2. Abstract Writing

However, we always provide, or we should provide, an additional mechanism for telling people whether to go ahead, and that is the ab­stract. Now, an abstract, as you know, is something that's about any­where, between, say fifty and two or three hundred words long, a de­scription of the work, really to supplement the title in telling people whether this is something that they ought to be pursuing further. One mistake that people often make is they don't give the results. They'll tell what was done – an experiment was carried out to measure this, that and the other, and to test some theory – but they'll never mention whether in fact the theory was true or not. So always make sure that your abstract does include the results that you've actually achieved. Don't give afterthoughts. The nature of things is such that abstracts are usually written right at the end of the paper, and often people who've written the paper and then a couple of days later they think of some­thing they should have said but didn't, will stick it in the abstract. Now that's not appropriate, rewrite the paper but the abstract should corre­spond to the paper and not be a further extension of it.

Abstracts are frequently published by themselves in Abstracting journals, so that a person may read the abstract, who doesn't have the paper before him. Often abstracts are read by very simple people –laymen, controllers, lawyers, directors – so try and keep the technical level of the abstract just a notch below that of the paper. I don't mean to say that, you know, make it such that a seven-year-old can read it. But don't make it as fiercely technical as you know how. I don't think that's appropriate in an abstract.

Well, these are the shorting mechanisms – by these means we re­duce the readers down to the number who ought to be reading the paper. It does no good to have a man read your paper who shouldn't be reading it. It just makes him angry and it retards the progress of sci­ence. So try and don't use it as a come on so much but make it a device to deter people who've no interest in what follows. However, those who ought to be reading it then are going to be with us and we're going to have to take them further.

3. The Structure of a Report

And now we have a fairly standardized line-up of things that happen after this point on. First of all there's going to be an introduction. In this section we outline why we did the work, what the state of knowledge is before we undertook our work – just put the reader in the picture what this is all about. How elaborate you make it, of course, depends on cir­cumstances. You know, some reports have to be, like a thesis would be, very lengthy; some reports to your boss on continuing an experiment that you started in the previous reporting period would be half a page, you know. But in any case this sets the stage for what is to follow.

In experimental studies there is not generally a section called theory – the theory that is needed to understand what happens is usually put in as part of the introduction. But where a study is both theoretical and experi­mental in nature there might, very appropriately, be a section entitled theory, in which the whole theory that is to be tested is developed at some length and evaluated and discussed. I've put a bracket round that to indicate that this is often omitted in experimental thesis, in experimental report.

Well, then there's a description of the apparatus, in which you discuss the equipment that you used. My general comment is: don't put in photographs of apparatus. It simply isn't worthwhile. By all means draw a line diagram in which you point out the appropriate features because anybody can understand a line diagram. And secondly, by all means point out the general weak­ness of the equipment, the features that make it less than ideal because this often helps the reader interpret what follows.

Well, let's carry on. We next come to the results section. Well, this is actually the important section of the paper describing experimental investigation because the results, of course, are the truly significant thing. This is what you did it for. The results are maybe valuable even if the rest of the paper is completely wrong, inappropriate. So perhaps this is the one that you really have to concentrate your attention on. Present your data. What do you present? I suppose one ideal is to present every data point that you ever took. Sometimes this is inappropriate, this may be far too lengthy. You have to then do some selecting. You have to display your re­sults, you have to produce tables, diagrams. All in all, you have to give the reader a flavour of the results that you obtained and really an appropriate description as to what happened when you undertook the experiment. So my gen­eral comment is that it's often appropriate to start writing up the results section of a report while you are still there, you're still tak­ing data and then it often becomes clear that there's a gap and then you can go back and fill the gap. But this is usually the section that is worth writing first, at least that's my own particular opinion.

And lastly there's a discussion section in which you relate the re­sults that you've obtained both in terms of the theory and the state of knowledge at the time that you started the work. Discuss whether what you did was technically or basically a success or failure. Discuss how you, you know, how your results fit in with work of others, with earlier results of yourself, and in general, you know, give the reader an im­pression as to your evaluation of the experiment. Again, the question as to how simple or complex to make this section depends a little bit on just what the purpose of this report is anyway.

Well, in a sense this is the end of the report. On the other hand, we haven't finished yet, we come to a few parts that are often rather tedious but often very important. The first one is recommendation. You don't gen­erally make recommendations in a paper intended for the general public, like a published paper, but you very frequently put in recommendations when you are producing a report that's to be read by a limited number of people, perhaps in your own organization. This recommendation or rec­ommendations are usually what is to be done next. And the time is really important is that if you know that you're going to continue to work in this particular area. Because if you don't make a recommendation then your boss is going to make a recommendation, and unless you tell him what you think ought to be done, don't be surprised if you finish up doing some­thing which you feel is useless. So a recommendation is always an oppor­tunity for you to make things easier for yourself the next time around. So, by all means put in a recommendation when it's a private report.