The Elements of Scientific Style | Hacker News Reader

The Elements of Scientific Style

worksinprogress.co

35 pointsgglitch7 comments

This is so painfully true, please, do yourself a favor and read the old papers of your field!

But, it takes courage to imitate our witty ancestors! I've been sneered at by many reviewers for my attempts ("unscientific style"?!). The best I've managed to pass through so far is starting a paper with a quotation from an old Agatha Christie novel: https://doi.org/10.1016/j.ast.2019.05.015 (free copy here: https://core.ac.uk/reader/237464116 or at your local Sci-hub domain).

The paper is delightfully laid out. What fonts are you using in the plots?

Thanks! For the plots it's Arial and for the text in the author version, Charter and Cabin Condensed. Elsevier uses instead the (very expensive) Gulliver font for the text.

I think the author conflates aspects of bad writing (which is rife, not only in academic papers, but everywhere), with the increased use of jargon and acronyms. I think jargon and acronyms serve two important purposes in academic writing. They help with conciseness and precision. For people working in the field, long, elaborate descriptions tend to just be noisy, when a more concise term would convey the same information. This can seem like gatekeeping, but in practice, scientific papers are not written for laymen, but for a (typically) rather small amount of other people working on the same, or closely related fields. Writing for dissemination purposes should be more explicit, and maybe avoid some acronyms and very specific jargon, but for scientific writing, please be short, sweet and to the point!

The cholera example, for me, shows the opposite of what the author is trying to put forward. The second paper conveys the information much more clearly, and in under half the space.

>The cholera example, for me, shows the opposite of what the author is trying to put forward. The second paper conveys the information much more clearly, and in under half the space.

Couldn't agree more with your comment. there's a reason why we do not use flowery language in scientific publications anymore. Science has come a long way since that time. Papers are not intended to be read by laymen, the targeted audience is the "specialists" working on that field and familiar with the jargons and acronyms.

If someone wants their scientific work to be accessible by a broader audience, write a blog post or popular science article.

Fermat's Library has a weekly newsletter wherein they showcase interesting papers. Their SAAS allows for easy annotations in the margins of papers, and they do a great job of adding context and insight to the papers they send out. I don't think I'll ever need their service, but I do enjoy reading the newsletters.

It does seem like one could use an LLM to summarize and explain harder to read abstracts and even content for those not in the field. This would be especially so if it were trained on the scientific library to start with.

This could be automated on a user by user basis avoiding what happened (FTFA) to Distill.

Reply on news.ycombinator.com