From the ChemBlog…
Writing the intro for papers is silly
I’m starting to write my second paper for JOC, fresh off the heals of my recent JACs submission* (fingers crossed) and I’ve sort of grown tired of the formulaic way the introduction of papers are written. I can basically boil it down with the Mad Lib here, so you can write your own paper in the future:The obligatory first sentences (Pick two, mix with proper punctuation to create an opening statement):
- (insert the germane specialty) has become of considerable interest recently.
- Efforts to develop an effective (means/method to do something) have emerged recently but currently there exists very few instances in the literature which (insert limitations they have that your novel crap doesn’t have)
- The (your family of chemical compound or compound class) has become the focus of research efforts based upon emerging applications in (list at least 3 things. I.E. nanofilms, antitumor agents, biowarefare detection shit)
- (Some chemical transformation which the world needs) has been challenging and often requires (list of awful conditions culled from carefully chosen literature).
- Here we present a method to (your method) using (some awesome feature, I.E. greater yields, higher e.e., lower temperatures, without the use of explosive peroxides/azides, autofellatio)
Now that the beginning is done, the rest of the article almost writes itself and you can fill it with data and facts and pictures.
Here is an example of a filled out Mad Lib using the above methods I made:
Crapping out awful introductions for papers has become of considerable interest recently. Efforts to develop an effective means to open a paper have emerged recently but currently there exists very few instances in the literature which don’t make me want to shoot myself in the face. Here we present a method to mock the fuck out of people that use these cliche openers using only a withered sarcastic spirit and little prospect for trying to write my JOC article anymore today.
I agree completely – and thanks to ChemBlog for raising this.
However it has one unexpected benefit – it makes it easier for machines to read it. So in our Sciborg project (see Peter Corbett’s blog) we are analysing chemical language for these types of phrases. We then use them to work out the motivation for writing the paper. For example:
“In previous work we have shown that foobarium iodide provides a facile and novel method of oxidising aldehydes[1], ketones[2], alkanes[3]. Now we show that it oxidises cycloakanes…”
is an immediate pointer to the machine that unless its owner is passionate about foobarium iodide (or is dependent on the goodwill of the authors for their next grant) it can inform the human reader that the paper is boring and can be skipped. There are many other stock phrases conveying motivation.
Even worse are the recipes and the stilted language:
“To 3 g of (1) dissolved in toluene were added 2 g of (2)”.
This is equivalent to “to my dog was given a bone by me” instead of “I gave my dog a bone”. Even better would be to encode it in CMLReact so the machine could read it and work out the quantities, mole fractions, etc. It’s easy to do – so let’s start doing it. It would make it much easier to check all those yields in the literature.
But first we’re liberating crystallography
Then spectra
Then reactions
BTW Many chemical syntheses are primarily competitive sporting events so the real motivation should read:
“we chose something to synthesise that was harder than previous work and put an army of grad students onto it” .
> “Here we present a method to (your method) using (some awesome feature, I.E. greater yields, higher > e.e., lower temperatures, without the use of explosive peroxides/azides, autofellatio)”
Autofellatio, huh? Doesn’t that preclude female authors from using the template by definition?
db
Coincidentally, I was just reviewing the literature on discourse analysis of reports of experimental research in science. This obvious pattern thing actually has a name (the CARS system by Swales), it turns out. I think what may be even more helpful for information extraction is the formally structured abstracts.