Annette Thomas, chief executive officer for scientific and academic research at Clarivate Analytics, was keynote speaker at Charleston Conference. Here, we present some highlights from her speech
"Most of us here are information professionals. These days almost every large organisation has an 'information director' or 'chief information officer' or 'head honcho of information'. And they're no longer just the people running the computer network, they're helping to shape organisations and strategies to make the best use of the 21 century's most valuable asset: information.
Of course, universities have always had information experts, though they're usually called 'librarians'. Sometimes it seems as if the rest of the world has been playing catch-up with academia when it comes to appreciating the role of information in our lives.
But this isn't to say that we've got it nailed – we still have a lot to learn. I'd like to talk about ways in which we can make things better, but first I want to describe some of our problems.
Universities' identity crisis
Over the last few decades, universities have enjoyed unprecedented growth. They've risen in number from about 500 worldwide just after the Second World War to more than 10,000 today. But while they should be basking in the glow of this success, many universities feel a sense of crisis. This is brought about by three fundamental questions: Who are universities for? Who should pay for universities? Above all, what are universities for?
Researchers also face their own set of challenges: How should research respond to political hostility? Should research serve the economy or human culture? Why is so much research impossible to reproduce?
Let me dwell for a few minutes on the last one. The popular image of research – at least among researchers themselves – has been of a process that transcends individual human biases by providing a series of checks and balances. But it is becoming increasingly clear that important aspects of the system are broken.
John Ioannidis's famous (or infamous) paper 'Why Most Published Research Findings Are False' showed that, at least in biomedical research, using current approaches and statistical thresholds for significance, we should expect most research findings to be wrong. Not "some" or even "many", but "most" research findings. And this isn't unique to biomedicine or even the natural sciences: in more recent work, Ioannidis and his colleagues have reported comparable problems in economics. And there's no reason to think the problem stops there. If this doesn't scare you then you haven't fully understood it.
Publishers' contribution crisis
Why publish so much that is read by so few? The ever-increasing rates at which papers are published aren't just a sign of higher rates of discovery. They also reflect the fact that researchers and publishers alike have become hooked on a publish-or-perish model that values quantity over quality. That's why we get vast numbers of papers that are referenced – and perhaps even read – by literally no one.
Why only publish 'positive' results? In a way, this is the opposite of the first problem. Though we publish a lot, we are in some respects too selective. By publishing only 'positive' or surprising results, we contribute to publication bias.
How can we be part of the solution? This is the key question. I don't have all the answers, but I'd like to share some thoughts. Over the last 25 years, the web has transformed society – including research. But it was also a product of research, born of the early academic Internet. Research and technology have always been tightly connected, and the web is a wonderful example of that.
It can also serve as a source of inspiration. The explosive success of the web was in large part thanks to three core principles: connectedness, openness, and seamlessness.
In the age of data we have unparalleled opportunities to construct views, maps and pictures of information – nothing in research stands on its own. Every paper cites it predecessors. Every researcher has collaborators. Every new discovery is connected to an array of associated facts. The genius of the web was to realise that knowledge itself is a network.
At its foundations, the web is an open system. The underlying protocols are public goods. Research, too, is founded on the principle of openness – transparency between individuals and organisations. It's about sharing your insights, showing your working and being willing to take blame as well as credit.
The web can seamlessly blend text, images, video, data and code into rich, interactive publications or applications. We're still at the beginning of the evolution from static to interactive content, and we have a lot to learn, but this is the future of knowledge dissemination. We need to become seamless too.
So where does research go next? Ours is an industry built on the vision and daring of people like Eugene Garfield. He was a true pioneer. He and his colleagues brought order to a chaotic world of research information. Now we need to rise to a similar challenge. Cynics might say that today's big innovation opportunities lie in different realms – especially in technology – and we shouldn't have such high expectations.
I disagree. The domain in which we operate is still full of opportunities and good ideas. These can drive progress for many more decades to come. We just need to have the insight to recognise them, the courage to act and the integrity to place the needs of researchers and scholars at the heart of what we do. The rest will take care of itself.