The introduction is certainly the most read section of any
deliverable, and it largely determines the attitude of the
reader/reviewer will have toward the work. Therefore, it is
probably the most delicate part of the writing of a report.
Unfortunately, many people (even very experienced ones) seem to have
difficulties at writing a good introduction. For some, it is a
In this short article, I present a very simple method for writing a
good introductory chapter. Actually, the core of this method was
taught to me many years ago by Krzysztof Apt. At that time, it
surprised me in its simplicity and efficiency. In ten years, I have
been happily applying it to all introductions I have written.
Of course, I am not the first one coming up with such a recipe: a
necessarily incomplete list of links to articles about scientific
writing is reported in the last section.
A Recipe for Writing an Introduction
An Introduction should contain the following three parts:
- 1. Background.
- In this part you have to make clear what
the context is. Ideally, you should give an idea of the
state-of-the art of the field the report is about. But keep it
short: in my opinion this part should be less than a page
long. Half a page should suffice in case of a normal 15-pages
- 2. The Problem.
- If there was no problem, there would be
no reason for writing a report, and definitely no reason for
reading it. So, please tell the reviewer why she should proceed
reading. A simple sentence like "So far no-one has
investigated the link..." or "The above-mentioned solutions
don't apply to the case ...", can sometimes be enough to clarify
the point you want to get at. Experience shows that for this part a few lines are
- 3. The Proposed Solution.
- Now - and only now! - you may
outline the contribution of the report. Here you have to make
sure you point out what are the novel aspects of your
work. There are probably zillions of articles out there on that
very subject: you can't expect the reviewer to know them all; so
make his life easier and clearly highlight what is the
difference between your method and the others. You can take your
time here, but I suggest to avoid getting into too much detail.
In addition there can be the following optional ingredients:
- 5. An anticipation of the conclusions
- This is very
difficult to do properly. I think that this part should be there
only in reports that have a strong position-statement nature. If
you decide to include this into the introduction, you might want
to (a) keep it as short as possible, (b) refer as much as
possible to the concluding section, and (c) keep it well
separated from the rest of the introduction.
- 5. Related work
- My suggestion is to postpone this
part to the end of the paper, unless there are good reasons for
doing otherwise. For instance, one good reason for not following
this suggestion is the presence in the literature of a very prominent
related work, in which case you might want to give immediately an idea of what are the differences
between your work and the prominent one. In general, though, I find it
much easier to have a related works section at the end of the
report: one reads it when he has already a good idea of the
technical contribution. In this case, include in the
introduction a line saying "Related works are discussed in
- 6. The outline (plan of the paper)
- Personally, I
find it useful only for long reports, otherwise
I think it is a waste of paper. But this is my very personal opinion.
Two Extra Tips
- Keep the parts well separated.
- Introductions in which parts 1&2&3 are not clearly separated from
each other are usually very annoying, and are certainly much more
demanding to the reviewer in terms concentration and energy. Use
"itemize", if possible: it helps.
- Keep it short.
- Removing everything that is not really
necessary is often a very effective strategy for improving an
There are many resources on the matter. In particular there is an
excellent website maintained by Toby Walsh with loads of links on on
scientific writing, on presenting scientific articles, etc. http://www-users.cs.york.ac.uk/~tw/phd/
My favourite links:
Other links I found: