Angry Reviewer
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
> Reviewer's suggestions for your text:
> Looks like this text is perfect!
If anyone would like to publish my paper, contact me. Email in bio.
This reminds me of a paper that was published in Nature a while back called “Use Machine Learning to Find Energy Materials”. Scientific writing is so heavily steeped in the passive voice that it was shocking to read a title that was active, even imperative.
The author goes on to write the entire article in the same style, with short, active sentences - subject, verb, period. It felt like they wrote it and proofread it to ensure that every single sentence was as abrupt and pointed as possible, almost as a game. Very interesting writing style all things considered.
I followed the review comments to the letter, starting from the default text. After two iterations I succeeded in a perfect text:
> Although quantity of innovative scientific output increased largely in the past 2021, the quality of academic writing has been declining as rarely. Click the button below to check this text!
[Edit: just to be clear, I didn't mean this in a snarky way! Was just having a bit of fun. I guess it does go to show that automated review comments need a human touch before applying.]
Uh, maybe include the text being reviewed in the results? It's weird to see all the comments and not necessarily have the text in front of you to see what it's talking about. Even better would be for each comment to somehow link to a spot in the text. There could be more than one "years" in line 1, and then how would you know which instance the comment is referring to?
See also this parody of the form, which gives notes on the poem Ozymandias ("Look on my works, ye mighty, and despair") as if it were a submission to an academic journal. https://www.nature.com/articles/268100a0.pdf
Note: this is not AI/GPT-3, just all heuristics surprisingly: https://github.com/anufrievroman/Angry-Reviewer
> The word "significantly" is often significantly misused and vague. It might mean statistically significant or significant to the author. State significance quantitatively, e.g. "increased by 42%". Other alternatives: "substantially, notably"
I think once "substantially" and "notably" become widely used, the goal-posts will shift and those too will lose meaning. I have a strong suspicion some of these standards are just plain fashion.
I would love to see the text you're working on also be on the review page. I also would love my text to persist between going back and forth from the edit screen.
> The passive voice is to be used exclusively in this abstract. Recognized as superior, the indirect method will be foisted upon the unwilling audience which will be assuaged by the copious delirium found in the general laity.
> Reviewer's suggestions for your text:
> Looks like this text is perfect!
Hmm.
> Consider if your readers know the Latin expressions "i.e."
If they don't know that one, good luck to them reading the rest of the paper.
Hello, developer is here. Come on, why are you people so picky? :D It's a one-man open-source project. I took most advices from the books about scientific writing. If you find some suggestions incorrect, I'll be happy to see examples on github and correct accordingly. Thank you for your interest.
I thought it would nitpick everything but it can be satisfied.
> This sentence is false.
No issues!
Interesting but largely incorrect, at least in part because it wants to correct everything to the "American English" patois.
Do such systems exist for writing and reviewing requirements and test objectives? This would be great to keep them consistent.
This is great! As a Developer, I think this could be handy for PR/MR reviews too.
This isn't the kind of angry reviewer I was expecting.