How to write a paper (part 1)

Hangers-on, the evil axis and ‘Chekhov’s gun’ – here Dr Matthew Partridge gives us some vital tips in Part 1 of… 

So you’ve just done some amazing research that is going to change the landscape of your particular field. What you need to do now is publish a paper and share this with the world. You could share it other ways e.g. sky writing or on the side of a blimp; but sadly people tend to only pay attention when it’s published in a journal. So here we are...

Now first off, you need to make sure you are actually allowed to publish a paper. In industry this might be difficult as most companies tend to fear writing down what they are doing and view every experiment as some kind of top secret weapons project. The key to getting a paper published while in a company is to find a way of persuading them it would be a good for marketing. Although, don’t go too far and actually have marketing write it or you might find your technique is renamed “Aspire” and rewritten so it can be available in several different colours.

In academia it’s easier – basically it’s less about persuading people to let you publish your research and more about escaping from the dungeon where you have been forced to write papers 20 hours a day. In academia you also face the problem of having to try and make sure that you don’t get hangers-on to your papers. Despite your colleagues’ protests, being in the room next door is not the qualification for being a co-author, not even if they “thought about it a bit”.

Once you’ve eventually got permission, your next step is to actually plan out what you are going to write. There’s a balance to how much work to include in one paper. I read plenty of papers that put everything and the kitchen sink in one paper. However there are also ones that publish a series of papers which, on their own, are little more than notes. Think ‘Chekhov’s gun’ – if you talk about a gun in the first page of a story then it has to go off before the last page. The same applies in papers – only put in the experiments that actually help your point, don’t go making several points. Unless it’s a paper about pointy things, in which case multiple points are probably okay.

Next, you might want to think a little about where to send it. Each journal has their own style, and writing it from the outset in this style can be helpful – especially if you’re using Microsoft Word. Converting from one style to another in Word is basically a day’s work, although it will probably lead you to develop a whole new set of swear words...

In theory using document preparation systems like LaTeX gets round this problem with its clever swappable style sheets. However, journals love their journal specific LaTeX code which makes it mostly pointless.

The final stage of planning concerns the pretty diagrams. Please, please, please spend time on this – nothing is more annoying than trying to read a paper with diagrams that are part child’s sketch and part MC Escher drawing. The best advice I have is this: The use of more diagrams is better than cramming everything possible into one ink splotch of illustration. The same applies to figures – before you create the figure, make sure you know exactly what you are trying to show – and show only that. And please don’t use 3D graphs, the only thing they show is that you’re terrible at choosing graph formats.

Now with all that sorted, you are finally ready to get to the last stage of writing – actually sitting down and putting words on paper...but you’ll have to wait until next month for this, as I reckon I can keep procrastinating until then…