Some tips for writing a good paper

If you are a student or a young researcher beginning to write a technical paper, you may find yourself in a difficult position: most likely you have received little or no specific training in how to write well, yet your paper will be the main outcome of your work by which other people will judge you. On this page I'm offering a few general tips, compiled from personal experience, which you might find helpful to consider.

• First of all, ask yourself whether this paper is worth writing. A good paper should contain new and technically correct results which solve a challenging and well-motivated problem. Will your paper meet all of these criteria? If not, maybe you should postpone the writing and return to the research. Conference deadlines and other such forms of peer pressure are poor excuses, and disappointed readers will be unlikely to take them into account.

• Before you actually start typing, take the time to think carefully about what you're going to write. What will be the main message of your paper? Can you clearly visualize the whole paper in your head? Have you decided on the best section structure? If you sit down in front of a computer too early, you'll end up wasting time. A similar suggestion applies if you get stuck in the middle of writing: get up and take a walk, and don't resume writing until it becomes clear to you how to proceed.

• Do not postpone working on a paper until you are no longer excited about the results and have moved on to other things. If the technical content of the paper is still fresh in your mind, writing the text will come more naturally to you and you'll produce a better paper with less effort. Avoid the common practice of putting a rough draft aside to polish it later, because returning to it will be painful. Careful writing also helps uncover errors and gaps in your reasoning.

• Writing a paper is a culmination of several months of work, during which you have thought about certain concepts and ideas until your brain became very accustomed to them. Many aspects of your problem which used to be fuzzy have gradually become clear to you, and it is easy to forget the steps that you struggled with. The readers weren't there with you through this process, though, and this is something you need to constantly keep in mind as you are writing.

• An important choice to make is regarding the level of generality at which you are going to present your results. The context in which you originally developed your ideas was probably somewhat specific, but they will have more value if cast in a more general conceptual framework. However, if you go too far in this direction, your results may start to look impenetrable to all but the most sophisticated readers. This problem can to some extent be mitigated by giving very specific examples.

• Motivating the reader doesn't end with the introduction. Remember that most readers will read your paper only if they want to, not because they have to. So, you have to provide continual motivation for them to keep reading. For example, it is usually not a good idea to start a long calculation that will eventually lead to an important result. It is better to first state the result, explain why it's important, and then give the calculations required to prove it.

• The complexity of your writing should be compatible with your English ability. For many of us, English is not the native language. Only use a fancy word or a complicated grammatical construction if you feel confident that you are using it correctly. Otherwise, substitute it with something simpler. You should be in control of the words that you're writing, not the other way around. And remember that many of your readers won't be native English speakers either.

• It is absolutely necessary to print out a hard copy and proofread it carefully, preferably after a break and away from the computer. If you're looking at the screen to check what you've just written, many mistakes will go unnoticed because your brain will automatically substitute what you wrote with what you intended to write. It is also easier to see the "big picture" when you're holding the page in your hands.

• Learn to be self-critical. Ask yourself not whether what you wrote is good enough, but whether there is any way to further improve it. No matter how much you try, other people will still criticize some aspects of your paper, but at least you'll be able to say that you did your best. On the other hand, if people see that your writing is sloppy, they will assume that your research is sloppy as well.

• Remember that you are also writing for yourself, because you will be frequently revisiting your papers in your future work. Getting stuck while trying to follow your own paper is very frustrating (and there is nobody to ask for help). This is another motivation for writing clearly and not covering nontrivial steps with phrases like "it is easy to see that..."

• Be responsive to reviewers' comments. Don't be combative. If a reviewer misunderstood something in your paper, chances are that many readers will misunderstand it as well. Use this feedback to improve your paper. Of course, this doesn't mean that you should make changes that go against your own taste and judgment just to satisfy the reviewers.

• It is certainly helpful to read books, articles, and online resources about good writing practices. Papers that you read for your research are also a great resource: while reading them, carefully note what you like and don't like. Just remember that everyone has their own writing style and not all advice will work for you. This is of course also true of the advice given on this page.