Learning scientific writing from great writers

For scientists, writing well (or well enough) is a critical skill, as written texts are essential for communicating research. Of course, not every scientist should be able to write well, as some may rely on collaborators. In a lecture on "Writing physics", David Mermin emphasizes the importance of language and writing through a famous example:

It is also said that even Landau's profound technical papers were actually written by Lifshitz. Many physicists look down on Lifshitz: Landau did the physics, Lifshitz wrote it up. I don't believe that for a minute. If Evgenii Lifshitz really wrote the amazing papers of Landau, he was doing physics of the highest order. Landau was not so much a coauthor, as a natural phenomenon — an important component of the remarkable coherence of the physical world that Lifshitz wrote about so powerfully in the papers of Landau.

SciPost two years on

There are two reasons why we should care about the journal SciPost Physics: it practices open peer review, by publishing the exchanges between authors and reviewers, and it is free to authors and readers, at a time when academia is struggling to escape the stranglehold of predatory publishers such as Elsevier.

Two years after it was launched, SciPost Physics is alive on well, with 155 published articles at the time of this writing. The journal has managed to attract articles of high quality, some of them by well-known physicists such as Cardy, Verlinde, Seiberg, or Rychkov. The main challenges are now to attract many more authors, and sustainable funding. The journal is funded by institutional subsidies, an economic model that works for arXiv but that is rare for journals. If the finances worked out, the journal could ultimately hope to become a megajournal, and publish thousands of articles a year like PLOS One or PeerJ. For the moment, the stated aim of the “Expansion and Sustainability Drive” is to publish 500 articles a year at a cost of 200.000 euros. So each published article should cost 400 euros, a very reasonable price for a lean electronic journal: an order of magnitude more than arXiv, an order of magnitude less than Elsevier. (Unlike Elsevier, SciPost does not pay executives, lobbyists, shareholders, salespersons, lawyers, etc.)

In this post I will comment on SciPost’s workflow and platform, updating my earlier assessment in light of two years’ worth of experience. The aims are to help authors decide whether this journal suits them, to determine which features of SciPost should be emulated (or avoided) by other journals, and to provide feedback to SciPost itself.

Will no one rid me of these tiresome Latin plurals?

The English language has inherited many scientifc words from Latin: a spectrum, an index, a torus, a formula. Then which plural forms should we use: the Latin plurals two spectra, two indices, two tori, two formulae? Or the English plurals two spectrums, two indexes, two toruses, two formulas? The Latin and the English plurals of these words are both considered correct, but the Latin plurals are more widespread. I will nevertheless argue that using Latin plurals is impractical and illogical, and should often be avoided.

Write for humans, not for robots

The recent San Francisco Declaration On Research Assessment (DORA) aims at improving how scientific research is evaluated. To do this, the declaration wants not only to improve the evaluation process itself, but also to modify how research is reported, in particular by having authors of scientific articles cite original articles in preference to review articles. But this would be bad scientific practice: researchers should not have to worry about the reliability of bibliometrics when they do research or write articles.

As a method of evaluating research, bibliometrics has many well-known shortcomings. The DORA denounces some of these shortcomings, and proposes remedies. Some of these remedies are common-sense ideas, such as stopping using journal impact factors, and evaluating the contents of articles (and other research outputs) rather than relying on bibliometrics.
Some other proposals aim at improving bibliometrics. It is not obvious that this would be good, because bibliometrics will always encourage bad practices, such as trading citations and authorships, and splitting results into small articles. Eliminating some of the flaws at the expense of making metrics more complicated may not be worth the trouble.

And the proposals 10 and 16, which want authors of scientific articles to cite  original articles in preference to review articles, are downright pernicious. The problem is no longer to improve bibliometrics itself, but to modify how articles are written, with the aim of making bibliometrics more reliable. So the DORA claims that bibliometrics is so flawed that its use should be much reduced, but that it is so important that great efforts should be made to improve it -- efforts not limited to modifying the metrics themselves. In this sense, the DORA is self-contradictory. 

This might not matter too much, if citing original articles was good scientific practice. But it is not. The original literature is a tangled mess of more or less reliable and understandable texts. Review articles play the vital roles of making sense of it, and of promoting common, generally accepted terminology and ideas. Writing review articles is in some cases more useful than doing original research, and should be encouraged. Moreover, researchers often do not learn of existing results from original articles, but from other texts, which may be clearer or more accessible than the original articles. Citing the original articles would often mean citing articles without having read them, which is obviously bad practice.

Remember why articles cite earlier works in the first place: to avoid repeating material which is available elsewhere, and to help readers find the origins and proofs of results which are built upon. To do this efficiently, one should cite as few articles as necessary, and select them based on clarity, reliability and ease of access. (Open access should surely be favoured.) There is no reason to favour original articles. Favouring original articles serves another purpose: as the DORA puts it, to "give credit where credit is due". But the purpose of research articles is not to give credit. An article's contribution to the history of its subject is only a byproduct, and should not take precedence over its primary purpose of reporting scientific ideas.

Bibliometrics was supposed to help with the task of giving credit, but now we are told that it will not work unless we think more about bibliometrics than about readers when we write articles. This should be resisted: write for humans, not for robots.

This blog's initial intentions

This blog is for discussing practices in scientific research as they are and as they should be, and the tools which can help doing research efficiently.

The focus is not to describe the practices which researchers should follow for the sake of their careers, but rather to discuss what should be done in order to do good and useful research. As a theoretical physicist, my definition of useful research will be research whose results are easily accessed, understood and discussed -- so scientific publishing will be a major topic.

It has been obvious for a long time that scientific journals could be replaced with a much more efficient and cheap system. But opinions vary widely as to whether and when this can happen: from the prediction of the imminent doomsday of commercial science publishing, to despair at the slow progress of open access. It is also not clear how a new system could emerge from the current situation: evolutionarily from existing journals? as an Arxiv overlay such as the Selected Papers Network? or as a new creation such as the journal PeerJ?

In some cases, there is a consensus on what the good and bad practices are, and the question is how to switch from the latter to the former: how do we escape rapacious scientific publishers? how do we stop using impact factors and the h-index? In other cases, the identification of good and bad practices is less clear: is it good practice to use Mathematica? should researchers contribute to Wikipedia? should publicly-employed scientists put their writings in the public domain?  

Some recommended reading:

• Gowers's Weblog.
• Two texts of mine on scientific publishing (in French): Journaux scientifiques: échapper aux abonnements ruineux and Les publications scientifiques au CEA : quelques questions.
• Nielsen's book "Reinventing Discovery" -- Citing it is admittedly bad practice as the text is not freely accessible, but this book is a compelling explanation of how research practices could be radically improved.