TechnologyQuarterly

Biohackers getto work

Safer and faster flights

When missiledefences fail

September 6th 2014

Smartphoneson wheelsHow the connected cargoes driverless

20140906_TQ_SEPT.indd 1 26/08/2014 14:44

“We want a banker who will work with us and be part of the team.”

When you need someone to strategize with, we’ll be ready to talk. Our relationship managers take the time to learn your business and gain a deeper understanding of your expansion goals. We’ve successfully partnered with mid-sized to large corporations to help them meet their global business needs. With our full suite of products backed by our time-tested strength and stability, we’ve never been more ready to support your business today and for years to come. To learn more about how our capabilities can work for you, visit national.wellsfargobank.com/EM20.

* Investment and insurance products: NOT FDIC-Insured • NO Bank Guarantee • MAY Lose Value1 Wells Fargo Asset Management is a trade name used by the asset management businesses of Wells Fargo & Company. Certain investments are distributed by

Wells Fargo Funds Distributor, LLC, Member FINRA/SIPC, a subsidiary of Wells Fargo & Company.2 Insurance products and services are offered through non-bank affiliates of Wells Fargo & Company including Wells Fargo Insurance Inc. and Wells Fargo Insurance

Services USA, Inc.3 Wells Fargo Securities is the trade name for the capital markets and investment banking services of Wells Fargo & Company and its subsidiaries, including Wells Fargo

Securities, LLC, a member of FINRA, NYSE, NFA and SIPC, Wells Fargo Institutional Securities, LLC, a member of FINRA, NFA and SIPC and Wells Fargo Bank, N.A.

© 2014 Wells Fargo Bank, N.A. All rights reserved. Deposit products offered by Wells Fargo Bank, N.A. Member FDIC. Deposits held in non-U.S. branches are not FDIC insured. WCS-1213801 9/14

wHOLESALEBANKING

Asset Management*1

Capital Finance

Commercial & Corporate

Commercial Real Estate

Government & Institutional

Insurance*2

International

Investment Banking & Capital Markets*3

Treasury Management

WCS-1213801-WSG-UmbrellaEconSpec-210x297mm-FNL.indd 1 9/12/14 8:03 AM

On the coverSmartphones and othermobile-communicationstechnology are changing theway cars are made, boughtand used. As vehicles begin totalk to each other and the roadaround them, a driverlessfuture comes closer, page 12

The Economist Technology Quarterly September 6th 2014 Monitor 1

1

IT LOOKS like an experimental cookingclass as participants taste a green pow-

der, pull faces and then mix it into a con-coction of fruit and milk. But the eventorganised by Open Wetlab in Amsterdamhas a more serious goal than to come upwith new recipes for smoothies: findingways to make spirulina—an algae which isfull ofproteins and vitamins, but tastesdisgusting—more palatable.

Welcome to the world ofbiohacking.In its original sense, hacking involvestaking things apart and putting them backtogether again in new ways. Such tinkerershelped to create the “maker movement”,which has grown into a worldwide com-munity ofpeople constructing thingsranging from 3D-printed jewellery torobots. Biohackers have also started toorganise themselves, under the umbrellaofa movement called DIYbio.

Nearly 50 cities, mostly in America andEurope, are now home to groups ofbio-hackers or amateur laboratories wherethey can meet and experiment. BesidesOpen Wetlab, these include Biocurious inSunnyvale, California, Genspace in NewYorkand La Paillasse in Paris. The numberofbiohackers around the world is any-body’s guess, but the movement’s mainonline-mailing list boasts nearly 4,000members and is growing rapidly.

What drives the movement is the

belief that “biology is technology” (toquote the title ofa bookby Rob Carlson, aDIYbio pioneer): that DNA is a form ofsoftware that can be manipulated to de-sign biological processes and devices. Butsome people worry that amateur labora-tories could create killer bugs or providetraining for bio terrorists. For the moment,at least, such fears seem premature. Theamateur labs are not yet very sophisticat-ed, according to a recent survey of359members of the DIYbio movement by theWoodrow Wilson International Centre forScholars, a think-tank. Most activitiesinvolve extracting DNA, for instance fromstrawberries. Only13% of the biohackershave synthesised a gene and just 3% havegenetically engineered a mammalian cell.Since biohackers often have a PhD, theyprobably did this in a professional lab.

Art and scienceNot all the groups are focused on syntheticbiology. In Europe, amateur biologistsoften workwith artists and designers, saysMarkus Schmidt, co-author ofa paper onEuropean DIYbio. A recent fair called“Synthetic Aesthetics” at the Victoria andAlbert Museum in London included pro-jects that use bacteria to colour tapestries,grow bags and encode music in DNA.

“Our goal is not only to advance biolo-gy, but democratise it,” explains Ellen

Biohackers of the world, unite

The DIYbio movement: Following the example of maker communitiesworldwide, hobbyists keen on biology have started to get together

Contents

Monitor

1 Biohackers organise,medical-technology startups,heating people rather thanbuildings, the language of theinternet of things, andsuspended animation

Difference engine

6 Where gadgets go to dieA mountain of electronic wasteneeds responsible recycling

Military technology

7 The unsheltering skyLong-range ballistic-missiledefences look doomed

Demolition technology

10 Bringing the house downNew ways to demolish oldbuildings in crowded cities

The connected car

12 Smartphones on wheelsMobile communications arechanging the way cars are used

Air-traffic control

15 Free flightHow pilots can fly direct routes,saving time and fuel

Brain Scan

17 Welcome to my genomeA profile of George Church, apioneer of genetics

2 Monitor The Economist Technology Quarterly September 6th 2014

2

1

Jorgensen, president ofGenspace. Found-ed in 2010, the community laboratory inBrooklyn is the model for the two dozenothers that have since opened around theworld. Genspace hosts all sorts ofevents,including “biohacker boot camps”, as wellas projects such as “barcoding” in Alaska,an attempt to catalogue plants.

Technological movements often arisewhen a critical mass ofenthusiasts getgreater access to information and findtools that are both cheap and widelyavailable. A similar thing is happening inmedical technology, where there is aflourish of innovative startups makingnew devices (see next story).

Many of the first makers were softwaredevelopers wanting to reconnect with thereal world by building physical things.Enough people are now interested inbiology and have knowledge to hackDNA,says Mr Carlson. DIYbio also has roots iniGEM, a successful annual synthetic-biology competition for undergraduates.

As in the case of the maker movement,websites such as YouTube and Instruc-tables allow tinkerers to share ideas. Simi-larly, access to information about biotech-nology has become much easier, says MrCarlson. The Journal of Visualised Experi-ments, a peer-reviewed online-videopublication, is one source.

The necessary laboratory equipment isno longer beyond the budget ofhobbyists.Many devices are now for sale on eBayand more specialised online market-places, not least because the recession hasforced a number ofcommercial laborato-ries to close. Such equipment can also bebuilt more cheaply by using off-the-shelfparts and open-source software.

My own PCR machineIf 3D printers are the tool ofchoice formakers, PCR machines are de rigueur inamateur labs. Using a biochemical tech-nology called polymerase chain reaction(hence PCR), the machines are used toidentify a specific segment ofDNA andmake multiple copies of it. “You can nowbuild these in a garage,” says Josh Perfetto,who is one of the founders ofOpenPCR, agroup which has developed a simple PCR

machine that costs only $600.DIYbio also benefits from the organisa-

tional infrastructure of the maker move-ment. Many laboratories start in hacker-spaces, which serve as clubhouses formakers. Amsterdam’s Open Wetlab, forinstance, is part of the Waag Society, anorganisation which also runs a shop formakers. Moreover, many tinkerers havestarted dabbling in biology.

All this raises the question ofhow bigDIYbio will become. The maker move-ment now counts tens of thousands ofmembers and hundreds ofstartups. Itsboosters say DIYbio could repeat the trick.It has already spawned its first firms.

OpenPCR has now become a startupbusiness and is working on an improvedPCR machine. Amplino, a Dutch startup,has built a low-cost device capable ofdetecting malaria, which it hopes will beused in developing countries.

Other firms are working on productsthat could make life much easier for bio-hackers. Autodesk, a big software com-pany, has a scheme under way code-named “Project Cyborg” which is devel-oping design tools for DNA. OpusLabworks, a startup, aims to build a fluid-handling robot starting at $2,000. Evenmore ambitious, Cambrian Genomics isbuilding a “3D printer for living things”—adevice that can cheaply synthesise DNA.

But there are barriers that will limitDIYbio’s growth. Building a biologicaldevice is a lot more complicated thanputting together a robot or designing anew circuit board. And whereas regulatorshave largely ignored the maker move-ment, they are a lot more interested in theworkofamateur laboratories.

IfEuropean biohackers are less focusedon synthetic biology, it is partly becausethey need to askfor permission. Geneticslaboratories require a licence and only afew are even trying to get one. In Americabiohackers used to riskgetting arrested,but in recent years the FBI has opted for amore enlightened approach: local specialagents talk to community labs; the agencyorganises an annual DIYbio conference; itis even a sponsor of iGEM. “The peoplewho practise DIYbio are best placed toknow what is going on,” says Edward You,who pioneered the FBI’s effort. He alsothinks that the agency and the DIYbiomovement have a “shared responsibilityto protect science”. In other words, ifthings go wrong there will be tighter regu-lations—making life more difficult for bothlaw enforcement and biohackers.

Most DIYbio leaders welcome all this(although some joke that DIYbio wouldnot be where it is today without the FBI’ssupport). “What you don’t want to do issurprise law enforcement and regulatorswith new technologies,” says Jason Bobe,co-founder ofDIYbio.org, a charity thatsupports the movement, who also workswith George Church, a pioneer ofgenetics(see page 17). For now it seems to be able toregulate itself. DIYbio.org has hired safetyexperts for members to consult. Andleaders of the movement on both sides ofthe Atlantic have developed a code ofethics which frowns on releasing geneti-cally modified organisms into the environ-ment. When one group wanted to launcha crowdfunding project to develop a glow-ing plant and send contributors the seeds,their laboratory showed them the door.(The project nonetheless went ahead andwill be the first biology startup to join YCombinator, Silicon Valley’s leading “ac-celerator”, which provides capital and

advice to new ventures.)As DIYbio grows calls for tighter regu-

lations will get louder��et clamping downwould be counter-productive, argues MrCarlson. Such rules would be hard toenforce and drive biohackers under-ground. It would hamper startups andlimit innovation. Much better, says MrCarlson, for governments to support com-munity labs where everbody—biohackers,startups and anyone who is interested—can experiment openly and safely.7

SOFTWARE helped startups flourish,with open-source programs and the

web providing easy collaboration and afast route to market. Tens of thousands ofsoftware developers will produce appsworth more than $30 billion this year.Now cheap computing power and theshrinking cost ofsmall-scale manufactur-ing are spawning more hardware startups.As examples from Europe show, a particu-larly thriving area is medical technology.

“Easy access to technology empowerscreative minds,” says Ulrich Weinberg ofthe Hasso Plattner Institute in Potsdam,Germany. Although Mr Weinberg oftenworks with big companies that makemedical devices, he has seen the costs ofentry tumble. “Even groups ofstudents

Doctoringdevices

Medical technology: Easy access totechnology and a lower cost of entryare creating a crowd of startupsmaking new gear

Babybe gets a cuddle

The Economist Technology Quarterly September 6th 2014 Monitor 3

2

1

with very limited financial resources areable to build functional prototypes withina couple ofweeks,” he says.

Turning ideas into products has neverbeen easier. It begins with cheap—and insome cases free—computer-aided-designsoftware, which allows devices to bemodelled and simulated to a high level ofaccuracy. Rapid-prototyping equipment,such as 3D printers, can quickly give theideas shape. And besides open-sourcesoftware, product developers now havelow-cost open-source hardware they canuse too, such as Arduino microcontrollers.

Medtech, as the business is known, isparticularly attractive to inventors be-cause the market is eager for new devicesthat can improve people’s lives, help withtreatments and make the jobs ofhealth-care workers easier. The field is also bigand growing: medical-technology salesare expected to reach $228 billion by 2015,up from $164 billion in 2010, according toMarkets and Markets, an American con-sulting firm.

The startups come from myriad areas.Babybe, for instance, was founded byRaphael Lang, an engineer and artist fromBarcelona, and Camilo Anabálon, anindustrial designer who lives in Santiago,Chile. The pair met while studying at theStuttgart State Academy ofArt and Design.After working on projects for a number ofcompanies, they got the idea for Babybeafter visiting a neonatal intensive-careward and being struckhow babies inincubators were surrounded by lots ofhigh-tech gear which, although helpingthem, separated the babies from the armsof their anxious mothers.

Copying motherThat led them to develop a system whichreplicates the heart beat and breathingpattern ofa mother for her baby while it isin an incubator. The idea is to maintain alinkbetween the two, which could helpreduce infant stress. A pad called the “tur-tle” is held by the mother to her chest. Theturtle contains various sensors which readher vital signs and it transmits those datato a soft, flexible mattress that is filled withgel. The mattress uses pumps and otherelectronic systems to create a gentle rhyth-mic movement which copies that of themother’s chest.

Developing the bionic mattress costless than $150,000. And being a two-manoperation based in different countries wasnot a handicap. “You can combine the besttechnology from all over the world withthe help of freelancers,” says Mr Lang.Hence the gel used in the mattress comesfrom Germany, miniature pumps fromAmerica and most of the digital innardsand plastic parts from China. Big compa-nies are far too bureaucratic to be able tooperate with the same speed and at suchlow cost, Mr Lang believes.

Many medtech startups are academicspin-offs. Scopsis, for example, is a collabo-ration begun in 2010 with Charité Univer-sity Hospital in Berlin and the FraunhoferInstitute. The company makes navigationsystems for minimally invasive surgery.These use various data, including imagesfrom the endoscopes used in such proce-dures, to ensure surgeons are followingthe correct path. One system allows thevirtual plans a surgeon has made on acomputer screen to be overlaid on thevideo from an endoscopic camera duringthe operation.

Ideas for new devices often emergewhen doctors team up with engineers.This was the case with TranscatheterTechnologies, founded in Regensburg,Germany, by Wolfgang Goetz, a cardiacsurgeon, and Hou-Sen Lim, a mechanicalengineer and entrepreneur. They devel-oped a new method of implanting artifi-cial aortic valves into a patient’s heartwithout open surgery. The artificial valve,which has a special sealing cuff, is placedinto position with a catheter. Importantly,it can also be repositioned, which greatlyreduces the riskofcomplications.

Some startups focus on the consumermarket, often with devices that connect toa smartphone app. One example is Ko-libree, a Paris-based firm which haslaunched an “intelligent” electric tooth-brush. It gathers data on a person’s brush-ing habits and informs him through theapp how well he is doing and which teethhe should brush a bit longer. Like manystartups, Kolibree used a crowdfundingplatform, in its case Kickstarter, to raisemoney. Delivery of its smart toothbrushesis planned for October.

Despite benefiting from the lower costofentry, medtech minnows may not bemuch ofa threat to the industry’s giants.Innovation has long bubbled up frombelow, though not necessarily in the formof independent startups. In the past, doc-tors were more likely to offer ideas to bigfirms to commercialise. Some would helpwith the product development.

Even though it is easier to get a business going, most startups tend not to remain on their own. Many large medical-equipment firms now acquire new products by buy-ing small companies that invent them—an exit strategy factored into the business plans ofmost startups. Selling out also helps when costly clinical trials and other regulations have to be met. And marketing a new product all over the world can be hard for a small company without a well-known brand.

But there is a third way, one that Baby-be intends to try. The company’s foundersdo not want to sell their firm but hope toget support from an established companyin return for a share of the profit. It is apartnership that might prove a good dealfor both sides.7

BUILDINGS are horribly inefficientconsumers ofheat. In winter, a vast

amount ofenergy is wasted heating emp-ty homes during the day, and warmingempty commercial buildings at night.Even when buildings are in use, unoccu-pied spaces are routinely kept at the sametemperature as those that are occupied.Some spaces in particular are spectacularheat wasters. The huge atriums that formany firms serve as corporate statussymbols are usually occupied by no morethan a handful ofpeople, yet every cubicfoot is kept warm around the clock. Smallwonder that the heating, ventilation andair conditioning (HVAC) ofbuildingsaccounts for13% of total energy consump-tion in America.

A 2012 paper published in the journalEnergy and Buildings unsurprisingly foundthat the operation ofHVAC systems in twobuildings at the Massachusetts Institute ofTechnology (MIT) closely tracked factorssuch as outside temperature. But it alsorevealed a distinct lackofcorrelationbetween building occupancy and theamount ofenergy supplied by its HVAC.This makes little sense, because apart fromthe modest ambient heat required to stopwater pipes from freezing, it is peoplerather than buildings who care aboutbeing comfortably warm.

Carlo Ratti, director ofMIT’s Senseable

In the momentof the heatLocal heating: One way to keep warmis to heat people rather thanexpending energy heating thebuildings they are in

4 Monitor The Economist Technology Quarterly September 6th 2014

2 City Laboratory and a co-author of thepaper, was musing on this while sittingoutside a restaurant being warmed by oneof those tall, mushroom-shaped infra-redheaters. Might it be possible, he wondered,to aim such heat more accurately at peopleas they move around? Thus was born hislab’s “Local Warming” project, a prototypeofwhich is on display at the 14th Interna-tional Architecture Exhibition being heldin Venice until November 23rd.

The installation in Venice, along withone which Dr Ratti and his team set up inthe exterior “porch” beneath MIT’s statelymain entrance, uses a combination ofpowerful infra-red lamps, clever opticsand servo motors to direct beams ofheatat people. The location and trajectory ofthe people are tracked by a Wi-Fi-basedsystem that was also developed by theuniversity. Another more compact andcost-effective prototype replaces the mov-ing lamps with arrays of infra-red LEDsand “targeting” optics that simply switchon and offas people come and go. The LED

arrays have no moving parts and could,for example, easily replace false ceilings inoffice buildings.

Warm encountersLocal Warming can workwith a smart-phone app that enables each person to settheir own preferred temperature. Eachindividual is warmed by a number ofLEDs to avoid the feeling—all too commonin the vicinity ofmushroom lamps—thatonly part of their body is being heated.“It’s almost like having a personal sunfollowing you around,” says Dr Ratti,although ifyou inadvertently leave yourphone on your deskyou get to shiver atwhatever baseline temperature the build-ing’s general HVAC system is maintaining.And if two or more people are chatting atthe water cooler, they may find their tem-perature preferences momentarily mixed.

Local Warming’s efficiency and costsavings are hard to calculate. If there areenough people in a locally warmed spacethe system will reach saturation, and itsefficiency will be no greater than a con-ventional HVAC installation. But in a large,intermittently populated space such as anatrium or lecture hall, Dr Ratti believes theenergy used for heating could in theory becut by up to 90%. As his team develops thetechnology and installs it more widely, itwill glean more accurate data on savings.

Whether such “active” heating systemsgain widespread adoption will depend onthose numbers, and also on how costlythey would be ifmass-produced. But LocalWarming is clearly sparking heated in-terest. The American government’s Ad-vanced Research Projects Agency-Energy,which backs research and development ofenergy technologies, has already set aside$30m to help get similarly heart-warmingprojects offthe ground.7

THERE was a time, not long ago, whenaccess to the internet could be gained

only through a computer. Now people canget to it using phones, tablets and somegames consoles. Increasingly, other de-vices are becoming internet-linked too, asconnectivity is extended to everydayobjects such as televisions, radios, watch-es and cars.

The “internet of things” promises atechnological revolution, but for it to workwell these things need to speak the samelanguage. Industry, however, tends toadopt common standards—ifat all—onlyafter jostling between rival producers withcompeting systems. It was so for trains,televisions, video recorders, mobilephones and the internet itself. And it willbe the same for connected devices.

A number of industry groups are tryingto standardise how things connect andcommunicate. One of the most prominentis the AllSeen Alliance, a consortium offirms from such diverse worlds as semi-conductors, white goods, consumer elec-tronics and retail. Its biggest membersinclude Haier, LG, Panasonic, Qualcommand Microsoft. Their proposed solution isAllJoyn, a free piece ofsoftware created byQualcomm but handed to the Alliance forits members to develop further.

AllJoyn is designed to sit inside devicesand, regardless ofmanufacturer or operat-ing system, mediate direct communica-tions over wireless links such as Wi-Fi orBluetooth. The idea is that objects broad-cast to each other descriptions ofwhatthey can do, in a common language.

A television might announce that it candisplay notifications, a speaker that itplays audio files or a clock that it can exe-cute commands at given times. A newlyreleased smartwatch, for example, shouldthen have no trouble communicating withthese other devices even though they havenot been previously acquainted or pro-grammed to workwith each other. In thecase ofan air-conditioning unit, say, thesmartwatch could control the airflow andthe air conditioner could send tempera-ture information to the watch’s screen.

Once more things connect, manyscenarios present themselves. Connectedsmoke alarms could enlist nearby lightbulbs to flash and speakers to sound analert. A warning about the smoke’s loca-tion could appear on a television. Anddoor locks could be automatically opened.

Many other not-yet-thought-ofapplica-tions will arise, says Liat Ben-Zur, chair-woman of the AllSeen Alliance. “Unex-pected capabilities pop up when devicesspeak the same language,” she says.

Simple as it sounds, the challengeremains considerable. The immediateproblem is accommodating the greatdiversity ofconnected things, varying notjust in their thousands ofdifferent uses,but also in their sophistication. Estab-lishing standards flexible enough to em-brace all devices is a daunting task.

So is trying to ensure forward compati-bility. Innovation in connected devices israpid, but futureproofing is difficult be-cause no one knows where the tech-nology and its myriad applications willlead. People also have to get used to smartdevices and installing regular softwareupdates even, perhaps, for ovens.

The real problem may turn out to benot a lackofstandards, but too many—anddisagreement over which initiative topursue. The AllSeen Alliance is just one ofmany groups working on solutions. AT&T,Cisco, GE, IBM and Intel have helped formanother group called the Industrial In-

The language of the internetof things

Communication standards: More and more devices are becoming connected,but will they speak the same language?

1

The Economist Technology Quarterly September 6th 2014 Monitor 5

2

THE picture that is immediately broughtto mind by talkofsuspended anima-

tion is usually one of the crew ofNostromowaking up from their deep sleep after along journey through space in the 1979movie “Alien”. Yet this gives a somewhatmisleading impression ofSamuel Tisher-man’s new trauma technique, which usesan innovative rapid-cooling procedure tosuspend life. Dr Tisherman and his col-leagues prefer to call it emergency preser-vation and resuscitation (EPR). The firsttrials using EPR on people are now beingheld at the University ofPittsburgh Medi-cal Centre Presbyterian Hospital. Earlynext year more trials are planned at theShockTrauma Centre, part of the Universi-ty ofMaryland Medical System.

Dr Tisherman’s EPR process, developedwith the help of$800,000 from the De-partment ofDefence, is mostly aboutresurrection. The idea at this stage is to useequipment like the catheters and pumpsthat can be found in any trauma centre tosuspend the life ofcritically injured peo-ple in order to buy more time for surgeonsto try to save them.

EPR works by lowering the patient’sbody temperature and replacing theirblood with a cold saline solution. Hypo-thermia is already induced in patients tohelp reduce bleeding during some surgicalprocedures. But cooling the body down sothat it goes into a suspended state has notbeen tried before. The idea came fromobservations that people have been resus-

citated having stopped breathing for halfan hour or more after falling into icy water.

Research on EPR was begun in Pitts-burgh by the late Peter Safar and Dr Tisher-man. Dr Safar, who died in 2003, was apioneer ofcardiopulmonary resuscita-tion, known as CPR. Dr Tisherman carriedon with the workand, after successfulexperiments on animals, got approval forhuman trials. The Food and Drug Admin-istration decided the procedure was ex-empt from informed consent, as patientswould be too ill to give it themselves and

might benefit because they were likely todie as no other treatment was available.For now, the patient has to be between 18and 65 years old, have a penetratingwound, such as a knife, gunshot or similarinjury, suffer a cardiac arrest within fiveminutes ofarrival in the hospital and failto respond to usual resuscitation efforts.

It is one thing to carry out EPR in anordered laboratory but quite another to doit in a busy emergency centre. The firstchallenge is for medical staffto insert acatheter into arteries to flush all the bloodout of the patient. The blood is replaced bygiving the patient 2-3 litres a minute ofsaline solution at a chilly10°C. The proce-dure has to be completed within 20-30minutes to have a chance ofworking.

Once the patient is in a suspendedstate, a surgeon will try to repair thewound within an hour to prevent anybleeding when blood circulates again.Finally, a heart-lung bypass machine isused to restart the blood flow and warmthe patient up. The medical team will tryto keep the patient slightly cooler thannormal, at around 34°C, for12 hours. Thenan attempt will be made to bring thepatient round. In animal experiments theheart did not always start by itself.

If the human trials are successful, theaim is to take the technique out into thefield. For example, a paramedic might beable to use EPR to put a dangerously illpatient into a suspended state until theycan be rushed to a specialist hospital. Sucha system could also be used on the battle-field to evacuate injured soldiers. PortableEPR systems will depend on the devel-opment ofnew technology to miniaturiseand automate equipment. That might bepossible with, for instance, smart cathetersthat use ultrasound to help non-specialistsguide them correctly into blood vessels. Itis another example ofscience fictionbecoming science fact.7

The big sleep

Suspended animation: Doctors have begun human trials of suspendedanimation to buy more time for critically injured patients

Wakey-wakey, there’s an alien about

ternet Consortium to “define commonarchitectures” for smart objects. Otherorganisations include the IPSO Alliance,which is promoting the use of the InternetProtocol, which eventually standardisedthe way that data could flow across differ-ent types ofnetwork; the Open Intercon-nect Consortium (Broadcom, Dell, Sam-sung and others); and the Institute ofElectrical and Electronics Engineers,which is working—perhaps ambitious-ly—to build consensus on standards.

The many approaches reflect the youthof the technology and it is unlikely thatany of the current efforts will succeedcompletely, says���elosa, a researchdirector at Gartner, a market-research firm.They will, though, nudge the industryforward by getting companies to talk toone another. It could take years beforefirms commit themselves to one solution.

Some companies are involved with morethan one group: Cisco, for example, is apartner ofboth the AllSeen Alliance andthe Industrial Internet Consortium, and amember ofseveral other initiatives.

The various efforts are both competingand complimentary, adds Mr�

elosa.AllJoyn, for example, is designed primari-ly for short-distance object-to-object com-munication within the home; alternativesolutions may involve a gateway, such as arouter, to intermediate communicationbetween devices—an approach bettersuited for outdoors, large buildings andcitywide networks. The Industrial In-ternet Consortium has a greater commer-cial and industrial focus. Given the hugevariety ofconnected devices, and nodoubt the desire ofsome firms to keepthings proprietary, it is possible that nosingle standard ever emerges.7

6 Difference engine The Economist Technology Quarterly September 6th 2014

WHAT to do with old computers,monitors, keyboards, printers,

phones and other digital paraphernalia?On no account should anything contain-ing a printed circuit board be put in therubbish bin for municipal collection. Notcountingall the other toxicmaterials usedin electronic products, the lead in the sol-dered joints alone requires such items tobe recycled professionally.

According to a United Nations initia-tive known as StEP (Solving the E-WasteProblem), electronic waste can contain upto 60 elements from the periodic table, aswell as flame retardants and other nastychemicals. Apart from heavy metals suchas lead and mercury, there are quantitiesofarsenic, beryllium, cadmium and poly-vinyl chloride to be found. All of these pose hazards to the healthof those handling them.

When burned at low temperature, the brominated flame retar-dants used in circuit boards and casings create additional toxins,includinghalogenated dioxinsand furans—some ofthe most toxicsubstances known. These can cause cancer, reproductive disor-ders, endocrine disruption and numerous other health problems.Meanwhile, the heavy metals released by incineration can accu-mulate in the food chain (especially in fish) and come back tohaunt future generations.

The trouble is that, even with respectable collection centres,there is no guarantee that e-waste will be processed responsiblydownstream. What little is known about recycling hazardouswaste in America, for instance, suggests thatonly15-20% isactuallyrecycled; the rest gets incinerated or buried in landfills, accordingto the Environmental Protection Agency (EPA). There is no evi-dence to suggest other countries are any better.

With few audits undertaken, even the EPA has to rely on as-sumptions and guesswork. Most observers agree that only 20% orso of the 9m tonnes of e-waste collected each year in America isprocessed domestically—either by reputable firms under con-trolled conditions, or by prison inmates with few, if any, handlingrequirements. In other words, the bulkof the waste—up to 80% byweight—gets exported to places in Asia and Africa where healthand safety regulations are less onerous.

Such exports are banned in Europe, but remain legal in Ameri-ca. The United States is the only developed country that has re-fused to ratify the 1989 Basel Convention, an international treatycontrolling the export ofhazardous waste from wealthy countriesto poorer ones. America has also refused, along with Canada andJapan, to accept the Basel Convention’s 1995 amendment that im-poses an outright ban on such trade.

Not that the Europeans behave any more ethically. Inspectionsof 18 seaports in the continent in 2005 found nearly half the e-waste destined for export was actually illegal. Shippers use va-rious dodges to circumvent the Basel ban. For instance, waste la-belled as goods for refurbishing or reuse can pass muster. It wouldbe nice to thinkthat scrapped electronic products are repaired andput back into productive use, but that is often not the case.

The Chinese city of Guiyu in Guang-dong province is the e-waste capital of theworld. Though container loads are stillshipped there from American, Europeanand Japanese ports, the bulk of the e-waste being processed in China nowa-days is domestically produced. Guiyu isreckoned to employ 150,000 people, in-cluding large numbers of children, disas-sembling old computers, phones and oth-er electronic devices by hand. Circuitboards are soaked in acid to dissolve outthe lead, cadmium and other metals. Plas-tic cases are ground into pellets, and cop-perwiring is stripped of its plastic coating.With costs so low, there is a ready marketfor most of the materials recovered.

Yet there is a price to pay for all this ac-tivity. Air pollution and contamination of the water supply inGuiyu are said to be horrendous. A medical researcher from near-by Shantou University found concentrations of lead in the bloodof local children to be on average 49% over the maximum safe lev-el. The highest concentrations were found in children living inhomes that contained workshops for recycling circuit boards onthe premises.

India is another big processor ofe-waste. All told, some 25,000workers in Delhi alone are estimated to be employed recycling upto 20,000 tonnes annually of computers, phones and other hard-ware. The preferred method for recycling circuit boards in India isto toss them into an open fire—to melt the plastics and burn awayeverything but the gold and copper.

With the mountain of e-waste growing at 8% a year, the20m-50m tonnes the EPA reckoned wasproduced globally in 2009could easily reach 100m tonnes by 2020. What can be done to re-duce the impact? Probably, not much at present. Recycling in anenvironmentally sound manner is expensive. For wealthier coun-tries it remains much cheaper to ship unwanted electronic goodsto poorer parts of the planet.

Dismantling at homeThe cost of recycling e-waste in America would, of course, comedown significantly iffirmsdoingthe workhad a greater volume ofelectronic trash to deal with. Thatwould also spur innovative newmethods of processing the material. But such a change would re-quire stiff penalties to be imposed on the export of e-waste, or atleast gettingmanufacturers to include a fee in the price ofelectron-ic goods to offset the cost of taking them backfor reprocessing.

In the meantime, people can do theirown dirty workby takingthe old television set, obsolete computerorbroken refrigerator to arecyclerwho is an accredited memberofone ofthe two voluntarycertification schemes: E-Stewards and Responsible Recycling Prac-tices. An interactive map giving details of certified recyclers is onthe EPA’s website (www.epa.gov/epawaste). In Europe the num-berofrecyclers accredited by E-Stewards is increasing steadily. TheBasel Action Network, an environmental pressure group, also listsrecyclers (www.ban.org). Owning an electronic device nowcomes with a responsibility for its afterlife. 7

Where gadgets go to die

Recycling electronics: A growing mountain of electronic waste needs to be disposed of responsibly by rich nationsrather than shipped to poorer countries to do the dirty work

The Economist Technology Quarterly September 6th 2014 Military technology 7

AS TEST flights go, FTG-06b was a daz-zling affair. The mission was part of a

programme called Ground-based Mid-course Defence (GMD), which is supposedto provide America’s main shield againstintercontinental ballistic missiles (ICBMs)with a range beyond 5,500km (3,418 miles).FTG-06b involved the launch (pictured op-posite) on June 22nd fr���

andenberg AirForce Base in California ofa hypersonic in-terceptor. It successfully annihilated an un-armed warhead which had been fired intospace from a US Army site on KwajaleinAtoll in the western Pacific Ocean.

The warhead was tracked bytwo Amer-ican naval vessels: a destroyer equippedwith an Aegis anti-missile system and a$900m floating offshore oil-rig, which hadbeen kitted out with a highly sophisticatedactive phased-array X-band radar. Farmore powerful than conventional radar,the X-band system can calculate—with thehelp of some big computers in ColoradoSprings—the size, shape and trajectory of abaseball-sized object 4,000km away trav-elling at 32,000kph.

Twelve years ago the United Stateswithdrew from the Anti-Ballistic MissileTreaty, a 1972 deal that limited the testingand deployment ofanti-ICBM weapons byAmerica, the former Soviet Union and, lat-er, Russia and some ex-Soviet republics.Since then, most technological advances insuch systemshave been in America, wherethe Missile Defence Agency (MDA) hasspent some $98 billion on various projectssince 2002. Although China appears to beworking on an anti-ICBM system, Russia isthe only other country with such a pro-gramme—and it is far less capable, says Jef-frey Caton, a former US Air Force coloneland space-warfare specialist.

Meanwhile, the threat grows as poten-tial attackers continue to acquire “morecomplex, survivable, reliable and accu-rate” ICBMs equipped with countermea-sures��ice-Admiral James Syring, theMDA’s boss, told Senate lawmakers inJune. Next year Iran could have a ballisticmissile able to reach America, he added.Butothers thinkthat isat least several years

away. North Korea is also testing rocketsand satellite systems which could carry anuclear warhead. Arun Prakash, a formerchairman of India’s Chiefs of Staff Com-mittee, sees the one-upmanship betweenoffence and defence systems as “a ding-dong battle” with the defender at a perpet-ual disadvantage because it is far easier tobuild a missile than shoot it down.

Despite the success of FTG-06b theprospects for a truly effective defenceagainst ICBMs appear as far away as ever.GMD alone has already cost America morethan $40 billion. Yet until June it had failedall five intercept tests conducted since2008, even though each was meticulously“scripted for success”, in the words of Phil-ip Coyle, a formerWhite House science ad-viser to BarackObama.

When things go wrongThe GMD system consistsofan “exoatmos-pheric kill vehicle” with steering rocketsand its own X-band radar system. The killvehicle is made by Raytheon, a big Ameri-can defence contractor. Other companiesinvolved in the project include Boeing, Or-bital Sciences and Northrop Grumman.The kill vehicle was used in two of thefailed tests. On two other occasions, notcounted as “tests”, a GMD interceptorfailed to leave its silo.

With such a record, FTG-06b was some-thing of “a make or break for the pro-gramme”, says Riki Ellison, chairman ofthe Missile Defence Advocacy Alliance, alobby group based in Washington, DC.When he addressed the Senate Subcom-mittee on Strategic Forces in April��ice-Ad-miral Syring admitted as much when hesaid that a failure ofFTG-06b would force areassessment of plans that are under wayto expand the programme.

So far, there are 30 interceptors at�

an-denberg Air Force Base and Fort Greely inAlaska. The MDA has begun work at FortGreely to prepare for a field of silos thatwill contain an extra 14 interceptors by2017. Even though the June test went well,GMD remains so unreliable that the expan-sion plans should be scrapped, says Fred-

The unsheltering sky

Missile defences: Even with new technology, America’s multi-billion-dollarefforts to build a shield against long-range ballistic missiles looks doomed

1

8 Military technology The Economist Technology Quarterly September 6th 2014

2

1

erickLamb of the University ofIllinois anda consultant to the Pentagon. In combatconditions seven or so GMD interceptorswould probably be needed to smash evena single rudimentary North Korean ballis-tic missile, reckons Mr Coyle, now with theCentre for Arms Control and Non-Prolifer-ation, also a Washington lobby group.

Money is being poured into developingnew radar systems that could improve theaccuracy of anti-missile technology. Butsalvaging GMD, some experts believe,might require an entirely new and largerkill vehicle. The MDA would like one, butthe project would take years. It took fouryears (and $1 billion) just to tweak the vi-bration frequency of the current vehicle’sfour thruster rockets because they were in-terfering with its inertial measurementunit, says George Lewis, a researcher atCornell University.

There are other missile defences. So far,30 of America’s warships carry Aegis anti-missile systems, but these were designedto strike shorter-range missiles. With re-cent upgrades, Aegis is thought to be capa-ble of intercepting warheads in space, inlimited circumstances. With additional ra-dar near America’s east coast, Aegis de-stroyers in the Atlantic could theoreticallyintercept ICBMs coming from Europe andAsia, says Henry Cooper, who was Presi-dent Reagan’s missile-defence negotiator.Japan has purchased the necessary kit forits warships and a land variant, AegisAshore, is due to be sited next year in Ro-mania and, in 2018, in Poland.

Shielding America from ICBMs will re-main impossible for the foreseeable future,reckons William Cohen, a former Ameri-can secretary of defence. A missile assaultfrom China or Russia would overwhelmeven flawlessly performing US defences.And defending against a limited attackfrom a sophisticated opponent would, headds, suffer from unresolved problems.

Among those problems are decoys.After leaving the atmosphere a big ballisticmissile can release, along with ten or sowarheads, dozens of decoys. In the vacu-um of space the decoys will travel at thesame speed as a warhead. Decoys can begenerated by discharging infra-red-emit-ting aerosols or clouds of thin wires or tin-foil strips known as chaff. A defender’s ra-dar will register many incoming objectsbut only a fraction contain a warhead, saysTheodore Postol, a missile expert at theMassachusetts Institute of Technology.Even ifdecoys can be identified, each radarblip may require several seconds or moreofanalysis. But time is short. With a closing

speed of more than 10km a second, an in-terceptor must typically commit to attack-ing a single object at least 50 seconds be-fore hitting it, says Dr Postol.

Among the most dangerous decoys areshiny Mylar balloons, similar to those soldby party shops, says Thomas Reed, a for-mer secretary of the US Air Force. Madefrom plastic with a metallic coating, thematerial reflects radar. Dozens can be re-leased in one go and inflated to look on ra-dar just like cone-shaped warheads, addsMr Reed. Worryingly, a warhead could beconcealed in a Mylar balloon.

It is possible that nuke-carrying bal-loons can be detected by heat sensors be-cause they would be warmer as a result ofthe slowly decaying plutonium inside thewarhead. But it would not be difficult tofoil such sensors on interceptors (or satel-lites) by fitting each decoy balloon with asmall battery-powered heater.

Multiplying the problemDecoys can also be generated by explosive“cutting cord” on the inner wall of the finalbooster stage of the warhead. Upon sepa-ration in space, the explosive breaks up themetal casing of the booster. “Now you’vegot 20 objects coming towards you” sogood luck identifying the warhead, saysCornell University’s Dr Lewis.

America’s National Intelligence Coun-cil said in 1999 that China and Russia haddevised numerous countermeasures toprotect offensive missiles and were proba-bly willing to sell the technology. A state-ment in May by the office of the assistant

secretary of defence for research and engi-neering noted that the proliferation ofsuch advanced countermeasures was ren-dering America’s missile defences “no lon-ger practical or cost-effective”.

Nevertheless, many proponents ofmis-sile defence believe more research couldmake even the most sophisticated decoysrecognisable. Decoyswere used in the JuneFTG-06b test, but the GMD engineers knewwhat to expect. Multiple interceptors couldbe launched, one after the other, for eachwarhead thought to be on its way. As thefirst interceptor draws closer to a flock ofdecoys, it could relay increasingly accuratedata to a following interceptor to hit a war-head that has been identified as real.

For now, though, no country has comeclose to defeating decoys, says KingstonReif, also of the Centre for Arms Controland Non-Proliferation. This view is widelyshared. Even if the hurdles are overcome,others would arise. Warheads in spacecould fire steel balls out in front of them toclear the way of interceptors, says MrCoyle. An interceptor’s radar might bejammed by electronic-warfare measuresor a nuclear warhead could be pro-grammed to detonate upon detection ofan approaching interceptor. A detonationin space would generate a powerful elec-tromagnetic pulse (EMP) which couldknock out electrical circuits and powergrids across a continent. America’s EMP

Commission, a body assembled by Con-gress to study such a threat, reckoned in2008 that two-thirds of Americans mightperish in the first year ofa societal collapse

It is far easier to build a missile than shoot it down

Stuck in a silo

The Economist Technology Quarterly September 6th 2014 Military technology 9

2 that would follow a nuclear blast in spaceabove the central United States.

Among nuclear powers, neither NorthKorea nor Pakistan is presently capable ofbuilding a ballistic-missile triggering sys-tem that is able to detonate a nuclear pay-load if an interceptor was drawing near,reckons Mr Reed, the former US Air Forcesecretary who has also designed nuclearwarheads for the Pentagon. With time andenough effort, this could change. At leastone type of nuclear device detonated byNorth Korea “is not inconsistent” with ef-forts to build a bomb designed for an EMP

attack, says James Woolsey, a former direc-tor of America’s Central Intelligence Agen-cy. (What is needed is not necessarily alarge blast, but lots ofgamma rays.)

Such an attack might not even require aballistic missile. In December 2012 NorthKorea launched a satellite on a southerlytrack. Although it may have malfunc-tioned, the launch reveals another vulner-ability in missile defences which could beexploited for an EMP attack, reckons MrWoolsey. If a nuclear device was fitted intoa subsequent southerly launched satellite,it would circumvent America’s defencesagainst long-range weapons because theseare positioned to hit warheads flying fromover the North Pole, not those coming fromthe south. Moreover, a nuke concealed in asatellite in an orbit used by many civiliansatellites could be detonated on a flyoverabove America. There is no point in havinga missile-defence system that cannot pre-vent such an attack, says Mr Woolsey.

It might, however, be possible to shootdown missiles or rockets before they reachspace and eject decoys or place a nuke-car-rying satellite in orbit. Proponents of“boost phase” defence, as it is called, pointout that during its ascent a missile is easierto hitbecause it travels slowlyand presentsa large, easier-to-pinpoint target thanks toun-jettisoned fuel tanks and the heat fromits exhaust plume. Another plus is that if itis hit by an interceptor soon after launch,

the missile’s payload and debris may fallbackon the country that launched it.

The tricky bit is placing interceptorsclose enough to reach the missile before itleaves the atmosphere. Ronald Reaganhoped to put them into low orbit, but the“Star Wars” scheme, as it was known,would have required legions of satellitescosting many billions of dollars. Anotherproblem with the Strategic Defence Initia-tive, to use its formal name, is that satellitescan be shotup orblinded with Earth-basedlasers. There is also a danger that the lasersmight “fall into radical hands”, says a mili-tary adviser to a European head of state.The adviser, who insisted on anonymity,added that there was concern about debrisfrom destroyed anti-missile satellitesknocking out other satellites. In a 2007 testChina shot up one of its defunct weathersatellites, creating a huge increase in thespace debris threatening satellites today.

The notion of arming satellites forboost-phase defence now has “zero main-stream adherents”, says Brian Weeden, aformer ICBM launch officerwho spent fouryears, as he puts it, “on alert in Montanawaiting for the end of the world”.

There is another technology taking tothe sky in increasing numbers that couldplay a role: using drones to launch inter-ceptors. Dale Tietz, a former senior StarWars official, says that North Korean mis-siles could be prevented from reachingspace by just three interceptor-armed Glo-bal Hawk drones flying above internation-al waters near the hermit kingdom.

David Trachtenberg, a deputy assistantsecretaryofdefence formissile defence un-der George W. Bush, believes that Americashould spend more on developing inter-ceptor-armed drones. But flying dronesclose enough to launch sites without pene-trating enemy airspace could be difficult.Iran is probably too big for drones to patrolsuccessfully because launch sites could belocated deep inside the country. Even inplaces that could be patrolled, drones

would need fast reactions. Last year Amer-ica’s National Air and Space IntelligenceCentre reported that the North Korean re-gime was developing a solid-fuel missile.Replacing its present liquid propellantswith solid-fuel would greatly reduce NorthKorea’s launch preparation time as well asthe time—roughly five minutes—which itsmissiles take to reach space.

Might aircraft-mounted anti-missile la-sers work? A few years back the Pentagoncancelled a Boeing-led airborne-laser pro-gramme, in part because the modified 747

airliner’s bulky chemically generated laserhad a limited range. Solid-state lasers mayperform better. The MDA believes thatdrones carrying lasers will “play a crucialrole” in defeating ICBMs during the boostphase. Experiments have begun with Gen-eral Atomics’ Reaper and Boeing’s Phan-tom Eye drones.

Sitting ducksBut slow-moving aircraft would be “sittingducks”, as Dr Lamb puts it, for anti-aircraftsystems like the Russian Buk that downedMalaysian Airlines flight MH17 over Uk-raine in July. Recent decades are “litteredwith the wreckage” of failed boost-phaseshoot-down schemes, says David Monta-gue, a formerhead ofmissile technology atLockheed (now Lockheed Martin). He co-authored a National Research Council re-port two years ago that advised the Penta-gon to give up on the idea.

A different approach could be the US

Army’s Terminal High-Altitude Area De-fence system (THAAD), which became op-erational last year in Guam, home toAmerican troops in the western PacificOcean. THAAD will also be exported; thefirst will go to the United Arab Emirates bythe end of the year. But THAAD, like Amer-ica’s Patriot missile batteries and othermissile defence-systems outside of the Un-ited States and Russia, such as Israel’s IronDome air-defence system, were developedto hit shorter-range threats and cannot in-tercept ICBMs in space. A THAAD mightscore a hit during the final approach of anICBM, but the launcher would need to bevery close to the targeted area.

Tellingly, in the remarkshe made to law-makers in Juneice-Admiral Syring re-ferred to the MDA’s “overridinggoal” asde-fending American troops and militarysites. That comment, together with the pre-sent state of the technology, suggests, fornow at least, the prospects for protectingmuch of the United States from ICBMs orsatellites secretly armed with nuclearweapons lookdoomed.7

THAADs let rip

10 Demolition technology The Economist Technology Quarterly September 6th 2014

1

FEW sights are as impressive as a tallbuilding neatly collapsing onto its own

footprint after being rigged with explo-sives by demolition experts. The spectacleattracts large crowds and lots of cameras.The levelling earlier this year of AfE Turm,a 32-storey skyscraper in Frankfurt (pic-tured below), showed how it is possible toreduce thousands of tonnes of masonry,concrete and steel to a pile of rubble with-out damaging surrounding buildings. Us-ing explosives might be quick, but withever more development in crowded cities,more discreet ways of demolishing oldbuildings are having to be found.

Bringing down a building with explo-sives is a carefully orchestrated event. The“blasters”, as contractors who specialise insuch work are known, position charges totear apart critical sections of supportingstructures. The detonations are staggered,so that material from above collapses intovoids created a moment earlier. This helpsto protect underground infrastructure inthe vicinity, such as sewers and fibre-opticcables, which might be broken if every-thing hits the ground at once.

Although blasting technologyand tech-niques have improved, the use of explo-sives remains limited, says Dario Trabucco,a buildings researcher at IUAV, an architec-ture school in Venice. This, he adds, is be-cause a quiet revolution is taking place inother ways to demolish buildings.

There are a number of reasons for this.In crowded cities there are tougher con-trols over demolition. These include pro-

tecting adjacent buildings and reducingdust, which is hard to prevent even whenknocking down walls with old-fashionedwrecking balls. An additional problem isthat when everything tumbles into a giantheap it is difficult to extract material for re-cycling, now an important way of defray-ing demolition costs.

Apart from dust, explosive charges canthrow out fragments called “flyrock”.Sandbags or netting are used to try to con-tain such projectiles, butaccidentshappen.When in August 2013 blasters demolisheda disused power plant in Bakersfield, Cali-fornia, debris was hurled beyond a safetyperimeter about 300 metres away. A spec-tator’s leg was partially severed.

Anothersort ofblastNow, though, it is possible to blow up alarge piece ofconcrete and produce almostno flyrock. The technique, developed byNon-Detonating Solutions of Cape Town,South Africa, does not rely on a conven-tional explosive material. The firm hasstarted producing a plastic cartridge calledAutoStem that is pushed into a borehole inrock or concrete just as a stick of dynamitewould be. When the cartridge is triggeredelectrically from a lead, a proprietary mix-ture ofmaterial which containsan oxidiserreacts to rapidly produce a high-pressuregas. Because the gas is tightly contained inthe borehole, its expansion splits the sur-rounding material. Even the smallest 20-gram gascartridge can breakup a cubic me-tre ofstone.

The noise from a gas cartridge explod-ing is not much louder than a firecracker.When explosives are used in quarries thearea has to be evacuated. When gas car-tridges are used, workers need to movebackby just 30 metres. Whereas explosivestend to pulverise much of the material intodust, a gas cartridge fractures it into chunkswhich are easier to move and reuse. Unlessthe cartridge is tightly confined it is almostharmless if ignited. So, according to JosyCohen, Non-Detonating Solutions’ foun-der, the gascartridgescan be shipped byairand stored on site without the restrictionsimposed on explosives.

Gas-generating cartridges are increas-inglyused to help demolish buildings eventhough they cost about 20% more than ex-plosives, says Jan Khlistovsky, co-owner ofGreen Break Technology, a Czech demoli-tions company based near Prague. The car-tridges sidestep the “45 days of paper-work” that the use of explosives requires,adds Mr Khlistovsky. This allowed the firmto charge just $1,600 to demolish a 20-me-tre concrete silo. The rigmarole of dealingwith conventional explosives would haveincreased the cost significantly.

Ferrari Démolitions, a French company,has worked out a way to fell tall buildingswithout explosives or gas cartridges. Itmakes use of the “house-of-cards princi-ple”, as a company technician put it. Engi-neers use remotely controlled powerfulhydraulic jacks to push supporting wallssideways on a mid-level floor. With thewalls gone, the top portion of the buildingimmediately drops down with enoughforce to pulverise, in rapid succession, thestoreys below.

“Jacking” buildings in this way gener-ates a bit less dust than explosives. Even so,the resulting clouds can still be thickenough to require buildings in the neigh-bourhood to be sealed. Other ways ofphysically pushing parts of the buildingapart are being developed. Excavatorswith hydraulic arms of unprecedented

Bringing the house down

Urban redevelopment: New ways are being found to demolish old buildings incrowded cities

The Economist Technology Quarterly September 6th 2014 Demolition technology 11

2 length and power are one alternative.Their use is growing in China—and not amoment too soon, says Wilson Lu, a pro-fessor of architecture at the University ofHong Kong. Many cities in China areplagued by air pollution and the pace ofdemolition and construction is frantic. It isnot unusual for buildings in China to lastonly a few decades, in part because theywere badly built.

Biting a bit offWith excavator arms that now reach up 12storeys (about 40 metres), demolitionfirms can “chew a building apart” from thetop down, says Richard Diven, an Idahodemolitions consultant. New arm “cou-plers” allow operators to switch tool at-tachments quickly without the bother ofleaving their cab. For instance, one toolmight be a pulveriser made by America’sCaterpillar. It can tear out big chunks ofconcrete and bite through reinforcing steel.A shears attachment can gnaw throughsteel columns thick enough to support a20-storey building.

For taller buildings beyond an excava-tor’s reach, smaller machines are hoistedby crane onto the roof. These use jackham-mers to break up material, with the rubblebeing placed into skips which the craneslift away and tip into lorries for removal. Abuildingnextdoor to The Economist in Lon-don has disappeared this way.

The process can be used to take giantskyscrapers apart in the most crowdedplaces on Earth. The 30-storey SunningPla-za in Hong Kong, built in 1982, is being de-molished like this by YSK2 Engineering, alocal firm. Thomas Wong, the company’smanaging director, says the job will takesome 60 workers 11 months. Dismantlingwith excavators from the top down alsoyields more recyclables than explosionswould, and much of the material recov-ered can be sold for reuse rather than scrap.Recycling, says Mr Wong, defrays his de-molition costs by as much as a tenth.

Little of mainland China’s demolitionwaste is recycled but the proportion isgrowing, says Hong Kong’s Dr Lu. This ispartly because it is becoming harder forhaulage firms to find places to dump rub-ble illegally without being photographedby angry locals with smartphones. Twonew types of disassembly could help in-crease recycling further.

One is a method called TopDownWay,developed by Despe, an Italian company. Itwas first used in 2012 to dismantle the TourUAP, a 25-storeyskyscraper in Lyon, France.Instead of spending three months erectingscaffolding, netting and sheeting (neces-sary to protect nearby buildings and thestreet below when demolishing a buildingfrom the top down with excavators),Despe built in just 20 days an exoskeleton“hat” that enclosed the top three floors.This hat was attached to the central core ofthe building and lowered as the structurereduced in size. It gave workers ample ac-cess to the building, reduced noise, con-tained dust and prevented debris from fall-ing to the street. Despe removed floors atthe rate of about two a week. All but 5% ofthe skyscraper was recycled.

An even more unusual method of de-molition was first demonstrated in 2008by Kajima, a Japanese construction com-pany, when it took down its 20-storey and

17-storey Tokyo headquarters buildingsfrom the bottom up. To do this, Kajima de-veloped a system which holds up a build-ing’s support columnswith gianthydraulicjacks. This allows workers to cut out slicesof the building from below—with the con-venience ofdoing so at street level where itis easier to remove material to be hauledaway. It is also quieter than hoisting giantskips aloft with cranes and dropping thecontents into lorries. Once a section hasbeen removed the building is lowered andthe process repeated.

What the company calls its “cut andtake down” method can remove roughlytwo storeys every ten days, says HitoshiUehara, one of Kajima’s technologists. Al-though it is 5-10% more expensive than tra-ditional demolition, the company says itsprocess is about 15% faster and well suitedto built-up areas.

Demolishing buildings with explosivesis likely to remain a spectacle in rural andmore open urban areas. It will be per-formed with what Mike Taylor, head ofAmerica’s National Demolition Associa-tion, describes as the precision with whicha brain surgeon operates. Elsewhere,though, buildings will be taken apart fromthe top down in increasingly careful ways.And some, as with Kajima’s method, willshrinkbefore your eyes.7

Kajima, a cut below the rest

12 The connected car The Economist Technology Quarterly September 6th 2014

1

Smartphones on wheels

Motoring: The way cars are made,bought and driven is changing withmobile communications. This pavesthe way to a driverless future

IN A generation from now, your journeyhome may go a bit like this. As you leave

your office, an empty car rolls up. Perhapsyou summoned it, or maybe this is a regu-lar pickup. On the way home you listen toyour favourite music, watch a televisionshow or catch up with the new�� ou bare-ly notice as the car slows down or speedsup to avoid othervehicles, except for whenit pulls aside to let an ambulance through.Some of the other cars have drivers using asteering wheel, but many of them, likeyours, have no wheel at all.

Despite that hold-up your journey ismuch faster, even though there are morecars on the road than in 2014. When you ar-rive home, the car heads off to its next cli-ent, or to park somewhere and wait for acall. You don’t know or care. After all, it’snot your vehicle: you summon a car onlywhen you need one.

Tantalising glimpses of this future arecommon today, most notably in Google’sbubble-shaped prototype of an autono-mous car. The internet giant has beenrunning Toyotas and other modelsadapted for driverless travel upand down Highway 101 in Sili-c���alley for a couple ofyears now, using on-boardsensors to keep the vehicleson the straight and narrow.

Other experiments usea different approach to en-sure safe journeys. Some3,000 drivers in Ann Arbor,Michigan, have had wireless inter-net connections fitted to their cars.These are used to feed informationto and from other vehicles andthe transport infrastructure. Thesystem will, for instance, warn adriver about to overtake a car ifthere is a chance ofa collision with anoncoming vehicle, or change a traffic lightto green if safe to do so. The number of ve-hicles involved in the project, run by theUniversity ofMichigan and largely fundedby America’s Department of Transporta-tion, could triple over the next few years.

What is happening in Michigan is partof a much broader trend: the rise of the“connected” car. This is the coming togeth-er of communications technologies, infor-mation systems and safety devices to pro-vide vehicles with an increasing level of

ble of doing a fair bit of autonomous driv-ing. For instance, the German company’snew “Intelligent Drive” package has a fea-ture which, in congested traffic moving atless than 60kph (37mph), allows the driverto let the car steer, brake and accelerate byitself. The system uses a combination oful-trasonicand radarsensorsalongwith cam-eras that monitor all around the vehicle.Because Mercedes drivers like to be com-fortable, it will even automatically adjustthe suspension before the car hits a pot-hole in the road.

Many features in modern cars are be-coming accessible to smartphones thatconnect to the vehicle. A smartphone appallows the driver of an electric BMW i3, forexample, to check the battery capacity ofhis vehicle while it is being topped up at arecharging station. Audi, part of the Volks-wagen group, is working on a systemwhich would allow a driver to get out ofthe car and use his smartphone to instructthe vehicle to park itself.

Connected cars are a marriage of twotypes of mobile technology: the mechani-cal sort, which revolutionised transport inthe 20th century, and the electronicvariety,which has transformed telecoms in the21st. A recent report by analysts at Citi-group, a bank, used data from IHS, a re-search firm, to divide the ways that mobiletelecoms are influencing motoring into

three useful groups.

The carappThe first bunch is made up of services

and applications delivered via mobile net-works to a car—either to systems that arepart of the vehicle or to devices, such assmartphones or tablets, carried by thedriver or passengers and connected to thecar wirelessly or with a cable. The most ob-vious example are “infotainment” sys-tems, which stream music, video, satellitenavigation and traffic information. Thesecond consists of services based on datasupplied from the car, such as advancewarning that a part needs to be replaced.And the third category brings togethermultiple vehicles, communicating witheach other and with smart infrastructure,from roadside sensors to traffic signals andremote data centres, to make traffic flowmore smoothly and safely.

Broadly speaking, services in the firstgroup are the most widespread already.“The cards in infotainment have beendealt,” says Andreas Mai of Cisco, a net-work-equipment giant. People alreadyhave their favourite services, like iTunes,Spotify or TripAdvisor, on their smart-

sophistication and automation. It is a pro-cess that will change not just how cars areused but also the relationship between acar and its driver. This, in turn, will affectthe way vehicles are made and sold. Even-tually, it is the connected car that may de-liver a driverless future.

The kit that enables this is starting to ap-pear in new vehicles. Some of the most ad-vanced driver aids can be specified in cer-tain Mercedes-Benz models.These cars are already capa-

2

1

phones. Surveys, though, suggest that carbuyers place a higher value on servicesthat make travelling safer, save them timeor money, or alert them to problems withtheir vehicle. These services lie mainlythough not wholly in the second and thirdgroups. But widespread availability maytake several years.

The number of cars with some sort ofnetworking ability today is small, perhapsonly 8% of the global total, according toMcKinsey, a consulting firm. But by 2020around a quarter of all cars, mainly themore expensive sort, will be online. Thebuild-up will be relatively slow becausemany old cars stay on the road fora decadeor so. But for new cars things are changingrapidly. BMW has been embedding SIM

cards for mobile connectivity in all its newcars since April. By 2020, around 90% ofallmanufacturers’ new models are likely tohave them, according to Machina Re-search, another consulting firm. The mar-ket then starts to look particularly juicy. Arecent report by GSMA, the mobile opera-tors’ trade body, says revenues from thesale of in-vehicle services, hardware andthe provision ofconnectivity itselfwill tre-ble over five years to reach $39 billion by2018. Machina reckons it could rise to astaggering $422 billion by 2022, most of itcoming from connected services to andfrom vehicles.

Car buyers are expected to be keen onconnected services once they get to knowabout them and see them in action. Thismuch is clear from the limited offerings al-ready available. The ability for the car itselfto call the emergency services automati-cally in the eventofan accident is reckonedby many drivers to be a valuable feature ofGM’s OnStar, a connected safety and navi-gation system which in effect enables a ve-hicle to function as a phone. A separate

app also allows OnStar users to lock andunlock the car’s doors remotely, start theengine and find the vehicle on a map if thedriver forgets where he parked it. GM aimsto have the service available in nearly all itscars worldwide by 2015.

But regulators are also forcing the pace.The European Union wants a system thatautomatically calls for help in the event ofa crash to be fitted to all new vehicles by2015. Russia has similar plans and Braziliancars will need to be fitted with trackers as away to reduce theft. Encouraged by theAnn Arbor test, in February America’s Na-tional Highway Traffic Safety Administra-tion said it would begin workingon a regu-lation to require vehicle-to-vehicle (V2V)communication to be fitted in all new cars.

On the digital dashboardDifferent applications require differenttechnologies. A search for a parking spacewould probably go overpublic mobile net-works from an app, whether on thedriver’s smartphone or one running on adigital dashboard. For safety features, suchaspreventinga car from pullingout in frontof another, V2V communication is essen-tial, says Kurt Sievers of NXP, a semicon-ductor company. Public networks will betoo slow for this and may lack the capacity.His company is making systems with dualantennae to cope with reception difficul-ties, because radio waves from moving ve-hicles tend to bounce off buildings andother surfaces. Authentication of signalsmatters too, to preventcars takingunneces-sary avoiding action.

With increased connectivity betweencars, driver aids will become much moresophisticated. A connected car would, forinstance, receive not just informationabout a hazard detected by its own sen-sors, but also alerts from a vehicle farther

along the road or around a blind corner. Connectivity can also help provide

more real-time information about traffichold-ups, beyond that already provided bysatellite-navigation devices. The additionof vehicle-to-infrastructure communica-tion (V2I) takes things further still. Where-as the connected cars in Ann Arbor canchange the timing of traffic lights, a combi-nation of V2V, V2I and automated drivingcould do away with traffic lights complete-ly. Cars could be co-ordinated so that theyavoid one another at road crossings. Nothaving to stop at road crossings would re-duce congestion.

The sensors in vehicles that checkthings like tyre and oil pressures, as well asbrakes and engine performance, will alsohave a role. Pavan Mathew of Telefónica, amobile-network operator, points out thatmany drivers dread the moment when adashboard warning light flicks on. Remotemonitoring and messaging can swiftlysend a note to the driver about the extentof the problem.

Vehicles’ diagnostic systems could alsopickup faultsbefore theyare manifested asblacksmoke pouring from an exhaust pipeor a horrible grinding noise from the en-gine. Cars could then be brought in for re-pair before trivial problems develop intobigones. Followingthe lead ofTesla, a Cali-fornian maker of electric cars, more faultsmight one day be fixed remotely over theinternet by a software upgrade.

Indeed, checking on cars remotely hasplenty of other possibilities that may re-duce (or worsen) stress levels. Online ser-vices will allow, for instance, closer moni-toring of the driving behaviour ofteenagersbeyond the basicwarningsof ag-gressive braking or exceeding speed limitsthat the “blackboxes” supplied bysome in-surance companies presently provide.And not just younger drivers. Insurers arelikely to offer any driver a lower premiumif technological monitoring of his drivinghabits shows he is being careful.

Exactly who will deliver all these newmotoring services is far from clear. It is byno means certain that it will be traditionalcarmakers, even though they are all busilydeveloping, making and marketing in-creasingly connected vehicles. In the pastconsumers have expected the new tech-nologies that appear in cars quickly to be-come standard features for which they paylittle if anything extra. Electric windows,anti-lock brakes and power steering arenow almost universal.

The connected car, however, hascreated powerful new competitors in the

The Economist Technology Quarterly September 6th 2014 The connected car 13

14 The connected car The Economist Technology Quarterly September 6th 2014

2 motor industry’s traditional supply chain.And some of those new competitors arekeen to win themselvesa bigslice of the ac-tion. These are mobile-telecoms operators,makers of networking gear, developers ofV2V and V2I technologies, producers ofconsumer hardware and systems, soft-ware firms and creators ofmobile apps.

Cars will become bundles of differenttechnologies, not only of devices but alsoof consumer brands, all vying for thedriver’s attention in a sometimes uneasyalliance with carmakers. Apple and Goo-gle are locked in competition for control ofthe digital dashboard. In response to Car-Play, a vehicle-infotainment system devel-oped by Apple, Google in June launched arival called Android Auto.

Mobile-phone operators see the con-nected car as yet another device to behooked up to their networks. In America,AT&T is letting drivers ofGM cars add theirvehicles to their data plans, alongside theirsmartphones and tablets, for $10 a month.In future, which mobile network you usemay affect your choice of car. In a recentpoll Nielsen, a market-research firm, foundthat half of Americans who owned carsmade since 2009 would be less likely tobuy a new car if it had a different data planfrom their smartphone.

Invisible competitorsNot everyone trying to get in on the act willbe visible to the driver. All the data going toand from cars and infrastructure will haveto be transmitted and processed. That addsto demand for chips, network equipmentand data centres. Cisco, for example, envis-ages a lot of processing taking place not inthe “cloud” of central data centres butmore speedily and conveniently within a“fog” of intelligent networks.

Carmakers know they will have toshare the benefits of the connected car.Some seem gloomy about their prospectsofgetting any of them at all. Fiat Chrysler’sboss, Sergio Marchionne, is worried that itwill cost his company money to “provide avenue to host other people’s parties”.Some carmakers see more of an opportu-nity to profit as they could benefit beyondtheir share of the monthly charges for con-nectivity. Using the data to tweak the de-sign and performance of their vehicles byidentifying components that are more like-ly to cause problems will both help themto improve the cars they produce and cutwarranty costs. Good connectivity shouldhelp to reinforce brand loyalty too.

The relationship between carmakersand their customers is at arm’s length at

Fiat Chrysler’s boss, Sergio Marchionne, is worriedthat it will cost his company money to “provide avenue to host other people’s parties”

present, operating through a dealershipsystem that is reminiscent of that betweenhandset-makers and operators. After sell-ing a car through a franchised dealer, fur-ther interaction with car buyers is limitedto a dealership visit every couple of yearsfor a service (or sooner if there is a pro-blem). Connectivity will bring the custom-er and carmaker closer together. Ship andforget will be supplanted by ship and up-date, which is what makers of computersand mobile devices do already. So far carcompanies seem unclear about what thiswill mean for how they do business.

Getting closer to their customersshould at least make the carmakers moreresponsive. The data can help manufactur-ers and dealers target customers more effi-ciently. As well as sending details of offers,dealers might better fit a particular car to adriver through an analysis of individualdriving habits. They could suggest extrafeatures that would suit some motorists,from hybrid technology to modest add-ons. Some carmakers are already milesalong this road. Elon Musk, Tesla’s boss,laughs at the suggestion that his customerswould accept anything less than a high de-gree of connectivity and interaction whenhe sells them an electric car.

The data could help customers knowmore about cars too. Motorists will havethe ability to find out the actual milesper gallon a car will do in thereal world rather thantrust the claims madeby car companies,which use a box oftricks to make their vehi-

cles unrealistically frugal during tests.Carmakers, usually conservative and

slow-moving, are getting ready. Aside fromthe engine, body and interior, cars alreadycontain lots of electrical architecture. Mostof the big firms have set up connected-cargroups to work alongside their electricalengineers to ensure that the hardware andsoftware required for connectivity fit. De-troit’s car guys are deferring to techies,poached from the software industry, whoare adept at dealing with app-makers andthe like. Carmakers are looking closely atTesla, which describes itself as a “softwarecompany that builds cars”, for inspiration.

Connectivity will eventually changethe way cars are integrated into transportsystems. Car sharing, either through carclubs run by the big rental firms or peer-to-peer services, will be far easier when com-munication between vehicles and poten-tial passengers is seamless and any car canbe accessed and operated securely by anysmartphone. Making journeys using sever-al forms of transport, including a car, willbe smoother if it is easier to find car-shar-ing locations or parking spaces close toconnecting points for trains or buses.

And with increasing automation andconnectivity there will be lessneed to haveto own or drive these vehicles yourself. To-day’s experimental autonomous carsstuffed full of on-board sensors are only

part of the solution. The develop-ment of systems that let cars

talk to cars, and to the worldbeyond, will be just as im-

portant on the road to a driver-less future.7

The Economist Technology Quarterly September 6th 2014 Air-traffic control 15

1

IN A windowless industrial building onthe outskirts ofMadrid a group ofpeople

are watching a series of coloured symbolsmove steadily across a bank of computerscreens. Each icon representsan aircraft fly-ing over southern Europe. In an adjacentroom another group are monitoring flightsover part ofAsia, and next door all eyes areon South America. These flights are not“live” butare simulated by Indra, a Spanishtechnologycompany, to train controllers inthe operation of a new generation of air-traffic-management systems that promisesto make flyingmore efficient by shorteningflight times and reducing delays.

What is different about these virtualflights is that some are “free-routing”,which means pilots have the freedom toset their own courses instead of followingone another along established flight corri-dors, as they presently do. Free-routing al-lows an aircraft to fly more directly to itsdestination, which for European journeysalone would knock ten minutes off aver-age flight times, thus saving fuel and reduc-ing carbon-dioxide emissions.

On the face of it, free-routing seems likea disaster in the making. If fast-moving jets

can fly where they want the risk of colli-sion would appear to rise. Yet the control-lers in Madrid are relaxed and, apart fromissuing a few course corrections to avoid aspotofbad weather, leave the aircraft to geton with it. This is made possible becausethe trajectory of each plane has beenworked outbycomputers some 25 minutesin advance and the pilots have alreadybeen informed ofany adjustments neededto prevent a potential conflict. Providedeach aircraft sticks to its flight plan, there isno need for the controllers to intervene.

Traditionally, air-traffic controllers haveplayed a more proactive role in keeping air-craft apart. The corridors which jets flyalongfunction much like lanes on a motor-way. They pass through sectors and eachsector ismonitored byair-trafficcontrollerswith the assistance of radar. When an air-craft is about to entera new sector, the pilotand controller communicate by radio. Thecontroller then gives the pilot instructionsto maintain a safe separation, both verti-cally and horizontally, from other aircraft.

It is a tried and trusted method, but onethat will struggle to cope with future de-mand forair travel. Thiswill be huge. In the

past 40 years the numberofairline passen-gers worldwide has grown tenfold to some3.1 billion in 2013. By 2030 it is expected toreach over 6.4 billion.

Flight corridors frequently follow his-toric routesand zigzagaround. Manyof theroutes which cross America are based onwhere hilltop beacons were lit to guideCharles Lindbergh’s mail flights in the1920s. Plenty of radar systems still resem-ble 1940s technology and provide only alimited “view” of what is in the air. And inEurope, flights have to negotiate a laby-rinth of64 air-traffic-control areasoperatedby different national authorities. All thisadds to journey times and puts constraintson the system because ofthe need to main-tain the safe separation ofplanes.

Flying aheadA much-needed revamp is under way. Al-though suffering from delays and budgetconstraints, America’s NextGen air-traffic-modernisation programme is slowly tak-ing shape. Europe is part way through theSingle European Sky initiative, which issupposed to increase co-operation be-tween a reduced number of control cen-tres. Japan also has a project in hand to ren-ovate its air-traffic-control systems.

One element of this modernisation in-volves fitting new kit to aircraft. This is asystem called Automatic Dependent Sur-veillance-Broadcast (ADS-B). It will becompulsory for jets in Europe by 2017 andin America by 2020. ADS-B uses satellitenavigation for pilots to determine their po-sition and is generally more accurate thanradar and radio-navigation aids. This al-

Free flight

Flight safety: As more aircraft take to the sky, new technology will allowpilots to pick their own direct routes but still avoid one another

16 Air-traffic control The Economist Technology Quarterly September 6th 2014

2 lows aircraft to be safely spaced closer to-gether, which permits more planes to be inthe air at the same time. Crucially, though,it also establishes a data link to control cen-tres and to otherplanes by regularly broad-casting an aircraft’s identification sign, itsposition and other information.

These data, when combined with theknown trajectory of the aircraft, meanflight-management can now be “based onwhere we know the aircraft will be at anyparticular time,” says Gonzalo Gavin, thedirector of a programme at Indra to installsuch a system at a control centre in Prest-wick, Scotland, run by NATS, a British com-pany. The Prestwick centre provides en-route services across northern Britain andfor flights crossing part of the busy NorthAtlantic. Known as iTEC (for interoperabil-ity Through European Collaboration), thetrajectory-based system was developed byIndra with air-traffic providers in Spain,Germany, Britain and the Netherlands aspart of the Single European Sky initiative.

In a typical flight, a pilot may climb, de-scend, change course at various points andspeed up or slow down numerous times.But with trajectory-based management,

the flight should be smoother and shorter,says Mr Gavin (see illustration). And it ismore likely to arrive on time.

The ability to predict the arrival timemore accurately should mean less circlingin holding patterns while planes wait toland, says Alastair Muir, operations direc-tor at Prestwick. That would allow moreaircraft to use what is called a “continuousdescent approach” when coming in toland. This is a procedure which involves alonger descent, more like a steady glide to-wards the runway. It requires less enginethrust than having to level out at variousstages of the approach, so it saves fuel andis also quieter.

A wholesale switch to free-routing willnot take place overnight—the aviation in-dustry is notoriously cautious in introduc-ing new technologies and procedures. Atfirst pilots are likely to pick from a numberof available routes before free-routingcompletely takes off.

It should enhance safety with an earlyalarm should something go wrong. Al-though no one knows what happened toMalaysian Airlines flight MH370, whichdisappeared in March, the new systems

would have alerted controllers that the air-craft was not keeping to its trajectory soonafter it changed course.

Beinghighlyautomated, the newgener-ation of air-traffic-management shouldalso help with the commercial use of civil-ian drones. Aviation authorities are facingincreasing pressure from companies to al-low drones to be used for a variety of ap-plications, ranging from aerial photogra-phy to surveying, search and rescue,delivering goods and providing temporaryWi-Fi. Guidelines are slowly emerging, butgenerally the operation of civil drones re-mains restricted in most countries, particu-larly the United States.

With data-driven systems like ADS-B

and trajectory management, monitoringthe flight path of a drone can be automat-ed, says Benjamin Trapnell of the Universi-ty of North Dakota, which was one of thefirst institutions to launch graduate coursesin operating drones.

Test flights by BAE Systems in Britainhave also shown how a drone can bemade to respond to air-traffic-control in-structions. This is done by having radiocommunications to and from the drone re-layed via a “pilot in command”, whowould be a drone operator on the ground.That operator might well be in charge ofmore than one drone. With the cost of op-erating a drone only a fraction of that of ahelicopter or light aircraft, the civilian useofunmanned aircraft is bound to make thesky an even busier place.

This is yet another reason why the long-awaited modernisation of air-traffic con-trol is welcome, says Andrew Charlton,head of Aviation Advocacy, a Swiss-basedconsultancy. But he thinks it should gomuch further. Organising more airspacealong functional lines rather than aroundnational borders, as much of it now is,would greatly improve efficiency—espe-cially in Europe. And with systems basedon data, he says more competition shouldbe possible, giving pilots a choice of air-traffic-management providers.

Nevertheless, the EU gushes about theprospects. It expects trajectory-manage-ment to enhance safety by leaving lessroom for human error. It hopes the SingleEuropean Sky initiative will provide theability to handle three times as manyflights, cut air-traffic-control costs and pro-duce savings for airlines worth some €9billion ($12 billion) a year. It also says thataircraft will on average land within oneminute of their scheduled arrival time.Weary air travellers will be forgiven if theythink that is a bit ofpie-in-the-sky.7

Flight paths follow historic routes and often involve stacking

Free-routing is possible, reducing fuel consumption and CO2 emissions

Flight levels given to pilots by controllers during the flight

A smoother trajectory requires less controller intervention

Smarter skies

Effect of trajectory-based air-traffic management

Sources: INDRA; The Economist

TODAY TRAJECTORY-BASED

With trajectory-based management, the flightshould be smoother and shorter

The Economist Technology Quarterly September 6th 2014 Brain scan 17

1

HE IS a professor ofgenetics at HarvardMedical School, but George Church is

also a vegan, cannot hold a tune, gave updriving due to narcolepsy and suffersheadaches after running. TechnologyQuarterly did not discover these intimatedetails through surveillance, interrogationor going through his trash, but simply bybrowsing a website where Dr Churchmakes such particulars freely available,and much more besides. The PersonalGenome Project (PGP), a medical studydesigned by Dr Church and for which hewas the first subject, even allows visitorsto download his entire genome and rum-mage through his DNA.

It is all part ofa grand experiment tohelp researchers explore the interactionsbetween genetics, environment, behav-iour and disease, with the ultimate goal ofdeveloping customised therapies forindividuals. The only way to do this, DrChurch believes, is with complete open-ness. “It really is hard to do good scienceon closed data sets,” he says. To compare,say, a study on autoimmunity in a groupofpeople with a study on their generalhealth could well prove impossible be-cause ofanonymity. But with open data,Dr Church and his colleagues have man-aged to conduct 25 different studies on thesame group in just one month.

The idea is that linking genes to out-comes, whether deadly diseases or talentslike singing, requires a huge array of rawdata about people’s lives, diets and theirsurroundings; data that do not sit wellwith traditional notions ofprivacy. DrChurch is seeking volunteers willing towaive confidentiality and lay bare theirgenetic code, medical records and dailyhabits to the world. More than 3,500 peo-ple have done so. While the PGP expects toeventually enroll 100,000 subjects, DrChurch is already dreaming much bigger:“We’re constructing it a little like Wikipe-dia, which has beat out all the proprietaryclosed systems. Ifenough people seeenough benefit from it, it could scale to abillion people.”

Such an undertaking would have beenunimaginable, and unimaginably expen-sive, a decade ago. Dr Church has donemuch himself to make it possible. His 1984Harvard PhD included the first method for

direct genome sequencing: determiningthe exact order ofnucleotides within DNA.These nucleotides contain four nucleo-bases (adenine, guanine, cytosine andthymine) that form base-pairs and give theDNA molecule its famous double-helicalstructure. He also came up with the ideaof“multiplexing”, where many pieces ofDNA are sequenced simultaneously ratherthan one by one.

In the late 1980s Dr Church helpedorganise the internationally funded Hu-man Genome Project, whose aim was tosequence all 3.3 billion base-pairs within ahuman genome. Even before it started, hewas not satisfied with the programmebecause the goal was one genome, and avery expensive genome at that. Thesprawling effort eventually took15 years,cost around $3 billion and delivered agenome that was a blend ofdifferentindividuals and riddled with gaps.

Although it was an historic milestone,it had relatively little value in practical,personal or medical terms, says DrChurch. “What I really wanted was foreverybody to have their genome andideally everybody to share their genome,and for that we needed to bring the priceway down.”

Genetics to goIn 1994 his automated technologies led tothe first commercial genome sequence,that ofa bacterium that causes stomachulcers, and later to dramatic improve-ments in the accuracy and cost ofsequenc-ing human genomes, reducing the cost toaround $1,000 today. Since 2007, DrChurch has co-founded 12 biotech compa-nies and advised many more.

One of these, Genia, is commercialis-ing a process called nanopore sequencingthat Dr Church first devised in 1988. Dis-tinct polymer tags are attached to each ofthe four nucleotides poised to contributeto a single molecule of replicating DNA. Asthey react, the tags are released near aprotein layer full of tiny holes called nano-pores. Each tag blocks the flow ofelectricalions across the layer in a different way.Because it relies on electronics rather thanoptics, nanopore sequencing promisesfaster, cheaper sequencing. Dr Churchholds up a fingernail-sized chip containing128,000 nanopores that he reckons willbring the cost ofsequencing down to $100.In June, Genia was acquired by Roche, aSwiss pharmaceuticals giant.

Sequencing a genome is one thing, butinterpreting it and understanding it isquite another. Dr Church’s endeavours in

Welcome to my genome

George Church is a genetics pioneerwhose research spans treatingdiseases, altering bodies and adesire to breed woolly mammoths

18 Brain scan The Economist Technology Quarterly September 6th 2014

Offer to readersReprints of Technology Quarterly are availablefrom the Rights and Syndication Department.A minimum order of five copies is required.

Corporate offerCustomisation options on corporate orders of100 or more are available. Please contact us todiscuss your requirements.For more information on how to order specialreports, reprints or any queries you may haveplease contact:

The Rights and Syndication DepartmentThe Economist20 Cabot SquareLondon E14 4QWTel +44 (0)20 7576 8148Fax +44 (0)20 7576 8492e-mail: rights@economist.comwww.economist.com/rights

2 that field are equally impressive. In 2007he co-founded Joule Unlimited, a biofuelscompany that hopes to use geneticallyengineered bacteria to convert wastecarbon dioxide directly into fuels. As a lotof the problems with the bioreactor andthe genetics have been solved, he says:“Making new petroleum should be assimple and straightforward as brewingbeer.” But the margins are tight and it isunclear yet whether the process will becommercially competitive.

Another project that excites Dr Churchis Revive and Restore, a controversial effortto bring backanimals and plants fromextinction. There are certain species, hebelieves, that are good for humanity andin order to keep them alive it is necessaryto conserve the whole ecosystem they livein, too. To do that, he says, may well re-quire reviving some species that are al-ready extinct.

For example, Dr Church thinks thatwoolly mammoths could help prevent theArctic permafrost from melting. Theirgrazing would invigorate the flora growingon the surface, which would provide moreprotection from the sun. His laboratory isdeveloping a robotic system called mul-tiplex automated genome engineering(MAGE) that can perform up to 50 differentgenome alterations at nearly the sametime, creating billions ofvariants in amatter ofhours. MAGE would allowscientists to start with an intact genome ofa living Asian elephant and change itwholesale into one that is comparable toan extinct mammoth, using informationpieced together from frozen fragments ofmammoths. Passenger pigeons, dodos,giant auroch cattle and even Neanderthalsmight follow.

But de-extinction is not ambitiousenough for Dr Church. Synthetic genomicshas the potential to change the course ofevolution, “with the difference that [it] willbe under our own conscious deliberationand control,” he writes in his book“Rege-nesis”. Ultimately, he thinks that willmean altering humanity itself.

A project that interests him would beengineering human cells to resist allknown viruses. He suggests two ways ofdoing this. The first would involve creating“mirror” humans—recoding a person’sDNA to switch the chirality, or handed-ness, of their entire body at the molecularlevel. It would render that person immuneto viruses, but at the cost of them beingunable to digest most normal foods orinteract with the natural microbiome thathelps to keep their gut healthy.

Slightly less drastic might be using atool like MAGE to make hundreds ofchanges to a person’s genetic code, whichconverts genetic information into usefulproteins. Altering enough of the rightnucleotides at once could eliminate thespecific sequences that viruses need toreproduce, but still produce the necessaryproteins required for that person’s well-being. “This process is probably easierthan changing ourselves into mirror peo-ple,” says Dr Church.

But would anyone want to go to suchan extreme? In the future Dr Church sees aworld in which individuals tinker withtheir DNA to eliminate diseases, give theiroffspring extra abilities or simply to lookmore attractive��ery few people currentlytry to obtain information about theirpropensity to carry diseases, which DrChurch reckons is one of the best ways ofcontrolling disease. “But we are willing tofix things that aren’t broken, for examplewith cosmetic surgery.” This is why DrChurch thinks there will be a desire forpeople to change themselves substantial-ly. “We already are radically different fromour ancient ancestors, augmented withcellphones, computers, cars and jets. Todraw a sharp line between physics andbiology doesn’t make any sense.”

To boldly goTo travel beyond the Earth, astronautscould also have their bodies altered to givethem a better chance ofsurviving thejourney. They could be genetically engi-neered to resist radiation and osteoporo-sis, a weakening of the bones whichwould result from prolonged weightless-ness. Those that remain on Earth could bealtered to reduce their carbon footprint.“All the discussion is about how to makebuildings bigger rather than people small-er,” he told a conference at the Massachu-setts Institute ofTechnology earlier thisyear. “We’re well beyond Darwinianlimitations to evolution. Evolution rightnow is in the marketplace,” he now says.

Even objects could be revolutionisedby genomics. DNA is a good way ofstoringinformation. It is possible to use it to storedata for 700,000 years at a million timesthe density of today’s disc drives, says DrChurch. In 2012, he encoded “Regenesis”into DNA and made 70 billion copies of it.Dr Church is beginning discussions withdata-storage companies about commer-cialising the technology.

The fantastical possibilities ofgeno-mics and the rapid democratisation of itstechnologies raise the spectre ofbioterro-

rism. Synthetic biology is potentially moredangerous than chemical or nuclearweaponry since organisms can self-repli-cate and spread rapidly, says Dr Church.“There has to be industry-wide surveil-lance, licensing and rigorous testing. Butit’s futile to try to keep things out of thescientific literature. Even ifyou could stopthe dissemination of results, it wouldmean that only the bad guys would doresearch and none of the good guys couldsee what’s going on.”

Now entering his seventh decade, DrChurch has no plans to retire. “Reversal ofageing is high on my list of things to do,and not just because I’m getting old,” hesays. The tools are now available to makerapid progress. “We’ve taken my 60-year-old fibroblast [connective tissue] cells,changed them into a different form andthen back to fibroblasts and they’re youngagain, most of the time. Turning that into awhole body therapy is another leap butmy point is that we can do it.”

Ifhis body fails before the science isready, Dr Church has a backup plan, liter-ally. He was one of the scientists behindthe BRAIN Initiative (Brain Researchthrough Advancing Innovative Neurotech-nologies), a $3 billion, decade-long projectannounced by President Obama in 2013 tomap the human brain. Innovation inbiotechnology is proceeding rapidlyenough to consider copying a brain. “Ifwecan come up with a way ofbacking up mybrain into another that I have in my back-pack, we’ll do it,” he says with a smile.“People talk themselves out of things veryeasily. Things that they thinkare a millionyears away or never, are actually fouryears away.”7

“Making new petroleum should be as simple andstraightforward as brewing beer.”