I normally post on my blog every Friday but the last Friday was a holiday here in Catalonia so I decided to take a break. It was also my birthday week so I wanted to take the time to reflect too on my life and the things I want to achieve moving forward. Anyhow, this are my top reads of the last week:
A hypothesis is a liability – compares day science (where ideas are tested) with night science (where ideas are generated). When we focus on testing a hypothesis, we may be blinded from exploring other important facets of the data. I especially liked the experiment where they asked students to analyze a certain dataset. If you simply plot the dataset, you will see a gorilla. However, most students miss this as they are too focused on running the regressions, without adequately conducting data exploration.
Learning from nature – Biomimicry innovation to support infrastructure sustainability and resilience – fascinating paper on how innovations in the natural world can be applied in developing infrastructures. The tables showing examples of biomimicry in different engineering products, processes and systems are really interesting.
What’s Wrong with Social Science and How to Fix It: Reflections After Reading 2578 Papers – a blog post by someone who skimmed tons of social science papers to participate in a replication market. A replication market is a market where one can predict whether a paper will replicate or not. Among the different fields, economics had the highest expectations of replication. Management and marketing ranked the lowest (honestly, unsurprisingly).
Illuminating the dark spaces of healthcare with ambient intelligence – a review in Nature on how sensors and artificial intelligence can improve the health of individuals in both hospitals (ICUs, operating rooms and even lobbies) and daily living settings (elderly homes, chronic disease management and mental health monitoring).
From Necessity to Opportunity: Scaling Bricolage across Resource-Constrained Environments – bricolage refers to working with what is available. Normally, this concept is applied to startups which may not have access to many resources. In this work, the researchers explore how a social enterprise was able to scale to different countries with bricolage.
Coevolution of cyberinfrastructure development and scientific progress – explored how developing cyberinfrastructure can impact progress in research, specifically in the field of biodiversity and ecology. The researchers show that the coevolution between this data infrastructure and the field itself led to evolutionary progress in terms of increasing both collaborations and democratization in this research community. Moreover, it also led to revolutionary progress in terms of connecting this field with other scientific disciplines.