There is no simple answer to this question. It is safe to say that we cannot think of any reason for not publishing but, when thinking if you should, let's consider the reasons for wanting to. If you are working on a topic which is hot and you know that many people are also researching it, it is important for you to stake your territory and start to establish a reputation in the field. Also, in fast developing areas such as computer science, it is of course vital to publish your results, since waiting until you have finished the whole thesis may render them obsolete. (Check whether your university will accept as a PhD thesis a series of articles published during your candidature.) Publishing will help you to judge whether your work is of a high enough academic standard, or contains an original contribution to knowledge--two concerns often raised by PhD students. Having to publish forces decisions about handling data and helps you organise yourself by providing an interim deadline. So writing a seminar or conference paper or an article for publication could be a very good idea. It gives you an opportunity to practise your academic writing. It exposes you to critical assessment of your work. And it gives you the psychological boost of knowing that you are achieving something worthwhile. However, the design and structure of your work may mean that isolating a discrete, publishable part is not possible. For example, it may not be until the end of your research that the value of the work can be shown. In this situation you mustn't feel pressured to publish. But you would be wise to find alternative forums for presenting your work to your peers such as at seminars, conferences and the like. Of course, time constraints will also govern your decision to publish. In addition you have to weigh the risks of rejection against the benefits of publication. These issues and also questions of where to publish and the protocol of possible joint authorship should be discussed with your supervisor.

	Wrestling with the idea of making an original contribution.
	Writing an abstract.
	Establishing a relationship with your supervisor.
	Publishing papers.
	Some writing tips.