'So much in the current system discourages best research practice'

Tell us a little about your background and qualifications...

The theme running through my odd-seeming career is communication. I grew up in awe of great communicators of biology like Gerald Durrell, David Attenborough, Richard Dawkins. I applied to study with Richard Dawkins and did my doctorate on communication and evolution in butterflies, knowing that I wanted my next step to be working in television.

It was. I was David Attenborough’s researcher on Life in the Undergrowth, and spent 17 years making documentaries. Then I saw a tweet from Prof David Spiegelhalter, whom I had previous directed when he was presenting a documentary. He wanted someone to lead a new centre for ‘evidence communication’ at the University of Cambridge.

I didn’t know what evidence communication was, but found myself leading an academic research team – and coming face to face with the scholarly publishing system!

Can you explain briefly how the Octopus model works?

Very briefly, Octopus, funded by Research England and supported by Jisc, allows researchers to publish in smaller units. So, instead of putting a lot of resources into finding a team to collaborate on a project that can be published in a traditional ‘paper’, then dedicating time to writing it up, followed by months of peer review and rewriting, researchers can be much more agile. They can publish idea, protocols, datasets, analyses etc, in a co-ordinated and connected way, and for free.

The idea is to minimise the costs of publishing to help get more work shared (and therefore used), and also to ensure that good work is rewarded – recognising that there are many different forms that research takes these days (such as writing code, designing protocols). It is hopefully more meritocratic and will produce more a robust research base.

We first interviewed you in 2020, in the early days of Octopus. How has progress been since then?

Well, Octopus.ac is now a fully functioning platform, and over the last year we’ve had about 50 new people a month coming on board to use it. It’s amazing to see people engaging with the site, which has been ‘just an idea’ for so long, and researchers are mostly very enthusiastic.

You recently talked of your ‘heartbreak’ over how researchers feel about the current publishing system/culture. Could you expand on that?

Although researchers are usually really sold on the idea of Octopus – they see that it is a much better way of doing things than the current system – many don’t feel they have the agency to change to using it. When the University of Bristol interviewed people to find out how they felt about their work and the current system, we heard such terrible stories. People felt that they had to ‘cheat’ and commit ‘questionable research practices’ in order to get publications. The ‘novelty’ of their work and their institution’s reputation were seen as critical to them being published and being able to continue to work as researchers.

It seems that so much in the current system discourages best research practice. Those that want to do things in the best way possible have to fight, to put in a lot of extra (unrewarded) work, to pay fees to make work open and so on. That causes inequities, because you have to be in a privileged position to do all that - such as having a secure job, secure funding, being at a well-resourced institution, or already having an academic reputation and contacts.

How does the Octopus system specifically alleviate researchers’ problems in this regard? And how does it benefit the advancement of science on a wider scale?

Octopus is designed to tackle a lot of those inequalities and reset the incentive structure to ensure that people get rewarded for good work and not for anything else. In the long run, it will change the culture to one that is more collaborative than competitive, and where open sharing and research best practices become the norm. This will mean that knowledge will advance more quickly, and findings will be based on more robust methods and data.

But, in the short and medium term, we have to ensure that researchers get a benefit from using Octopus alongside the traditional publishing system which still controls their careers. Part of that is down to us to make it easy to use and useful, but part of that is also down to institutions and funders, who control researchers’ career prospects, to recognise and reward those who are demonstrating best practice within Octopus (such as those sharing data and code, pre-registering their hypotheses or methods, or peer reviewing).

Lastly, looking forward, how would you like to see the scholarly communications ecosystem develop over the next few years?

I’d like to see a move to recognising that a pdf of words and graphs is not the best way to share, or to assess the quality of, a lot of contemporary research work. We should think about the vast range of skills and roles that we need in our research community, how best to reward people for developing them, and how best to bring them together to work on research problems.

The digital tools at our disposal are incredibly powerful – people can work together across space and time, communicate despite language differences, and AI can help do what would have taken humans inordinate amounts of time. But, if researchers are not getting rewarded for sharing digital assets or communicating their real experience, this will all be lost and we’ll be condemned to repeat mistakes and learn from scratch over and over.

Researchers will continue to be unhappy and feel they cannot ‘afford’ to do things ‘right’.