turned up.

⁴ It takes about a half million dollars of lab gear including very high speed photography  to sort out some of the models and build better ones.  Only a few labs have such equipment, but the influx of money into track events is beginning to make a difference.  The same thing happened in swimming earlier.  Although that is a technically challenging problem, much of it had been solved by people who worry about the flow of fluids around objects like airplanes and boats.

Expect to see the barefoot vs shoe question solved as well as the type of shoe. The problem of double amputee efficiency running on lightweight blades has been solved (guess what? - they are unfair)

__________

Recipe Corner

 

An attempt at Tatale - a sort of plantain pancake from West Africa.  I've had these in restaurants and they can be wonderful.  This is a modification of a recipe that works well, but experiment as you see fit.  I clearly have no idea about what is/isn't authentic or what to serve them with (usually a bean dish in my case), but they can be very good.

 

Tatale

 

Ingredients

 

° 2 very ripe plantains

° 1/2 c cornmeal

° 1/2 tsp finely chopped garlic

° 1/2 tsp sea salt

° 1 tsp freshly grated ginger

° chopped scallion - green and white parts

° 1 tbl parsley

° 1 tsp cayenne pepper (however much you want for a bit of heat .. you could use a wicked hot sauce if you wanted)

 

Technique

 

° Puree the plantains

° in a bowl mix the corneal, garlic, ginger, salt and cayenne to the plantain goo.  Stir in the scallion and parsley

° heat a heavy skillet (cast iron for example) with a bit of canola oil to medium high  (I probably use 2 or 3 tbl)

° plop in about a half cup of the goo at a time (I form them into thick pancake shapes) and cook until golden brown - maybe 5 or 6 minutes.  Turn as required.

 

is it breakeven yet?

About a week ago news spread through the Internet and conventional media that scientists at the National Ignition Facility.  It surprised me, but I wrote it off to poor reporting.  Yesterday someone asked me about this "huge energy breakthrough", so a bit of where things stand...

 

Breakeven is the point where the energy output of a power station is greater than the energy necessary to operate it.  Scientists usually speak in terms of gain - energy output divided by energy input.  Breakeven comes when the gain is equal to or greater than 1.0.

 

In nuclear fusion two atoms come together to form one.  In the process energy is liberated.  Many fusion reactions involve isotopes of hydrogen - deuterium or tritium.  Bringing them together is difficult as the protons in each nucleus are all positively charged and want to repel each other.  Fusion doesn't happen until the nuclei are under great temperature and pressure to force them together.  Stars do this with their mass and the force of gravity. 

 

On earth we can achieve controlled fusion by confining its fuel of deuterium and tritium under extreme temperature and pressure.  The two main approaches are magnetic and inertial confinement.  Both are incredibly challenging.

 

The NIF is an inertial confinement experiment that uses 192 enormous lasers that ultimately crunch a two illimeter pellet filled with liquid deuterium and tritium raising its temperature and pressure to the point where fusion occurs. There are lots of gee whiz statistics - during the laser pulsing the power is about five hundred times more than the total electric power generation capacity of the United States. This is exotic stuff and has sparked a good deal of invention and spin-off technologies.¹

 

What took place at the end of September was a  record shot that produced 5*10¹⁵ neutrons carrying a total of about 13.5 kilojoules of energy - about the same amount of energy as a baseball traveling about 40 miles an hour.  Unfortunately this is only about 0.7% of the energy of the laser pulse.  While it wasn't close to a practical breakeven,  the energy of the X-ray blast was a bit less than 13.5 kJ so that part of the experiment achieved a technical breakeven.  A big deal and celebration worthy.

 

It will take a lot of work to go from 0.7 to over 100 percent to achieve real breakeven, but practical reactor is even farther down the road.  The energy required to fire the lasers is much greater than what they emit and you have to catch the neutrons and turn their kinetic energy into electricity somehow ..  the conventional way would boil water making steam to run a generator.³

 

It would be great to have terrestrial fusion - there is an enormous supply of deuterium and it should be easy enough to find enough tritium to supply enough energy to the world.  At the same time the Sun has been doing it for over four billion years - it cheats by using a reaction that requires a much lower temperature.  Unfortunately its power density is far too low to be practical so we are forced to use reactions that require higher temperatures and pressures.⁴

 

Researchers started serious work on fusion reactors in the 1950s it has always been said practical power stations are at least several decades away.  Currently people talk about 2050, but history suggests that  expect that might slip for technical and practical reasons. Renewable energy sources that are basically solar power (solar and wind) are becoming much less expensive and it is likely that progress in energy storage may make tapping the sun practical long before we get a fusion reactor.  

 

The Sun is about 500 light seconds away you might say fusion is 50 years in the future or 500 seconds in the past.⁵

 

And this post from phdcomics pretty much sums up the sciences news cycle...

 

_________

¹ To call the facility sophisticated is total understatement.  Amazing precision is required - much of the facility had to be invented and custom built for the task.

The lasers are fired simultaneously in a 20 nanosecond pulse (light travels about 20 feet in that time).  The combined energy is about  1.8 megajoules (about the same kinetic energy of a large family car traveling at 70 mph) and goes into a special spherical vessel called the hohlraum, heating it to give off X-rays which strike and crush a plastic capsule surrounding the fuel pellet.  The fuel pellet heats to about 100 million °K (six times the core temperature of the Sun) and a pressure of 100 billion atmospheres.

It should be noted magnetic confinement is closer to being practical than inertial confinement, but both are problematic.

² More properly ²H + ³H -> ⁴He + n where the neutron carries off about 17 MeV (17 MeV is about 2.7 * 10^-12 joules, but there can be a lot of nuclei. Also note chemical bonds are in the single eV range.)

³ The lasers are about 10% efficient and are followed by a step that converts their infrared output to ultraviolet that is about 30% efficient -- overall efficiency generating the light pulses is about 3%.  A steam turbine power generator is about 40% efficient, so the breakeven is farther off than the 0.7% figure suggests.

It is interesting to note that plants average a few tenths of a percent efficiency turning solar energy into chemical energy.  The process of making fossil fuels is enormously inefficient - tens or even hundreds of thousand times worse.  But that doesn't matter to us as we mine what required millennia to produce at no cost to us.

⁴ The Sun's reaction is non-trivial and parts of it proceed very slowly.  If you work out the energy density of the core of the Sun you find it is much less than the energy use density of a human - in fact it is roughly about the same as a lizard.

⁵ An astrophysicist would point out that energy produced by fusion in the core of the Sun takes a long time - something like 100,000 years - to migrate to the surface where it then travels at the speed of light.

__________

Recipe corner

tips rather than recipes this time...

Halloween approaches.  The large pumpkins suitable for carving taste awful - don't even think about it.  Compost everything but the seeds. The seeds are wonderful when you roast them.  Don't worry about separating the glop.  Put them on a cookie sheet (you may want to cover it with foil first) and toss them in oil and sea salt.  Pop 'em into an oven at 325°F and wait until they begin to brown around the edges.  The relatively cool oven lets them cook to their core and they will be wonderful.  The glop will dry out and is very easy to separate at this point.

And that brings us to pumpkin pie.   Use butternut squash (acorn squash isn't bad either) instead for a much better pie.  Don't bother with condensed milk - it was a WWII substitute for heavy cream.  Try using whole milk, half and half or even heavy cream depending on how rich you want the pie to be.  Remember the ginger and nutmeg are best if very fresh.  You'll be surprised how much better a pie with great ingredients is than the traditional canned pumpkin and condensed milk variety of the 1940s...  and you can make great ones will less sugar than you might think (just the thing for Jeri)

 

 

 

 

 

the english carrot crop and my danish cousin

In the mid 1930s the British Air Ministry was seriously worried about death rays - there were rumors that the Germans had developed such a weapon and the notion of death rays was strongly embedded in the popular culture of the day.  Rather than dismiss the notion, the director of scientific research at the the air ministry asked Robert Watson-Watt for his assessment.¹ 

 

WWI was something of a turning point as applied science had emerged as a potent tool.  It was clear to many in German, Britain and the United States that being able to create large applied science and research engineering projects.  Britain, in particular, had taken this very seriously and had a network of research labs throughout the nation.  

 

Watson-Watt was trained as a physicist and spent part of WWI working on radio trying to crack the problem of thunderstorm detection.  After the war he rose in stature and was leading the radio research department at the National Physical Laboratory when the curious death ray request came.  He pointed out such a device, using any current or emerging technology, would be ineffective - perhaps you could raise the temperature of a pilot in an approaching aircraft by an almost unmeasurable amount, but nothing more.  

 

But then the brilliant bit.  He noted that a signal would reflect off the plane and could be measured elsewhere.  By using the fact that radio waves propagated at the speed of light and measuring elapsed time it would be possible to determine the distance of the airplane.  By sweeping a beam you could also determine a heading.  Within a month he submitted a proposal to build what later became known as radar and a few weeks later, before funding had been approved, he was able to show off a proof of concept.

 

The early units required a large amount of power, but it was clear they would be a game changer for air defense.  An airborne radar would be even more useful and a separate project was started to build a very powerful compact and efficient transmitter  that could be carried in an airplane.  The result, and it required only a very small amount of time, was the magnetron.   It not only revolutionized aircraft detection and tracking, but is at the heart of your microwave oven.

The British were worried that the Germans might catch on.  Some details had emerged on the German radar effort, but their units were primitive by comparison and so large as to require a dedicated station.  It was worried the Germans would realize the RAF always seemed to turn up where they weren't expected and give away the advantage.

 

Enter the carrot.

 

It was well known that vitamin A deficiency resulted in bad vision.  A story was manufactured that implied taking large doses of Vitamin A by eating carrots would improve night vision - something that had no scientific basis.  The government had citizens growing carrots and supplying the RAF.  Pilots and civilians believed that RAF aircrews had superhuman eyesight courtesy of the carrots.  The Germans may or may not have fallen for the deception, but the myth endured.

 

The English excelled at going for the low hanging technological fruit.  They didn't have the time or resources for enormous efforts like America's Manhattan Project. Instead they sent relevant physicists to Los Alamos and concentrated on projects that were relatively inexpensive and would have a short term payoff. Some crazy and risky shcemes were tried, but all within fixed limits.  

 

Just in time research - but science has the important side effect, scientists would say it is central, of asking new questions...

 

Francis Crick was a young physicist working on mine detection at the end of the war.  LIke many physicists of the day he had become obsessed with information and decided to bring that background to biology.  As it happened he was something of a disaster in the lab, but had one of those great synthesizing minds and, after teaming up with James Watson, sorted out the basics of DNA starting a major revolution.

 

Maurice Wilkes was another young English physicist who had been tapped to work on radar and operations research during WWII.  Puzzling information became one of his core interests and after the war he returned to Cambridge to start their Mathematical Laboratory.  In the US we tend to have a bias about the early history of computing.  Much of it took place at Princeton's Institute of Advanced Studies - but Wilkes invented microprogramming - the idea that a central processing unit could be controlled by a fast highly specialized program that existed in read only memory - and a variety of memory schemes.²  Some of the early packet network ideas that led to ARPANET and, much later, the Internet came from the Cambridge lab.

 

This brings me to my Danish cousin.  

 

Bjarne is a great friend and brilliant computer scientist.  His Ph.D. is from Cambridge and his thesis advisor's advisor was none other than Wilkes.  Since Wilkes was a physicist, Bjarne decided to do a bit of digging and found his great great great great grand-advisor was J. J. Thomson who, among other things, discovered the electron.

 

My thesis advisor's advisor was also from Cambridge and going back the same genealogical distance one finds J. J. Thomson.  Bjarne noted this in an email where he addressed me as "cousin".    It turns out the path from Thompson backwards marches directly to Newton, but the notion of what science was had changed dramatically by the mid 19^th century into what is familiar today.  In fact the term scientist wasn't in use until about that time.

 

The industrial research science based lab became important in the 30s and projects of a scale and diversity not dreamed of before became national priorities.  A key to their success was the mixing of backgrounds and the emergence of entirely new disciplines.  Some of this continued after the war fueled by the cold war and, in some cases, large corporate monopolies.  That approach has faded, but is showing signs of life again in rather curious places.

__________

¹ The British were better prepared for the overlap of science early due, in part, to Winston Churchill who was a great fan of science fiction and science fact.  In 1931 he wrote a popular piece on the atomic bomb following up on an HG Wells idea and realized early on that such devices were possible in theory.  

² Both of these laboratories were central to the development of digital computing.  I strongly recommend George Dyson's recent book Turing's Cathedral for a look at the American effort at the Institute of Adanced Studies.

 

Recipe corner

 

Honey Roasted Carrots

 

Ingredients

 

° 8 large peeled carrots.  Interesting varieties if possible.  I find organic carrots often are superior

° 3 tbl olive oil

° 1/4 cup honey  - I've also used a good grade B maple syrup

° sea salt and ground pepper - a finishing salt like Maldon is best (Maldon is also English, staying with the theme)

 

Technique

 

° Heat the oven to about 375°F

° Put the whole carrots into a baking dish and drizzle with the oil. Pour on the honey and then season with salt and pepper mixing until the carrots are more or less uniformly coated.

° Bake until tender or perhaps a bit more if you like (I do, but others seem to prefer the just tender stage) - anywhere from 45 minutes to an hour

on big data

The other day someone asked my opinion on Monsanto getting into agricultural "big data"..   People don't seem to agree on what that means, so I'll offer my definition, but first I should note that "data" means something different to someone in the physical sciences than the common usage.¹ 

 

"Big" is also loaded.  Big as compared to what?  I've seen datasets for some Fortune 50 companies that are trivially small compared to a week's worth of post trigger data from the LHC, but they are still sufficient to be overwhelming.  When I was a grad student in the 1970s my thesis experiment generated about 250 gigabytes a data and required a few years to analyze.  That was enormous at the time.  Our grant paid for several hundred thousand dollars worth of time on a supercomputer of the day that had less computational power than my iPhone 5.

 

Big data, for me, is dealing with the collection, processing and analysis of information that is at the frontier of what is possible.  

 

The devil is in the details and lurks at every step.  In the early 90s a few of us looked into the recording and billing system that AT&T used for domestic long distance calls.  This involved call detail records from 4ESS and 5ESS switches made by AT&T and DMS 100 and 200 switches made by Northern Telecom.  Call detail records are simple things - time stamps associated with the beginning and end of the call, some routing information, the type of call and so on.  Much of this had been cobbled together over several decades and too many assumptions had been made along the way.  We found error rates on the order of nearly two percent and were able to sort out quite a bit of what was going on in the system (the main technical report ran nearly 200 pages).  None of this had been properly characterized before and we underbelly missed bits and pieces, but we were able to characterize the system well enough to supply confidence levels on our measurements (we were a couple of physicists and a mathematician). 

 

Model building is extremely important.  You have to understand not only what you area measuring, but what you think it means.  This is can be extremely difficult in the social world and some huge leaps in logic are frequently made.  I've seen very robust work as well as work that is completely off base.  Sometimes there is enough information in your datasets to work with and sometimes other techniques are much more powerful.  Part of the success of Trader Joe's comes from being able to distill customer feedback that comes from their workers and be able to make good use of the information.  Big data analysis is not a big part of their strategy - it doesn't have to be as they are smart enough to make use of other more appropriate tools.

 

There is a danger associated with the misapplication of current and emerging tools.  Too often the underlying mechanisms are poorly understood and sometimes it is difficult or impossible to know what the tools are exactly doing.  The user may be well versed in their field but may not understand the problem at hand, details of the models being used, the quality of the data or several other critical components.  Compounding the problem is many of the newer tools is they have very powerful visualization components.  People tend to be more accepting and less skeptical of beautiful visual design. ...  lipstick for the pig

 

Too many organizations are jumping on the bandwagon simply because they think there is gold in their haystacks and, after all, isn't Hadoop cool?

 

Some have a solid understanding of their domain and can produce results that can be characterized.  I'm guessing there will be something of a bubble with a lot of less serious organizations getting burned in the next few years and a smaller number finding it is worth the serious effort to get things right.

 

Beyond that there are a few interesting things to note.  Particle physics is sometimes called the elder statesman of big data ... a lot of serious work understanding data collection, trigger analysis, modeling and analysis with results being compared with the work of others who were using their own techniques.   The complexity of the debris from a collision event increased with the collision energy of the accelerators and the resulting computational challenges have been matched (so far)  by a combination of Moore's Law and rapidly advancing algorithmic approaches.²

 

The Large Hadron Collider only keeps track of a tiny fraction of the data pre-selected by hardware and software based models called triggers that determine what is important. About three petabytes of data a month were kept going out to a network of several hundred thousand computers in three dozen countries.  The LHC is being modified for higher energies as is the computational environment.  When it turns on the data rates will increase significantly.  But even though this is a lot of data, it is relatively simple.  The events are isolated - there is no correlation between events.  Event correlation is a major challenge in astronomical surveys - the Square Kilometer Array, the Sloan Digital Sky Survey, the Dark Energy Survey and the Large Synoptic Survey Telescope come.

 

The LSST records at about 500 attributes of 20 or so billion objects in the sky.  A time series of these objects are recorded.  Attributes of these objects can change with time and there is a good deal of complexity associated with the early stages of capture and categorization - the transparency of the sky from night to night can dramatically change what something looks like.  Finding unexpected correlations in 500 dimensional spaces is at the edge and beyond what we can do at the moment ..   by my definition this is big data.³

 

A lot of research is going into exotic areas like compressed sensing, maximal information coefficient, and topological analysis and people are beginning to progress.  Some of it will be useful in other domains.

 

As a grad student I spent time with a Swede and a Californian hacking a fax machine to increase its resolution by a factor of about four.  We needed to communicate the sub and superscripts in our equations more clearly.  At the same time people were using the emerging computer internetworks to move computer typeset documents.  This became very common as the 80s went on, but order needed to be made of it.  There was a guy known as TimBL at CERN with an interesting proposal that he later implemented in the form of an HTTP client and server.  All to support the need for particle physics documents - usually time sensitive preprints - be be accessible by collaborations of groups around the world.  Sir Timothy Berners-Lee OM, KBE, FRS, FREng, FRSA had invented the web.

 

Mark Twain supposedly said something like "history doesn't repeat itself, but it does rhyme..."  I suspect folks doing the pure research in particle physics and astrophysics will create tools that will have an enormous impact on society.  

 

___________

¹ Not to go into detail, but information is the more basic term in physics.  Data generally has meta-information attached to it - how it was measured and to what accuracy for example.  Within the field it is assumed that someone has a good characterization of the meta-information when they invoke the term data.  

It should be said that physics is deeply into detail and understanding an apparatus, models and mathematical manipulations are of fundamental importance to the sport.

² Particle physicists learn about the very small by slamming particles into each other at nearly the speed of light.  The amount of energy during the collision is similar to early periods in the formation of the universe and allow the creation of exotic forms of matter. A friend notes that you wouldn't want to have one of us study a car or an airplane.  

³ The complexity is beyond linear it goes as 2^N, where N is the number of dimensions.  This is far faster than improvements from Moore's Law.  There are approaches to reason with this, but much remains to be done.

__________

Recipe Corner

 

A pressure cooker recipe.  If you don't have a pressure cooker and regularly cook, drop what you are doing and get one.  I did quite a bit of asking around and ended up with a Kuhn Rikon Duromatic and wouldn't hesitate to recommend it.  A bit spendy, but it is bulletproof and just works - an acceptable price for one of the most important pieces of kit in my kitchen.

 

Red Lentils and Jasmine Rice

 

Ingredients

 

° 2 tsp olive oil 

° 1 tbl minced ginger

° 2 cloves garlic, minced

° 2 cups jasmine rice (I used a brown jasmine)

° 1/2 c red lentils (you could use other varieties)

° 3 cups water or vegetable broth (really rich if you have a good broth around, but ok with water)

° sea salt 

 

Technique

 

° heat the cooker over medium heat, add the oil.  Sauté the garlic and ginger for about a minute.  Stir in the rice, lentils and add the fluid.

° lock the lid and bring to pressure.  Lower the heat to maintain pressure and let it cook for 18 minutes.

° move pot off the burner and let it come to atmospheric pressure naturally

° season to taste (a bit of salt for me)