All You Need ... is All You Need

The most important paper of all time came out in June of 1951. It changed science. And science is about to change again.

All You Need ... is All You Need

Before getting started, two things:

1/ Supporters get full archive access and premium notes.

2/ Know someone who might like to read this? Share it with them via the links at the bottom of the newsletter.

Here is something I ve been thinking about. It is an essay about the nature of progress, and science as an evolving and social human endeavor.

As usual, the footnotes are first and free, and the essay is (after the first few paragraphs) for premium subscribers only. I hope you enjoy either or both and it gets you thinking.


  1. Lowry, Oliver H., et al. "Protein measurement with the folin phenol reagent." Journal of Biological Chemistry 193.1 (1951): 265-275.

The paper "Protein measurement with the Folin phenol reagent" by Lowry et al. is a foundational scientific work that introduced a new method for measuring the amount of protein in a sample. Before this paper, accurately measuring proteins was challenging and lacked a reliable, simple technique. The Lowry method, as it came to be known, uses a chemical reaction that changes color in the presence of proteins. This color change can be measured using a spectrophotometer, an instrument that detects the intensity of colors, to determine the protein concentration.

  1. Vaswani, Ashish, et al. "Attention is all you need." Advances in Neural Information Processing Systems 30 (2017).

The 2017 paper introduces the Transformer model, revolutionizing how machines understand and generate human language. Before this, machine learning for language tasks relied heavily on complex models that processed data sequentially, which was time-consuming and inefficient for capturing the nuances of language. The Transformer model changed that by using a mechanism called "attention," allowing the model to focus on different parts of the input data simultaneously. This approach improved the efficiency and effectiveness of machine learning models in translating languages, understanding questions, and generating text. It set a new standard for natural language processing (NLP) tasks, leading to rapid advancements in AI's ability to interact with human language, from better translation services to more responsive chatbots.

  1. Simkin, Mikhail V., and Vwani P. Roychowdhury. "Read before you cite!" Complex Systems 14 (2003): 269-274.

The paper "Read before you cite!" examines an intriguing phenomenon in scholarly research: many scientists cite papers they haven't fully read. Through statistical analysis, the authors estimate that only about 20% of citations are based on a thorough reading of the cited work. This observation suggests a certain level of superficiality in how research papers are referenced, potentially leading to the spread of misinformation or a misunderstanding of scientific findings. The study highlights the importance of critical and careful citation practices in academia, urging researchers to ensure they truly understand and have read the works they cite. This work sparked discussions on the integrity of academic referencing and encouraged more rigorous standards for scholarly citations.

  1. Raafat, Ramsey M., Nick Chater, and Chris Frith. "Herding in humans." Trends in Cognitive Sciences 13.10 (2009): 420-428.

The paper "Herding in humans" explores how individuals in groups often align their beliefs and behaviors with those of the majority, a phenomenon known as herding. This behavior is observed across various contexts, including financial markets, consumer choices, and even scientific research. The authors discuss psychological and social mechanisms driving herding, such as the natural human tendency to conform to social norms and the reliance on the wisdom of crowds when individual decision-making is uncertain. The study highlights the dual nature of herding: it can lead to the efficient dissemination and adoption of useful behaviors and information but also to the propagation of errors and biases. The findings underscore the importance of understanding herding mechanisms to mitigate its negative impacts, particularly in areas where independent critical thinking is crucial.

  1. Kuhn, Thomas S. The Structure of Scientific Revolutions. 4th ed., University of Chicago Press, 2012.

Kuhn's classic "The Structure of Scientific Revolutions" argues that scientific progress is marked by paradigm shifts rather than linear knowledge accumulation. Science operates under a prevailing paradigm, which dictates the norms for legitimate research. Anomalies that the current paradigm can't solve steadily accumulate, leading to a crisis and eventually a paradigm shift. This shift redefines the field's fundamental principles, a process met with resistance due to the deep entrenchment of paradigms in scientific practices. Kuhn's theory, emphasizing revolutionary rather than evolutionary progress, as well as the sociology of science, has significantly influenced the philosophy of science.


The Most Important Paper in the World

The most important paper of all time came out in June of 1951. Called "Protein measurement with the folin phenol reagent", it introduced the Lowry method for determining the concentration of proteins in a solution. Now a standard laboratory technique, it is widely adopted—and the paper is dull as dirt.

Lowry isn't just dull because it is lousy reading, although it isn't much fun to read. Here, to give you a flavor, is the opening paragraph, which comes at you old school, without any of that Gen Z academic "Hey, love me, here is what this is all about" abstract or any such coddling.

Even a biochemist won't find it very exciting. The paper is Ikea-instructions-level material on a particular lab technique. While useful, it is hardly the stuff that research dreams are made of.

So, why is the Lowry, et al paper so important? It is important because it is, at present, the most-cited academic paper of all time, with more than 300,000 citations in other books and papers. And why is it so highly cited? Because of the Lowry method's usefulness and wide adoption in biochemistry and molecular biology, making it part of the white noise of wet lab work, thus requiring a citation in any paper based on experiments using it. And there are a lot of wet lab papers in these and related fields.

A refresher: Being widely cited is important in academia. There are at least five reasons, only one-and-a-half-of-which of which is gross and careerist:

  1. Impact and Influence: Citations measure the impact and influence of research work. All else equal, a more cited paper is a sign that it is recognized and used by other researchers in the field.
  2. Research Quality: Highly cited papers tend to be ones with high perceived quality. Granted, many poorly cited papers are well done, but rarer yet to find a highly cited paper that is complete crap.
  3. Academic Careers: Citations are part of most academic evaluations as they try to obtain tenure, promotions, and funding. High citation counts suggest your field thinks what you are doing is, for want of a better word, good.
  4. Visibility: Think of citations like Google links in the world of early Google. More citations suggest higher quality, which makes you more visible in search. Similarly, more citations mean a quality paper is more likely to be discovered by other researchers, increasing the visibility of the research and the authors.
  5. Network and Collaboration Opportunities: The rich get richer, and the most highly cited researchers tend to attract more collaborators and coauthors and conference invitations. It feeds itself.

There is a New Paper on the Block

But it is all about to change. And not because citations are less important, even if their role in science is evolving. Citations are changing, but they are still very important. The change is that sometime in the next few years (at the current rate of citation growth) a quirky 2017 paper will pass Lowry by and become the most referenced in the history of science.

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to Paul Kedrosky.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.