IELTS Academic Reading Practice Test 165

For most of human history the deepest parts of the ocean were beyond reach. A diver on a single breath can descend a few tens of metres, and armoured suits added little. The obstacle is pressure: at a depth of 10,000 metres every square centimetre of a surface carries a weight of about a tonne, enough to crush an ordinary steel chamber. The first craft to take people into this environment was the bathysphere, a steel ball with thick quartz windows, lowered on a cable from a ship. In 1934 the American naturalist William Beebe and the engineer Otis Barton descended in one to 923 metres off Bermuda. It could move only up and down.

The man who freed deep diving from the cable was Auguste Piccard, a Swiss physicist who had already become famous by rising into the stratosphere in a sealed aluminium gondola hanging beneath a hydrogen balloon. Piccard's insight was to treat a diving craft as an underwater balloon. A balloon climbs because the gas inside it is lighter than the air outside; a vessel could sink and rise if it carried a liquid lighter than water. The craft he designed, which he named the bathyscaphe from the Greek words for 'deep' and 'boat', had two parts: a large, thin-walled float filled with petrol and, slung beneath it, a small steel sphere, called the gondola, for the crew.

Petrol is less dense than water and therefore provides lift, but its second property mattered even more. Like all liquids it is almost incompressible, so a float filled with petrol keeps its shape at great depth. A hollow float full of air would have been flattened long before it reached the bottom. Because the petrol carried much of the load, the float could be built of relatively thin steel. It was open at the bottom, so that seawater could move in and out as the petrol changed volume with pressure; on the way up the petrol expands and a little of it escapes, and the tank is never split.

The second element was ballast. Iron shot was stored in hoppers and held in place by electromagnets that the crew could switch off. With the shot on board, the craft was heavier than the water it displaced and sank. Releasing part of it allowed the pilot to hover at a chosen depth, and dropping all of it sent the bathyscaphe back to the surface. Here lay the great advantage over the bathysphere: no cable was needed, and the vessel could stop wherever its crew wished.

Piccard's first bathyscaphe was built in Belgium with money from the country's national research fund, whose initials gave the craft its name, FNRS-2. Launched in 1948, it made a series of unmanned descents off the Cape Verde islands to about 1,400 metres, but the trials were troubled and the vessel was damaged in a storm at sea. Belgium lost interest and handed it to the French navy, which rebuilt it as the FNRS-3. In 1954, off West Africa, that craft carried two French officers below 4,000 metres, a record at the time.

Piccard meanwhile had designed a larger vessel, the Trieste, built in Italy and launched in 1953. Its float was well over ten metres long; beneath it hung a steel sphere with thick walls and a single window of acrylic plastic, shaped as a cone so that the growing pressure of the water forced it more tightly into its seating. The United States navy bought the Trieste in 1958 and shipped it to California. Auguste's son Jacques, who had worked on the design since boyhood, went with it as its pilot.

The most celebrated dive took place on 23 January 1960, when Jacques Piccard and the American naval officer Don Walsh descended into the Challenger Deep, a trench in the western Pacific. The descent took nearly five hours. At about 9,000 metres the window cracked with a loud report, but it held, and the two men went on to the bottom, roughly 10,900 metres down. They stayed for about twenty minutes and reported seeing a flatfish, a claim that later researchers have questioned.

The dive proved that people could reach the lowest point of the sea, yet the bathyscaphe was an awkward research vessel. Having no means of propulsion in the horizontal plane, it could not explore a trench but only descend and rise again in one place. It needed calm weather, a large escort ship and hours of preparation for each dive, and its crew had very little room to work.

Within a decade, lighter submersibles driven by batteries and propellers, among them the American Alvin, had taken over most scientific work, and the bathyscaphes were retired. The principle of variable ballast survives in every deep-diving vehicle built since. In 2012 the film director James Cameron reached the Challenger Deep alone in a vertical craft of his own design, a place that Piccard and Walsh had taken nearly five hours to reach.

The bathyscaphe was therefore a machine of its moment: the first vessel that made the deepest ocean a place where people could arrive, even if they could not yet work there.

Virtual reality (VR) is a computer-generated world that a user enters by putting on a display which covers the eyes. What separates it from cinema or television is that the image responds to the viewer: turn your head and the scene turns with you. The pictures are produced in stereo, one for each eye, and the headset carries sensors that report its position many times a second. The aim is not simply a large picture but 'presence', the sensation of being inside a place rather than looking at a screen.

The technology is older than its recent popularity suggests. In 1965 the computer scientist Ivan Sutherland described what he called the ultimate display, a room in which a computer controls the existence of matter, and in 1968 he and a student built a head-mounted system at Harvard University. It was so heavy that it had to be suspended from the ceiling above the wearer, and it was nicknamed the Sword of Damocles. The phrase 'virtual reality' was popularised in the late 1980s by Jaron Lanier, whose company sold headsets and a glove that could be used to grasp virtual objects.

Depth is created in the same way as in human vision. Each eye receives a slightly different view of the world, and the brain reads the difference as distance. A headset therefore draws two images from viewpoints a few centimetres apart, matching the spacing of the eyes, and displays them on separate screens or on two halves of a single screen. Lenses between the panel and the eye bend the light so that a display only a few centimetres from the face can be focused on comfortably, and they widen the view so that the image fills as much of the visual field as possible; the best headsets approach a hundred degrees.

Tracking is what makes the illusion convincing. Accelerometers and gyroscopes measure how quickly the head is turning, while magnetometers provide a reference direction; cameras fixed around a room, or built into the headset itself, pin down its position. The software must then redraw the scene before the eyes notice that it has failed to keep up. Delays of more than about twenty milliseconds are perceived as a lag, and a headset that lags feels wrong in a way users describe as swimming or drifting. The conflict between the picture and the vestibular system of the inner ear, which registers balance, is thought to explain much of the nausea that some users experience. Displays that refresh ninety times a second or more help to keep the picture in step.

A subtler problem comes from the way the eyes work. To look at something close, they turn slightly inwards, or converge, and they also change focus, or accommodate. In ordinary life these two responses happen at the same distance, but in a headset the eyes converge on an object that appears to be two metres away while their focus stays fixed on a panel a few centimetres from the face. The brain receives contradictory signals, and many users report fatigue or headache after a session. Researchers are testing displays whose focus can change, and screens that reproduce the directions from which light would arrive in a real scene.

Headsets are not the only route to immersion. In the early 1990s a group at the University of Illinois at Chicago built the first CAVE, a room whose walls and floor are screens onto which images are projected. Several people can stand inside it at once and see the same scene from their own positions, which makes it useful for engineering and architectural design, but the equipment is costly and needs a large, dedicated space.

Vision and movement alone are not enough for every task. Spatial audio, delivered through headphones, tells a user where a sound comes from, and tactile feedback is supplied by controllers that vibrate or by gloves fitted with small motors. Haptics of this kind remains crude beside the richness of touch in ordinary life: a trainee surgeon can feel that a tool has met resistance, but not the texture of the tissue, so researchers continue to look for ways of simulating it.

The most successful applications are those in which a mistake is expensive or the real situation is hard to arrange. Pilots and surgeons rehearse procedures in simulators, and firefighters practise entering smoke-filled buildings. In clinical psychology, VR is used to expose patients gradually to the situations they fear, such as heights, flying or public speaking, and to help people with post-traumatic stress disorder revisit painful memories under controlled conditions. Burns units have used virtual scenes to distract patients during dressing changes, and physiotherapists use games in which stroke patients must reach for objects. Architects walk clients through buildings that do not yet exist.

The obstacles that remain are practical as much as conceptual. Wireless headsets have removed the cable that once tied users to a computer, but the processor and battery must still be carried on the head. Drawing a high-resolution image across a wide field of view is expensive, and one promising answer is foveated rendering, in which cameras follow the eye and only the small area at the centre of gaze is drawn in full detail, since the rest of the retina cannot resolve fine detail anyway. How far such systems should be used with children, and how much data about a user's gaze should be collected, are questions researchers have only begun to address.

Examinations that were once written by hand in silent halls are increasingly taken at a keyboard. Several large testing organisations now allow candidates to choose between paper and screen, and international assessments of school pupils are delivered on computer. The change is usually presented as a matter of logistics, but it is not: the medium in which a candidate answers alters what the examination measures. My own view is that the shift is on balance right, but that it will be fair only if papers are redesigned for the screen rather than merely transferred to it.

The case for the change begins with practicalities. A digital script cannot be lost in the post or mislaid in an examiner's office, and it reaches the marker within seconds. Illegible handwriting, which has always penalised some candidates for reasons unconnected with their subject, disappears. Accessibility improves markedly: text can be enlarged, colours and contrast changed, a screen reader used or a spellchecker approved for the purpose. Because the answers arrive as data, awarding bodies can also analyse results more quickly and in finer detail.

There are costs as well. Reading a long text on a screen is not identical to reading it on paper. A meta-analysis that pooled more than fifty studies found a small but consistent advantage for paper, and the gap widened when readers were short of time. This should not be exaggerated: the difference is modest, it is smallest for the short texts used in examinations, and readers who have grown up with screens show less of it. Even so, an awarding body ought to be able to show that it has checked, rather than assumed, that its digital paper is no harder than the printed one.

Then there is writing itself. Students generally produce more words when they type than when they write by hand, and where a mark scheme rewards development, those extra words may turn into marks. That is not necessarily unjust, since fluency with a keyboard is a real skill, but it becomes so if candidates who type slowly choose paper while confident typists choose the screen; the medium then becomes a second competition running alongside the subject being examined. Handwriting speed, after all, was never taught systematically and varied enormously, so the old arrangement was not neutral either.

The features of a computer alter the task further. Automatic correction, word counts, copy and paste and access to the internet all change what is being assessed. Boards disable most of them, and a locked browser makes a digital examination remarkably secure. Yet a candidate who can delete a sentence and begin again without trace is working in a different way from one who must cross out and carry on; the digital paper is never simply the printed paper in another form.

Nor do all subjects fit the same mould. Mathematics and the sciences depend on symbols, diagrams and worked calculations, and although on-screen tools exist for all of these, many teachers argue that an equation written by hand is quicker and clearer than one assembled from a menu of symbols. Essays and source-based questions transfer far more easily.

Fairness therefore depends on familiarity. A pupil who has used a laptop since infancy is at ease with a trackpad and a scrolling page, while a pupil who has written only in exercise books is not, and the difficulty has nothing to do with the subject being tested. Where equipment is shared or unreliable, a power failure can cost a candidate the recognition of years of work, so every digital system needs a printed plan for the day it breaks down.

Security is raised by both sides. Printed papers can be photographed and sold before the examination; digital papers can, in principle, be intercepted, but they can also be encrypted and released only minutes before the start. Neither medium is inherently safe, and in both cases the real safeguard is the conduct of the people who handle the material, together with the penalties that await those who abuse it.

What would a defensible transition look like? Keyboarding should be taught as a basic skill, as handwriting once was. Candidates should sit practice papers on the same platform they will meet on the day, and both formats should remain available during the transition while schools and pupils adjust. Results should be scrutinised for bias, comparing the performance of groups who chose each medium. Above all, questions should be written for the screen from the outset, using its possibilities instead of reproducing a printed booklet.

So should examinations be taken on computers? For most subjects and most candidates, yes. But a medium is never neutral, and a fair examination on screen demands more than a server and a room full of laptops: it demands teaching, practice, evidence, and a plan for the day the technology fails.