It is reviewing time again. I am wading through a number of papers for various conferences in applied language technology. Most all are interesting to read. Some are fun. But a reviewer does get annoyed at some things.
1st gripe: "This paper is organised as follows"
A general annoyance I find common to many papers – unrelated to other qualities of the paper – is a paragraph wasted on verbiage such as "This paper is organised as follows: Section 2 describes related work, Section 3 our hypotheses … ".
What a waste of space! A paper the length of most conference papers should be able to stand on its own.
While writing a research paper is more of a craft than an art and needs rules of thumb and codified practices for the novice author, there is no need to use up space in the finished product and time and attention of the reader. The paper should be able to support its structure on its own.
2nd gripe: "Future work"
Another general annoyance common to many or even most research papers is a discussion on "future work". A research report should, again, stand on its own. The results should be useful for others as they stand, not with respect to future work, yet undone. The implications of a "future work" section can be understood as
- that this work is only preliminary (in which case it should not be published in a serious conference or journal) or
- that the authors wish to do something else in the future (which of course is not very interesting to the reader of the research report unless that reader is the supervisor of the author) or
- that there are important choice points along the road left unexplored.
This last motivation is the only real value of such a section. It would be better formulated in a less exclusionary manner. Writing it up as "future work" doesn’t invite others to participate: it is a way for authors to stake their claim in directions they have yet to explore. And that is very unfair to the reader!
And of course, thinking about the future of the written paper itself: a section on future work doesn’t age well.
3d gripe: SAD (Spurious Acronym Disorder)
Many or even most papers – or more correctly, authors – in my field suffer from SAD – Spurious Acronym Disorder. Or actually their readers do the suffering. Using acronyms of any kind is ill-advised in a text intended for other than the immediate colleagues of the author, such as a text presented at an international conference. With modern typing aids and word-processors, acronyms serve no purpose – they mess up the reading
of the text and give an unwarranted and often undeserved impression of technical specialization. Typed words are neater in print, easier to process both when reading continuously and skimming, and less intimidating to the unwary and inexperienced reader. And by contrast, silly acronyms give an unprofessional and sub-technical tone to a scientific paper. For most people in the world IR stands for Infrared, not for Information Retrieval.
Oh, and in case you wonder and before you start downloading my collected publications and tech reports … yes, I have myself at times written stuff just as annoying as all of the above.