The Crowd and the Cosmos: Adventures in the Zooniverse
Page 11
here on in on carefully controlled experiments, and this early scientific attempt at involving a crowd cannot, it’s true, be said to have contributed much to the sum of human knowledge. It was,
in fact, about as much of a damp squib as could be imagined,
though perhaps that was inevitable given the fate of the display
itself. Fog prevented spectators from enjoying a proper view, and
the fireworks set fire to the pavilion which had been specially
constructed in the centre of the park. Even on a clear day, though,
I don’t think many would have participated in the experiment;
having struggled to find the right questions to ask Galaxy Zoo
volunteers, it is nice to think that prospective citizen science
practitioners could have learned lessons about writing clear
instructions even at this early stage.
No Such ThiNg aS a New ide a 79
Robins’ efforts, abortive though they were, are often cited as
the first scientific crowdsourcing effort, but there is at least one
earlier example. It involves a total eclipse of the Sun that took
place in 1715, and no less a figure than Edmund Halley, famous for
his eponymous comet and a leading light in much that was good
about eighteenthcentury science. A total eclipse is a freak of
nature, the result of the apparent coincidence that the Moon and
the Sun appear the same size in the sky. (I say ‘apparent’ coincidence because my Oxford colleague Steve Balbus has argued that the tides that result from this arrangement are particularly fortuitous for the development of complex life; if he’s right, then rather than advertising Earth’s total eclipses as a wonder of the
galaxy and expecting aliens to flock to our otherwise modest
Solar System to enjoy it, we should expect alien life to grow up
only on worlds where total eclipses are common. Rather than an
attraction, our total eclipses would be a signpost pointing extraterrestrial astrobiologists in our direction.) This coincidence means that when, as happens once or twice
a year at most, the Moon crosses in front of the Sun as seen from
Earth, we experience not only the temporary extinction of the
Sun but also get to see its beautiful, tenuous pearly white outer
atmosphere, the corona. The view is spectacular—and everyone
should try and see at least one total eclipse in their lifetime—but
the downside of the close parity between Moon and Sun is that
the shadow of the Moon makes only a narrow track on the surface of the Earth. If you’re not under the shadow, then all you receive is a partial eclipse—and even a 99 per cent partial is a pale shadow of the experience of totality itself. The narrowness of the
track makes a total eclipse one of the rarest of phenomena from
the perspective of any particular place; if you stay put, you would
be lucky to get one total eclipse in a millennium. The British Isles, therefore, views totality only rarely. The last total eclipse to cross
80 No Such ThiNg aS a New ide a
the British mainland happened in 1999, when the track covered
Cornwall and South Devon including my then home in Torbay,
where it was cloudy, and the next opportunity for British observers won’t arrive until 2090.*
Halley was well aware of the rarity of such an eclipse, and he
thus viewed the British eclipse of 1715—the first for 500 years—
as something of an opportunity. Because the geometry of the
eclipse track depends precisely on the positions of the Moon and
the Sun, recording the track’s dimensions to great accuracy
would, he reasoned, allow him to make accurate measurements
of the scale of the Solar System. Making such a calculation
requires records of the beginning and the end of the various
phases of the eclipse, and so Halley prepared to make his scientific observations from the slightly incongruous vantage point of a side road just off Fleet Street in the heart of London, the then
headquarters of the Royal Society. He was a good enough
researcher to realize that a single set of observations would be
sadly lacking, and so in addition to his own preparations he wrote
to the professors of astronomy at Oxford and Cambridge to
enlist them in his endeavour.
It was somewhat fortunate that Oxford, Cambridge, and
London, major concentrations of the scientifically aware, were
all predicted to fall within the track of totality. Yet Halley realized that the track crossed a large swathe of the country from east to
west, and that restricting himself to just three sites would be
amazingly shortsighted. He took action, publishing a map of
the track alongside tables of likely timings, distributing it far and wide alongside an appeal for help.
* It’s 23 September. I’ve already marked my calendar, and by coincidence it’s a trip to the West Country again as the track covers Cornwall and Devon. The weather prospects are uncertain.
No Such ThiNg aS a New ide a 81
The map was published as a broadside, a sort of singlesheet
newspaper common at the time, and was a commercial proposition. Halley worked with a publisher, John Senex, who specialized in mapmaking, and they sold for six pence each (Figure 14).*
Trusting in the ability and efforts of a public he referred to in
the instructions he wrote as ‘the curious’, he asked that anyone
who was capable of making suitable observations should send
him timings of the eclipse. The leap of faith required for this
establishment figure to ask for help perhaps shouldn’t surprise
us too much; Halley’s great genius was in handling data, which
he manipulated in order to bring mathematical rigour to his view
of the cosmos. Rather than wanting to rely on a small number of
‘professional’ observations, no matter how reliable, it seems
obvious to me that a great datasmith such as Halley would want
to have on hand all the available information. After all, he could
always decide later how much weight to place on each record he
received.
An eclipse is a oneoff. Whatever is planned, there is only one
chance and a few short minutes in which to execute even the best
laid of plans. Given how things went on the day, it was probably
a good thing that Halley cast his net widely. The Royal Society
party in central London were successful in their attempts to view
and to record the eclipse, as were the observers organized by
John Flamsteed, the first Astronomer Royal, a few miles down
the river at Greenwich, but the university observatories do not
come out well from this story. In Oxford, it was cloudy. In
Cambridge, poor Reverend Cotes, then in charge of the observatory, was blessed with clear skies. He did, though, have ‘the misfortune to be oppressed by too much company’ and was thus
* Not a model we’ve yet considered for more modern crowdsourcing projects, but maybe we should.
Figure 14 Halley’s 1715 eclipse map, with instructions for making scientific measurements of timings.
No Such ThiNg aS a New ide a 83
unable to contribute anything of use. This seems remarkably lilylivered to me; even if we’re charitable and assume that the company in question consisted of crowds of eager eclipseseekers rather than merely of a friend or two calling round for tea, this
distraction seems rather neglectful of the potential for scientific
progress.
These varsity failures turned out
not to matter, though, for
observations flooded in. More than 200 people, as far from
London as Plymouth, sent data to Halley. An especial mention is
necessary for the particularly heroic souls who eschewed the
temptation to head for the centre of the track and a longer eclipse,
but who stayed near the edge of the track bearing witness to either
no totality at all or a total phase which lasted only a few seconds.
A trip of a few miles would have afforded them a more spectacular eclipse, but it was these liminal places that provided the information critical for Halley’s experiment. Written up in the pages of the Royal Society journal, the collated results make for an impressive sight, and certainly the 1715 event represented a leap forward in eclipse observation, but ultimately the experiment didn’t
amount to much. The main result was to confirm that eighteenthcentury astronomers already had a pretty good grasp on where things in the Solar System were, and there is no record that I can
find of anyone bothering to repeat the experiment when, just a
few years later in 1724, a total solar eclipse again darkened parts
of southern England. Negative results are, of course, of great
importance in the progress of science, marking as they do the
roads not taken and the theories not overturned in the search for
scientific progress, but it is hard not to be disappointed that this
wasn’t the start of something more. Rather than standing as spectacular successes which encouraged everyone to reach out to crowds of volunteers, these eighteenthcentury stories of Robins
and Halley are interesting but lead us to no glorious triumph.
84 No Such ThiNg aS a New ide a
A few decades later, what we would now call citizen science
was to trigger an explosion in scientific knowledge, as the age of
the grand amateur provided a kickstart to observational astronomy, geology, meteorology, and more. As each of these subjects became established as datarich and observational sciences, so
they came to rely on distributed networks of observers to provide their raw material. That’s most clear in cases like that of systematic weather observation, where having data gathered from as many different places as possible makes a world of difference.
The very idea of weather forecasting was somewhat controversial during the nineteenth century, a tale told brilliantly by Katharine Anderson in her book Predicting the Weather: Victorians and the Science of Meteorology (seriously—it’s one of the most readable scholarly books I’ve come across in years). In one chapter, she talks about the meteorological efforts of two of my heroes,
both of whom built up networks of thousands of weather recorders
across the British Isles.
The first, James Glaisher, was an astronomer at the Royal
Observatory at Greenwich, back when the site was still supporting cuttingedge research rather than the museum and tourist attraction it is today. Glaisher was a flamboyant polymath with
equal tastes for both adventure and publicity, a combination
which led to him pioneering the art of scientific study from hotair balloons in the 1850s and 1860s. Flight then was still a novelty—
the stuff of fairgrounds and balls, and passengers clutching
champagne—and the distinguished Glaisher who was already, in
his fifties, a scientific person of quite some standing who might
at first have required some persuading to fly himself. We’re told
he only took to the air when he became dissatisfied with the work
being done by the students and technicians he had deputized to
stand in his place. Despite this reluctance, once airborne Glaisher
quickly realized the twin advantages of ballooning; not only did
No Such ThiNg aS a New ide a 85
it provide access for the experimenter to the atmosphere far
above the ground, but it also provided plenty of material sure to
attract a crowd to his public lectures. In the process, his transformation from scholar to dashing adventurer was complete, and adventure he certainly had. His descriptions of scientific
flights to trace the temperature profile of the upper atmosphere
include accounts of episodes in which both he and his pilot all
but lost consciousness, becoming ‘insensible’ and struggling to
read their instruments (Figure 15). Even at lower altitudes, the
chance of drifting irretrievably out to sea caused constant worry
and conflict between the scientist, always persuading himself to
take just one more measurement, and the pilot who would have
been understandably concentrating on getting back on the
ground in one piece.
The balloon used for these ascents was owned by Glaisher’s
regular pilot, a man named Henry Coxwell, and it came to a surprising end. In 1864, it was ripped apart by a rioting crowd that had gathered in Leicester for a demonstration flight that never
happened. Somewhat ironically, it was the sheer size of the
crowd that prevented a safe ascent being attempted, but the incident put an end to airborne experiments. The overlap between public display and scientific experiment goes back to the beginnings of Glaisher’s scientific career. He had, since 1844, been producing weather statistics as part of an ongoing study into possible connections between disease and the weather. As well as publicizing the results of his investigations in the press, he actively collaborated with the newspapers in order to gather in observations from further afield. London’s Daily News, for example, helped arrange for station masters to make daily weather reports
that could be carried back to the capital and printed in the paper
for general edification as well as for scientific use. Others contributed observations too, and contemporary reports are careful
Figure 15 Coxwell, the pilot, climbs into the balloon rigging as Glaisher lies ‘insensible’. Nineteenthcentury meteorology looks dangerous.
No Such ThiNg aS a New ide a 87
to explain that the ranks of Glaisher’s observers were swelled not
only with ‘zealous meteorologists’ but also with ‘the servants
and gardeners of landed gentlemen and noblemen’.
The snobbery of that division is striking, at least when we look
down from the comfortable heights of a twentyfirstcentury perspective. All may of course contribute, but only those who do so backed by real means can be meteorologists; others contribute
data which can be used in the common cause. This division could
be the result of blinkered stereotyping, or it might reflect the
simple practicalities of nineteenthcentury scientific life, an age
the astronomical historian Allan Chapman has called the ‘age of
the grand amateur’, a time when those with the resources to
achieve leisure invested it in serious pursuits. There are also signs of a hierarchical view of science: servants and gardeners can make
observations but that work isn’t really meteorology, which
involves analysing the results. Whichever it is, it’s amazing to me
how clearly these thoughts reflect an argument about what it
means to ‘do science’ that persists right up to the present day.
Once Galaxy Zoo had become an overnight success, we found
ourselves casting about for words with which to describe it. Plenty
of options were available: crowdsourcing, which I used for Robins’
efforts above, was coined by an editor at Wired magazine, Jeff Howe, as a portmanteau derived from ‘outsourcing to the crowd’.
Howe’s definition draws on an analogy with ‘outsourcing’, which
is what a company does when it replaces
employees who would
otherwise be engaged in a particular task with a call to outsiders
to complete the task for them. Crowdsourcing would then be
what happens when a company or organization asks the world to
complete a task for them which would otherwise be handled by
employees; a good example might be the recent call by Transport
for London, which runs the city’s deep tube lines, for ideas as to
how to install air conditioning in cramped tunnels.
88 No Such ThiNg aS a New ide a
As with the meteorology example, when you use ‘crowdsourcing’ it seem that you’re making a distinction between those who direct the operation and those who provide the work. Even in the
early days, it was clear that Galaxy Zoo seemed to be different;
we didn’t want to replace Kevin’s work with that of the public,
but rather to extend the scope of the investigation beyond what
could be done by professional astronomers. In any case ‘crowdsourcing’ seemed a little unambitious, and I found myself leaning towards a more aspirational term—‘citizen science’—as something that was more inclusive.*
Rather than placing myself on a pedestal, viewing my position
as Eminent Scientist as apart from the crowd—someone who
assigns tasks and reserves only for themselves the right to analyse the results—it instinctively seems to me that it is impossible to draw a clear line between where the supposedly menial tasks
of data gathering, classification, and exploration stop and some
sort of Proper Science starts. It’s all just science. Glaisher clearly took much of the credit for the joint enterprise which involved
information gathered by observers waiting down the track at
remote railway stations, but the value of the whole enterprise is
none the less a collective one.
This argument about status and contribution is also visible in
the story of the other character that stands out in Anderson’s tale
of Victorian weather observers. While Glaisher used his status
and, somewhat to the distress of his bosses, the name of the
Royal Observatory to promote his work, George James Symons
* This term has its problems. The prominence of immigration in political debate, particularly in the US, is a reminder that not everyone is a ‘citizen’, so