Learning on knowledge graph dynamics provides an early warning of impactful research – A paper that got a lot of attention from reddit. The paper aims to predict which papers would be breakthroughs, by aggregating 29 metrics including from author and journal data, paper citations, and network characteristics. Using the method retrospectively, they were able to identify 19/20 seminal biotechnologies from 1980 to 2014. I need to spend more time reading the paper but the first thing that popped into my mind is this joke about economists predicting nine of the last five recessions.
Prestigious European grants might be biased, study suggests – In the study cited in the Nature article, the authors find that “applicants who shared both a home and a host organization with one panellist or more received a grant 40% more often than average.” Not really surprising. Scientists and grant funders are humans in the end.
Waiving IP – I’m teaching a class on life science patents and it’s impossible to talk about patents without discussing the various calls to waive patents for COVID vaccines. I’m also of the opinion that to some extent, patents can be beneficial to society.
Making the hard problem of consciousness easier – the idea of adversarial collaborations is really interesting. It’s when two groups with opposing ideas collaborate to come up with the experiment to test which theory is correct. In this case, it’s two groups having different theories of consciousness. I wouldn’t claim to understand global neuronal workspace theory and integrated information theory though.