I find the goal of this study quite commendable. It is an interesting idea to study empirically which writing software (1) minimizes the effort of the writing process and (2) produces the most professionally-looking output.

However, I have to agree with the previous commentators that this study did not actually test either of these questions. The users in this study did not actually produce any text of their own. Rather, they had to REproduce content and layout of already published journal articles. I do not doubt that Latex is not very useful for this purpose, but it does not pretend to be and it also does not matter. I do not think that in my academic life, I ever had to do this kind of task—copying other people’s work is not what scientists usually do. Thus, while the research question itself is valid, the experimental design does not actually address the question at all.

Other commentators have already elaborated on this point, so I will focus on a different concern: What are the implications of the authors' conclusions from a reader's perspective? The authors admit that text produced by Latex may look better, but "we believe that the appearance of text matters less than the scientific content and impact to the field." They also argue that it is a better use of taxpayer money to write academic documents with Word. In my point of view, this argument largely neglects the role of the people who have to read these manuscripts: the reviewers. I have reviewed a sizeable number of manuscripts for journals, grant applications, Master's and PhD theses, etc. In my experience, the appearance of the manuscript does matter a great deal; bad font choices and layout, odd placement of figures, inconsistent typography, inaccurate within-paper references, inaccurate literature references, just to name a few problems, can make the review process quite frustrating and long-winded. I submit that slowing down the speed of reviewing is indeed a bad use of taxpayer money. It definitely is a bad use of my time.

To clarify: as a reviewer, I do not expect that authors emulate the looks of a specific journal or style (as was required in this study). What I do ask for, as a professional courtesy, is that the manuscript be as easy to read as possible. I am well aware that Word and other word-processing software is totally capable of producing professionally-looking manuscripts. However, as it turns out, a great number of word-users seem to be clueless how to pull that off. On the other hand, I have never reviewed a manuscript produced by Latex that had problematic "looks".

To conclude: in my opinion, the authors’ recommendation that journals should not accept manuscripts produced in Latex and to disregard a manuscript’s appearance altogether neglects the reader’s point of view.