It’s a privilege and honour to edit a highly regarded biomedical journal. Though it’s considerable work, you learn a lot from reading the submitted typescripts. You also learn how to distinguish good writing from bad. Taking advice from George Orwell we consider these points in evaluating a typescript: (1) What is the author trying to say? (2) Does he/she express it clearly and succinctly? (3) Which edits will make their typescript clearer? and (4) Is their typescript new enough to have an effect?

Over our 12 years at Leukemia we have learned a lot. The quality of submitted typescripts has improved substantially, perhaps aided recently by artificial intelligence, but this is not the whole story. We have also seen writing that drives us crazy. To help future authors, we discuss a few.

Use of jargon: Language peculiar to a particular trade, profession or group of people. Other definitions include unintelligible or meaningless writing, gibberish, writing that one does not understand and language characterized by uncommon or pretentious vocabulary and convoluted syntax, often vague in meaning.

Some common examples. Overall survival. Why overall? Doesn’t survival have the same meaning? Is there partial survival? We suspect authors use overall survival so as to have a convenient abbreviation like others: progression-free and relapse-free survivals (PFS and RFS). “S” alone will simply not do. How about: A total of x subjects? What does a total of add? Another is hypo-methylating agents. Why agents; aren’t these drugs? How about stem cell transplants? Given we cannot accurately identify haematopoietic stem cells in humans isn’t haematopoietic cell transplants more accurate? How about peripheral blood? Is there a central blood? Another pervasive jargon term is novel. To some authors, everything is novel, including drugs approved 20 years ago. Won’t new therapies suffice when the intervention is really new? Lastly, pediatric patients seem very much like children. There are many more examples we discuss elsewhere [1, 2].

The abbreviations curse: The main purpose of abbreviations is to save time and space by shortening words or phrases, avoiding repetition, making text easier to read, and making complex or long names read and sound better. However, we find the converse is often true. When an abbreviation is first used, it needs defining, and when the word or phrase is used only once in a typescript, there is no role for an abbreviation. (There are some exceptions like US, UK, L and km.) Midway through a typescript, readers struggle to recall what an abbreviation means. Some abbreviations, like OS are jargon (see above). Abbreviations should not be in the typescript title and need defining in legends to tables and figures.

Death by percentages and a thousand decimal points: First, percentages should only be used for sample sizes ≥ 100. We don’t need a percentage to understand 7 of 21 subjects. Compounding this abuse, authors with small sample sizes seem to think adding a decimal makes their point-estimate seem more precise. For example, consider a response in 7 of 22 subjects in a clinical trial where the authors write 31.8%. Crazy. The 95 percent Confidence Interval (CI) is huge, 14-55%. Does the 0.8% help with this CI? Another abuse of decimals is in indicating P-values. An example, P = 0.0496. The convention for P-values is 2 decimals unless <0.001; hence, P = 0.05.

Lost in a 1000 words...Introduction, Results and Discussion:  The Introduction serves to introduce the topic of the study, not to the place the study in the 5,000-year history of medicine. Results should not repeat details which should have been in the Materials and Methods. The Discussion should not repeat the Results which immediately preceded it. The purpose is to discuss what the results mean, to compare them with what others have reported and explain any discordances. Limitations of the study and conclusions should be stated (not left to the reviewers). Remember, people want to know the time, not how to make a clock. One would do wise to heed the advice of Blaise Pascal: I would have written a shorter letter but I didn’t have time.

The Associate Editor and reviewers are giving you a huge, mostly unrewarded gift, helping you improve your typescript, avoid errors and identify limitations. The process is complementary, not adversarial. Carefully consider what reviewers suggest and, when appropriate, make recommended edits. An email thank you is more often appropriate compared with a death threat.

Never say: To our knowledge ours is the first study ….. First, to our knowledge, is meaningless; perhaps you were sleeping under a rock. How can we know how diligently you searched for precedents or even what you consider a precedent? Second, claiming priority is a bad idea; someone was always there or thereabout before you. You can only pray it’s not one of the reviewers.

What’s in a name? (Romeo and Juliet; Act 2, Scene 2): How to identify participants in a research study. Often we see patients. Patient derives from the Latin verb patior, meaning to suffer or to endure. It also implies someone under the care of a physician. However, often people in a research study are not suffering nor under the care of the authors. As such, these people are properly designated subjects.

Death by a thousand figures: Often, each datapoint in Results is supported by a complex figure resulting in huge compound figures mirroring a laboratory book, sometimes with 10–20 sub-panels. These complex figures are not informative for most readers. Figures should be carefully selected and unnecessary ones relegated to the Supplementary Materials. Still worse, these compound figures are used in a 10–15 min oral presentation. Understanding even one such figure would take the entire presentation time, during which the audience stops listening to the speaker, instead trying to decipher the figure. Hopeless task under these conditions.

Who needs the Guide to Authors? (https://www.nature.com/leu/authors-and-referees/gta): The Guide to Authors should be strictly followed in developing and formatting the typescript, including word count, requirement (or not) for an Abstract, number of Tables, Figures and References, as well as the need for statements of Author Contribution, Conflicts-of-Interest, Acknowledgements, Funding and Data Availability. References should be current. Most reviewers like being cited when appropriate. Letters and Correspondence are different from Original Articles. They should be brief and concise. Don’t try to fit everything appropriate for an Original Article into a Letter.

Random [C]apitalization?: Often, seemingly out of the blue, authors capitalize names of diseases, drugs and the like. We cannot discern why. Acute Myeloid Leukemia, Busulfan or Lenalidomide appear suddenly, unannounced. Perhaps authors think that if letters in an abbreviation are capitalized, the words should also be. This is not so.

Klingon? (Language developed by linguist Marc Okrand for Star Wars): There are an estimated 20-25 fluent speakers and 100-200 conversational speakers worldwide.) Accurate communication in science relies on conventions. Units of measure should be used appropriately and are mandatory in Tables and Figures. The International System of Units, the SI (Système international d’unités), the modern form of the metric system, is preferred in scientific communications. HUGO Gene Nomenclature Committee terms should be used for cytogenetic and molecular data (https://www.genenames.org/). Human genes are in capitals and italics, and fusion genes are designated by a double colon as in BCR::ABL1 [3, 4]. Likewise, disease names should follow the 2022 World Health Organization classifications [5, 6].

What author position for generative artificial intelligence (AI), 1st, last or both? The proper role of generative AI in scientific writing is language editing, not reviewing a topic, nor summarizing data nor conclusions. Being an author means accepting responsibility for the content of a typescript. Most computer algorithms are reluctant to accept authorship responsibility.

Interesting. But to whom? Often Results  are spiced up with words like interestingly, remarkably, unexpectedly, paradoxically and the like. These are opinions. What may be interesting to you may be prosaic to others.

Shakespeare again: A typescript or a manuscript? Manuscript is a late 16th century term from medieval Latin manuscriptus, manu (by hand) and scriptus (written; past participle of scriber). Shakespeare’s plays are manuscripts. Most current authors are typing on a computer keyboard, not a folio or lined yellow legal pad, using pencil, pen, an eraser and white-out. Hence, typescript is the correct descriptor.

The Oxford comma heresy: There are sometimes bona fide disagreements in science, for example, the heliocentric model of the universe, which nearly cost Galileo his life. History will judge the Oxford or serial comma. The US Government Printing Office says you should use the Oxford comma; The Economist Style Guide says no. Given the current state of the US Government, we side with The Economist. No typescript is ever rejected from Leukemia because of use of the Oxford comma. We simply delete the little buggers for better aesthetics. Let history be our judge.

Presently, English is the language of science. But who’s English? The King’s English or Don Rodrigo’s (Dare we speak his name)? Churchill commented: British and Americans are two people separated by a common language. In Leukemia, we are agnostic, honouring or honoring both. What we require is consistency within a typescript.

On Being Certain: Believing You Are Right Even When You’re Not [7]. Science is a human endeavor and as such is imprecise and error-prone. This is especially so of medicine. To quote Osler: The science of medicine is uncertainty. Failure to acknowledge uncertainty is a critical flaw which can be overcome, in part, by giving Ranges, CIs (frequentist) or Credibility Limits (Bayesian). A related mistake is confusing association with cause-and-effect. Rarely in medicine can we accurately ascribe causality, even in the context of randomized controlled trials (RCTs) and much less so with other settings [8]. Try avoiding the post hoc, ergo propter hoc fallacy whenever possible.

These are our piques. Some, many, or even most authors and Editors may disagree with us. We are not curmudgeons, always open to new ideas. When Orwell met his adolescent sweetheart, he was standing on his head in a field. When asked why, he replied: You are noticed more if you stand on your head than if you are right way up.

Finally, more advice from Orwell, presumably given when he was the right way up: (1) Never use a metaphor, simile or other figure of speech which you are used to seeing in print; (2) Never use a long word where a short one will do; (3) If it is possible to cut a word out, always cut it out; (4) Never use the passive where you can use the active; (5) Never use a foreign phrase, a scientific word, or a jargon word if you can think of an everyday English equivalent; and (6) Break any of these rules sooner than say anything outright barbarous [9].