Thoughts on Electronic Research Notebooks as a national service for the academic community.

My interest in research documentation began in 2012, when I thought that an institutional transition from paper to digital would be painless, easy, and universally welcomed – a quick win for everyone. In industry, they’ve been using electronic research notebook systems for the last 20yrs – the benefits are clear and proven. But it turns out that the world of academia is quite different and, despite our best efforts, it has been more difficult than we imagined to identify a system that would be appropriate for institutional deployment. I’ve worked and spoken with many researchers, groups and institutions, trying to establish what the challenges are, and how we might overcome them, and I think the following bullet points summarise the current position and explain our paralysis quite well:

• Researchers want to switch;
• Institutions want to switch;
• Researchers are dazzled by specialist, discipline-specific software features, but research groups are becoming increasingly multi-disciplinary, comprised of researchers with a diverse range of individual workflows, interests and preferences;
• Institutions cannot identify a product that will cater for all of their researchers’ interests; nor can they support a multitude of diverse products;
• Everyone is fearful of disruption, and fearful of committing to a service over which they have little control, and from which it will be painful to disengage in future.

I believe that a solution is possible that will overcome the obstacles described above, and also provide many other benefits for the entire academic community.

My proposal is that, instead of searching for a perfect product that does everything for everyone (there will never be such a thing), we should instead try to build a product that perfectly satisfies all of the most basic requirements and benefits of an electronic research notebook. Something that is discipline agnostic – equally and undeniably useful to all researchers, in all departments, in all institutions – musicians, chemists, historians and programmers alike. BUT… we should build it in such a way that optional interfaces, features and tools can be ‘bolted on’ to satisfy individuals and groups who have specialist requirements. This will allow all researchers, in all disciplines, to switch from paper notebooks to an electronic documentation platform easily and with complete confidence. Furthermore, it will allow all researchers to experiment with optional features quickly and easily – musical notation tools or chemical structure drawing tools for example – to create a bespoke documentation workflow that perfectly suits their needs, without having to disengage from the underlying infrastructure – they would be able to toggle alternative interfaces and tools on and off as easily as experimenting with WordPress themes.

Given that this ideal system will be designed to be equally relevant and useful for all disciplines, it seems sensible, to me, for it to be delivered and managed as a national service by an organisation that is part of the academic community – such as JISC in the UK. In my view, this would provide many additional advantages:

• Economy of scale, and reduced local complexity and support;
• Democratised access to resources;
• Consistent experience, processes, and quality for all researchers;
• A common platform, culture and language for all researchers, improving collaboration opportunities;
• A framework for rapid deployment of innovative tools to the entire academic research community;
• Promotion and strengthening of ORCID and its advantages as an institutionally-independent research identifier

Software vendors have a wealth of experience and expertise to offer, and it would be extravagant, inefficient and probably unsuccessful for the academic community to try to reinvent their products within this national documentation platform. I believe that commercial partners have an essential role to play, in the provision of those specialist features and tools that will extend the functionality of the basic service. Their products will inevitably be more lightweight, but the trade-off is that they will be immediately visible and easily accessible to the entire research community, on a trial or subscription basis. With all obstacles and anxieties removed, I believe researchers’ uptake of those well-designed commercial products will be substantially greater than their current customer base – everybody wins.

Although it seems sensible for JISC to deliver and manage this service for the UK academic community, and perhaps to build the infrastructure to support it, the matter of whom should develop the basic software platform is less clear. Perhaps, if JISC designs and controls the specifications, standards and evolution of the platform, then it would be appropriate to outsource the development work. Some time ago it seemed like a good idea to encourage a discussion with Microsoft about creating a research-specific version of OneNote, addressing its (very few) current shortcomings, because it’s almost a suitable product already. But I confess I feel a bit uncomfortable about recent developments that I think reveal Microsoft’s ambition for OneNote to evolve into a much more ‘ecosystem’ product, integrated with its Teams software features. I’m also not sure that Microsoft would want to feel hamstrung by an external organisation’s specifications.

Many challenges will have to be overcome to turn this idea into a reality – one of which is the design of an integrated, national data storage service that will enable persistent links between the documentation and the datasets to which it refers. Clearly this will, in the first instance, have to be built at a scale that would limit usage to very small datasets, but, having proven that data storage as a national service can work, I have no doubt that this storage platform would become a good working foundation for a very much larger project that will address the problem of all institutions (not to mention many departments and research groups) buying, building and maintaining their own data storage infrastructures – twice, for both live and published data – despite all of these institutions sharing the same goals. That’s the subject of another blog post, I think.

If you’ve read this far, I’d be very grateful to hear your own ideas and comments. If you’re able to leave a comment below, please do so; otherwise, or if you prefer, please don’t hesitate to send me an email: it-blog@bio.cam.ac.uk