4 ms·
Yup. And in CS papers, we are told to very early on (within the first 2 or 3 paragraphs, never later), to introduce our contribution. It may use some undefined
by sn41 3y ago
Yup. And in CS papers, we are told to very early on (within the first 2 or 3 paragraphs, never later), to introduce our contribution. It may use some undefined terms and so on, but we should give the reader enough to go on. The "punchline" cannot be 3 pages in. Of course, the rest of the paper is to flesh out the details.
There's also the pragmatic advice given to most graduate students: read the abstract, then the conclusion, then the introduction, then the rest of the paper.
- marcosdumay 3y agoFunny, I usually find it best to jump from the abstract to the methodology. Or, maybe pass through the introduction in between if there's any word I do not understand. The only papers I think the conclusion is more useful than the methodology are on the social sciences and software engineering.
- ilyt 3y agoIf your job is to read the paper, sure, but if you're wondering whether it is even relevant to what you're looking for abstract is useful