Ask HN: Why do tech journals have such poor writing?

Technology is an international undertaking. We should be grateful that English has become the common language. It would be so much harder if we needed to be polyglots in addition to our technical knowledge.

I have read several thousand academic papers in my career. In general I don't find the quality of the writing to be as poor as you suggest. Which makes me question whether you are selecting the papers for technical quality.

My process is to read the abstract, look at the authors' affiliations and the references. Those three items give me a sense of how much weight to give the paper. Then I skim the paper, reading the method, results and conclusion. Only strong papers get a more careful reading. So, yes at that stage I might notice the quality of the actual writing. Understandably some international institutions may not have had access to competent English technical editors.

> Which makes me question whether you are selecting the papers for technical quality.

I’m wondering that too. I don’t usually have issues finding good sources in my usual studies, but I am definitely struggling with this academic cybersecurity stuff! In psychology (my usual discipline at the moment) I have journals, authors/academics, and books I know of and trust.

Here’s an example of a sentence from an article I’ve read today about the Eternal Blue Vulnerability used in the WannaCry Ransomware attack: “Since the discovery of the vulnerability, the cybersecurity industry has taken a number of steps to try the issue from the negative.”

- I can’t understand if that is specific lingo to the cybersecurity industry, or just bad English.

Academic researchers don't use bro-talk, nor hacker jargon. It is a mark of professionalism to communicate in a precise manner and to present reproducible research.

Perhaps you should start your research from papers published in ACM and IEEE journals.

I’m far from an expert on the topic, but when I think of a peer reviewed paper, I assume the review is for the content… is the methodology sound, can it be reproduced, etc… I don’t think they are checking for grammar and sentence structure. That’s not their area of expertise and they probably don’t want to waste their time playing English teacher. If the person who wrote the paper is a non-native speaker, I would expect them to work with a native speaker to proofread and edit, if they care about the paper reading well.

I would expect a 'technical writer' to have reviewed before publication. But then again, to save money, they might just rely on Grammerly.

I have published multiple papers to top CS conferences and never had anyone outside of our lab group review the paper before submission.

Peer reviewers and editors are often also non-native speakers. And because they are almost always volunteers, they may not care that much about language, as they have better things to do. It's enough that they (as experts in the field) can understand the paper.

Most academic journals also don't have actual editors. Their editors are basically managers who facilitate the review process and make decisions on acceptance. Outside prestige journals such as Nature, authors are usually responsible for the final text.

Thanks for that insight. I’m going to have to stop assuming that poorly written articles are not worth the read!

Isn't it more important for academic type authors to have their papers just be published, with higher priority than to be as readable as possible?

Plus for it to really be peer review, you would have to have non-native speakers reviewing the papers of non-native authors :)

Which is probably more like what's happening.

Peer review does not mean that the actual experiment, study or computer simulation was replicated by peers. It's more of a gatekeeper function. For example, if a paper entitled "Password Security Best Practices" suggested that it was OK to use your mother's maiden name as a password, then the peers reviewing the paper would flag it as nonsense and require the author to revise the paper accordingly.

But the point about readability is a good one. Papers are meant to communicate. Poor writing increases the friction readers experience, which diminishes communication.

> some of the writing conventions, sentence structure, grammar, spelling etc is so terrible, I can’t believe it was actually peer reviewed.

The key problem is /peer/.

I.e. people no better :)