First the link to this week’s complete list as HTML and as PDF.
Nearly all of the total methane emission from abandoned wells comes from very few highly emitting cases, according to Kang et al., suggesting that little effort might make a big difference here. On the other hand their total sample size is a mere 19 wells, so before acting a much more comprehensive look is in order.
Craftsmen tend to become emotionally attached to their favourite tools and no doubt I am biased regarding Knauff & Nejasmic. That said there are some points to make:
The task set was to exactly reproduce the layout of a few pages of professionally typeset material, where each element occurred only once or very few times. This is the ideal task for a point and click graphic-like interface.
The normal task is quite different: How to create a pleasant, harmonious, and legible document from scratch beginning with pure text or handwritten notes. Using LaTeX and its default settings you are hard put not to achieve that, while Office suites tend to encourage the horrible results you see all too often.
Adapting LaTeX to the very specific rules of one journal is a big job but once done you can churn out articles by the dozen. The Cologne Institute for Prehistory requires me to format bibliographies according to the standard set by DAI. Years ago it took me several days (on and off) to get all the details right, but now I have them typeset perfectly by the dozen or even hundreds at a single key press. The PLOS one article in question, which was itself set using Arbortext priced from four to ten thousand US-Dollars, has eight references, only one of which bearing a link, and that link does not work. Using LaTeX it’s impossible for that to happen to me, and I have by now completely forgotten half the DAI rules and no longer know how to do it manually.
For reproducing a given layout just once and never again even I might prefer an Office tool, were it not that I’d have to learn using it from scratch. Only because of that, I’d probably accept the extra effort and still employ the tool, I’m at home with.
That test was not about efficiently in writing down results to send them off for publication, but rather a task no scientist ever engages in (except perhaps for a hoax) and heavily rigged to achieve a desired result.