Technical papers follow a fairly standard format. You'll have a title, abstract, introduction, body, conclusions, recommendations for future work, appendices, appendices and a bibliography. Each section has its own particular pitfalls.
• Use a title which clearly states the subject of your paper.
• Your abstract should be both brief and comprehensive.
• Your introduction should explain your problem; why it warrants exploration, why it's not been solved before, and the major points of your research.
• It's always important to describe other relevant research.
•If necessary, include a "Preliminaries" section to introduce the reader to unfamiliar terms, definitions or notations.
•Make sure the body of your paper follows a clear and concise narrative.
•Your conclusion should be a brief summary of your findings.
• Include a section on future research, either to describe your own work plans, or to inspire others looking for new directions.
•Reserve appendices for technical information which someone might find interesting, but isn't necessary for the reader to understand your work.
Technical writing demands clarity. Your audience should be able to understand what you mean, in the exact way you mean it. To ensure this happens, use simpler sentence structures whenever possible, particularly when you're trying to describe complicated processes. Is it possible that a sentence could be misunderstood? If so, rewrite it. Provide figures, graphs, and equations, both to illustrate your findings and to effectively communicate with those who think visually. Make sure your narrative remains linear, that each step follows the previous one in logical order, so that your conclusion seems inevitable. When you've finished writing, put your paper away for a day or so, then take it out to proofread. If possible, have a few "beta readers" read and critique your work.
Get to the point simply and clearly. It might help to remember the point of technical writing: you're writing to inform, rather than entertain. When you're explaining your motivation for creating a new kind of hernia mesh, don't feel obligated to give a mini-refresher on hernia types; your audience knows this. Similarly, when introducing your new search engine protocol, don't digress into a history of the computer. It may help to outline your work before you begin writing. Simply complete one point, and then move on to the next. Your paper will be shorter, easier to write, read and to understand.
Communication always has a human component, and there are mistakes you'd do well to avoid. First, make sure you follow the guidelines provided by your committee, conference or professor. When citing others' work, make sure you do so accurately; don't read into it, finding a conclusion that isn't really there. If you must criticize another's work, do so constructively and with consideration. And, while it might seem calculated, include (and read) relevant research conducted by those on your conference panel, or published in the journal to which you're submitting. Finally, make sure your research and your writing are original. Don't give authorship credit to someone who has not contributed intellectual content to your work, and don't seek it in another's work if you have not earned it. Academic and scientific communities are small and have long memories. If you wish to be successful in them, treat your colleagues with respect.