I have recently completed a part-time PhD with Southampton University investigating note-taking behaviour of scientists in the lab and the design of tools to help them capture and make use of their experiment information. I have been conducting this research as part of the Chemistry department in Southampton, with Professor Jeremy Frey. The research is multi-disciplinary with elements of science, e-science, computing, usability, and psychology. There are a number of themes within the research as discussed below, but all of them come together to help inform the design of tools to support scientific researchers, in particular electronic lab notebooks and digital research notebooks. The research has also involved investigating user behaviour using traditional note-taking methods and using digital systems such as ELNs and other information capturing platforms such as Flickr, WordPress, and MyExperiment.
Note-taking behaviour in lab scientists and user research activities
In order to design tools that support and facilitate scientists in their research it is necessary to understand how they work now, both using paper-based and electronic means to design and enact their experiments, to gather and analyse their data, and how they share their work with collaborators and the scientific community. It is also valuable to observe how researchers use existing tools to understand the strengths and weaknesses of these existing platforms, and how they might be improved to better serve their users. A variety of user research methods were used to investigate not just how scientists record their experiments, but also the whole process of how they conduct their day to day research and the individual challenges that they face, including semi-formal interviews, observations, and user studies. We also investigated what researchers felt were the main reasons for note-taking and who they were for. The results of these activities have then been used to produce personas, user stories, and generate user requirements that can be used to contribute to the design and development of tools to support scientific researchers in the future. A variety of unexpected requirements were generated as a result of these activities that indicates that the needs of researchers extend beyond the capture and management of experiment information.
Designing interfaces for recording experiments in the lab
Although there are many perceived advantages to using paper-notebooks – for example, they are easy to use, very portable, flexible, don’t need batteries or network connections – there are disadvantages compared to digital systems, such as being difficult to share, difficult to ‘back-up’ and keep the contents secure, relatively fragile, and – particularly in these days of digital instruments – often separated from the data generated during an experiment. ELN systems can help to overcome these issues as well as provide additional functionality such as automatic generation of table of contents and provide searching and linking capabilities that are impractical in a paper notebook. The design of the interface for an ELN or other experiment capture tool is likely to have an impact on the information that is captured by the researcher. ELNs typically provide either a template for an experiment to complete, or a more ‘free text’ style of interface where the user can record whatever they want to. What is not known is if using a template will significantly change the information that is recorded compared to a ‘blank’ page. A number of studies were conducted with undergraduate chemistry students and other researchers to investigate whether using templates changed the information that was captured for the write-up of experiments, and whether changing the design of the template had an impact on the information that was recorded. We found that using templates did in fact change the information that was recorded by the students and researchers with the questions/headings in the template acting as cues for responses. The effects of using templates were both negative and positive, with some findings matching our expectations, but others rather unexpected. The results of these studies have important consequences for the design of interfaces for tools for supporting scientists to ensure that the most useful information is captured for the experiment record. The results of these studies have been published in this paper:
- Effects of Using Structured Templates for Recalling Chemistry Experiments. Willoughby, C., Logothetis, T., & Frey J.G. (2016) Journal of Cheminformatics, 8:9. DOI: 10.1186/s13321-016-0118-6
Generating and capturing metadata for experiments
Metadata is often considered to be important in the curation and deposition of materials for archiving purposes, but metadata can be extremely valuable within electronic systems enabling improved access, search, and organisation of information within a system, which can then be used to provide valuable functionality for the authors and users of the information. In our initial interviews and studies with users we found that use of metadata within our ELN was perceived as an area of difficulty and confusion. To investigate this further we analysed the metadata from over 100 notebooks from our ELN to examine how metadata was being used within the ELN, whether it was used effectively, and how the interface might be improved to encourage the creation of more appropriate metadata. The metadata from the ELN was also compared with the metadata from other platforms that enable the creation of user-defined metadata including Flickr, NASA and chemistry-related blogs, and the scientific workflow sharing tool, MyExperiment. Similarities were found between the different platforms, but some of the interface designs from the other platforms could enhance the capture of metadata within ELNs. A number of studies were conducted, based in part on the template studies, to find out whether using cues or asking the user to change their perspective would change and perhaps improve the metadata that was captured over the more traditional method of tagging – or asking the users to record words or phrases that come to mind when they think about information or their experiments. We found that using these different methods generated significantly different information that could be used as metadata, with again positive and negative benefits for the different mechanisms, suggesting that a combination approach may be useful. Many of the results for the studies described in this section have been published in these two papers:
- User-Defined Metadata: Using Cues and Changing Perspectives. Cerys Willoughby, Colin Bird, and Jeremy Frey. International Journal of Digital Curation. 2015 Vol. 10, No. 1, pp. 18-47 doi:10.2218/ijdc.v10i1.343
- Creating Context for the Experiment Record: User-Defined Metadata: Investigations into Metadata Usage in the LabTrove ELN. Cerys Willoughby, Colin Leonard Bird, Simon J. Coles, and Jeremy Graham Frey. Journal of Chemical Information and Modeling. 2014 DOI: 10.1021/ci500469f
The role of mobile ELNs
With ease of use being one of the main barriers to ELN adoption within academic environments, mobile devices are likely to have an important role in enabling the transition and acceptance of ELNs by researchers. More and more students and researchers have access to mobile devices in the form of phones, tablets, and laptops, together with their familiar and easier to use interfaces compared to desktop computers. Other properties of mobile devices such as long battery life, portability, and access to data networks, make them easier to carry in and out of the lab, and access experiment information and results in the locations they are required. As part of this research we developed a proof of concept mobile ELN for the iPad called Notelus that was designed specifically for synthetic chemists based on our user research. As well as the app enabling the capture of sketches, notes, and text, users can capture photographs and sound, and also search for chemical compounds from ChemSpider. Researchers can optionally create a plan for their experiment that can be imported into the app to help guide them through the experiment and provide information about the materials, equipment, measures, and steps in the experiment. Researchers can then share their experiment via email or archive the notes and associated resources into the LabTrove ELN, including capturing useful metadata that can be used for search and navigation within the ELN. One of the features of Notelus is that it provides a notebook like interface metaphor with pages that can be turned and flipped, with the aim of providing a bridge between the paper notebook and electronic form. The research related to mobile ELNs is currently being preparing for publication.
The future of digital tools for supporting scientific research and Smart environments for research
The adoption of ELNs and other digital tools in academic environments will only happen when those tools meet the needs of those users and are easy to use or at least provide enough value that is in the interests of researchers to spend the time learning to use them. Interface designs are also critically important to the ease of use of digital systems and user research can provide insights into the needs of users and how tools can most effectively fit within their workflows and provide additional value. The tools that are needed, however, extend in scope beyond digital replacements for paper-notebooks. Tools are needed for managing the entire experiment lifecycle, for collaboration between researchers and stakeholders, and for providing additional value through the creation of wisdom by enabling the data and information produced by the researchers to accessible and transformable in a variety of ways. Although there will always be a need to capture the specific thoughts and ideas of the researchers in their own words, smart digital systems – such as lab and even researcher augmentation – can provide support to capture other information automatically that may prove to be useful for interpreting information or finding patterns that extend beyond what was ever possible through the use of paper notebooks and instruments alone.