Whilst the first post on Paper Writing was focused on delivering a paper, and providing structure for peole who have yet to write a paper, one year on, what would I change about it? Is it still relevant?

In the original focus, we being with no topic, no ideas, and spend half the time:

  • Reading
  • Crafting a proposal (the why)

My Thought Process

Often times writing a paper is iterative. One could easily come up with ideas (theoretical or otherwise) whilst reading reasearch. However it is still important to realise that the split in time is more than reasonable! If you don’t have results, your paper is worth nothing!

Steps:

  1. Writing your first outline
  2. Run your experiments

These steps are both worth half the amount of time you should expect to spend! That is, if you spend 2 months writing your outline, before working through empirical results, expect to spend another 2 months working through those empirical results.

Writing your first outline

The paper you write should be addressing new contribution. This might only be 2 or 3 points, but the outline of your paper should do everything in its power to address those three points! The simple structure which is adhered to many papers are:

  1. Introduction - the motivating purpose and outline of your contribution
  2. Related works and background material - the framework (or justifying why) for your new contribution. This is where you can write up what you’ve learnt and assess from your readings. At this point in time, if you realise that your idea isn’t sufficiently new - that is okay, you’ll just reset.
  3. Framework of your contribution - even though you have not run any experiments, you should be able to provide at least a brief outline of what you expect. It need not be correct, but simply to address what gaps you would have to fill after your experiments

Experiments and Results

Once you’ve completed your first outline, this would represent roughly half-way point of your paper writing. The second half is coming up with the experiment and results. There isn’t much to prescribe with respect to how one might do it, but it is always important to compare results with what has previously been done, and ensure that the metrics chosen are not misleading.

Getting Feedback

An important part of this whole process is continually getting feedback from peers and supervisor to ensure that you are on the right track. Without feedback, we can lose sight of the larger picture and end up with a sub-par paper.