In addition to the various tips and guides posted here, an important missing ingredient is actually putting in the time and effort on writing. In computer scien

In addition to the various tips and guides posted here, an important missing ingredient is actually putting in the time and effort on writing. In computer science, at least, I know many people who spend 4-8 months doing research and getting results, and then 1 week writing. This is incredibly foolish, because you're selling yourself short. It takes a lot of time and energy to write (and rewrite) well, and unless you're a top-notch writer, 1 week is not going to be enough. It's just a matter of setting priorities -- writing should not be an afterthought. You may have done the best research in the world, but if no one can understand it easily from the paper, it won't matter. I try to spend a month writing and editing my papers (often going through 15-20 drafts) to make them as good as possible. It seems like a lot of time, until you realize that if your paper is not well-written, it's not going to be read and understood, and so you've lost 8 months of work, not 1 month. The same thing applies to preparing presentations for academic conferences. Most talks are terrible, and so it's actually not that hard to make one that stands out. The main thing to remember in talks is that it's just an advertisement to get people to read your paper; it's not a substitute! From this, you can derive many of the rules for giving good talks: provide strong motivation and background, skip very technical stuff, have a polished presentation, keep it simple, don't be boring, etc.