写好论文、策划文案及工作报告的十大技巧

夜神◕_̯͡◕

Top 10 tips for writing a paper
写好论文的十大技巧
1.Every paper tells a story
每篇论文都讲述了一个故事。
what is the “elevator pitch” of your story? elevator pitch = summary that is short enough to give during an elevator ride
换句话说，你的“电梯陈述”（elevator pitch）是什么？如何利用搭乘电梯的短暂时间向别人陈述足够精炼的总结？
the story is not what you did, but rather what you show, new ideas, new insights？why interesting, important?
这个故事讲的不是你做了什么，而是你要展示什么。有没有新的想法，新的见解，它们为什么有趣、重要？
why is the story of interest to others? universal truths, hot topic, surprises or unexpected results?
为什么别人会对你的故事感兴趣？发现了普遍的真理，触及了热点话题，让人吃惊，还是有出人意料的结果？
know your story!
了解你的故事！

夜神◕_̯͡◕

2.Write top down
按照自顶向下的顺序写作
计算机科学家（以及大多数人）以这种方式思考问题
computer scientists (and most human beings) think this way!
先陈述宽泛的主题或想法，然后再进入细节。一定要交代背景！
state broad themes/ideas first, then go into detail. context, context, context
即使进入了细节，也要自顶向下来写！
even when going into detail … write top down!

夜神◕_̯͡◕

3.Introduction: crucial, formulaic
Introduction：至关重要，已经程式化
如果读者没有被介绍部分打动，论文就失败了
if reader not excited by intro, paper is lost
结构：
recipe:
第1段：写作动机：很宽泛，问题领域是什么，为什么重要？
para. 1: motivation: broadly, what is problem area, why important?
第2段：缩小范围：你具体考虑的问题是什么？
para. 2: narrow down: what is problem you specifically consider
第3段：“In the paper, we ….”：最重要的一段，写出你的电梯陈述
para. 3: “In the paper, we ….”: most crucial paragraph, tell your elevator pitch
第4段：和其它研究结果相比，有怎样的不同/更好/相关之处？
para. 4: how different/better/relates to other work
第5段：“The remainder of this paper is structured as follows”
para. 5: “The remainder of this paper is structured as follows”

夜神◕_̯͡◕

4.paragraph = ordered set of topically-related sentences
段 = 一些相关话题的句子的有序集合
lead sentence: sets context for paragraph, might tie to previous paragraph
每段头一句：为这一段设定背景；有可能起到承上启下的作用
sentences in paragraph should have logical narrative flow, relating to theme/topic
段落中的句子应该扣住主题，按照逻辑关系进行陈述
don’t mix tenses in descriptive text
不要在描述性的文字里混合不同的时态
one sentence paragraph: warning!
只有一句的段落：慎用！

夜神◕_̯͡◕

5.Put yourself in place of the reader
将自己摆在读者的位置
less is more:“I would have sent you less if I had had time”, take the time to write less
Less is more：花些时间写得更简练
readers shouldn’t have to work, won’t “dig” to get story, understand context, results
读者不应该花费太多精力去理解你的故事、背景、结果
need textual signposts to know where ‘story” is going, context to know where they are
需要一些路标性的文字让读者知道故事进行到哪里了以及相关的背景。
good: “e.g., Having seen that … let us next develop a model for …. Let Z be ….”
好: “e.g., Having seen that … let us next develop a model for …. Let Z be ….”
bad: “Let Z be”
糟: “Let Z be”
what does reader know/not know, want/not want? write for reader, not for yourself
了解读者知道什么，不知道什么，想要什么，不想要什么。论文是为读者写的，不是为你自己。
page upon page of dense text is no fun to read
一页又一页密密麻麻的文字会让人读起来索然无味
avoid cramped feeling of tiny fonts, small margins
避免使用难以辨认的字体和太小的页边距
create openess with white space: figures, lists
利用图表在文字间开辟一些空白区域
enough context/information for reader to understand what you write?
是否有足够的信息（包括背景知识）让读者理解你所写的
no one has as much background/content as you
没有人拥有和你一样多的背景知识
no one can read your mind
没有人能读懂你的心思
all terms/notation defined?
有没有未定义的术语/符号？

夜神◕_̯͡◕

6. No one (not even your mother) is as interested in this topic as you
没有人对这个话题和你一样感兴趣
so you had better be (or appear) interested
所以你最好表现得很有兴趣
tell readers why they should be interested in your “story”
告诉读者为什么他们应该对你的故事感兴趣
don’t overload reader with 40 graphs:
别逼读者读你论文中的40幅图
think about main points you want to convey with graphs
考虑一下你想用图传递的主要信息
can’t explore entire parameter space
不可能探讨整个参数空间
don’t overload reader with pages of equations
别逼读者读满篇的方程式
put long derivations/proofs in appendix, provide sketch in body of paper
将很长的推导或证明放在附录中，只在正文部分提供摘要

夜神◕_̯͡◕

7.State the results carefully
谨慎地陈述结果
clearly state assumptions (see overstate/understate your results)
清晰地说明假设（见9）
experiment/simulation description: enough info to nearly recreate experiment/description
对实验/仿真的描述：需要足够的信息来重现实验以及相关描述
simulation/measurements: statistical properties of your results (e.g., confidence intervals)
仿真/测量：你的结果的统计学性质
are results presented representative? or just a corner case that makes the point you want to make
试验结果有代表性吗？还是说只是为了达到你的目的而进行的一个特殊用例

夜神◕_̯͡◕

8.Don’t overstate/understate your results
不要夸大你的结果，也不要太保守
overstatement mistake: when only actually shown for one/small/limited cases
夸大型错误（如果只是针对一个/很少/有限的用例做出的结果）：
“We show that X is prevalent in the Internet”
“We show that X is better than Y”
understatement mistake: fail to consider broader implications of your work
保守型错误：没有认识到你的工作引申出来的意义
if your result is small, interest will be small
如果结果不引人注意，读者就没有兴趣
“rock the world”
“rock the world”

夜神◕_̯͡◕

9.Study the art of writing
学习写作的艺术
writing well gives you an “unfair advantage”
高超的写作技巧可以让你拥有“不公平的优势”
writing well matters in getting your work published in top venues
写作技巧关系到你的论文能不能发表在顶级的期刊上
highly recommended:
强烈推荐（书籍）：
The Elements of Style, W. Strunk, E.B. White, Macmillan Publishing, 1979（PDF格式下载 ）
Writing for Computer Science: The Art of Effective Communication, Justin Sobel, Springer 1997.
你认为谁是你们领域最会写作的：学习他们的风格
who do you think are the best writers in your area: study their style

夜神◕_̯͡◕

10. Good writing takes times
写作需要花费一定时间
give yourself time to reflect, write, review, refine
给自己一些时间来构思、写作、审查、修改论文
give others a chance to read/review and provide feedback
让别人阅读/审查你的论文然后给你反馈
get a reader’s point of view
了解读者的视角
find a good writer/editor to critique your writing
找一个好的论文写手或者编辑评价你的写作
starting a paper three days before the deadline, while results are still being generated, is a non-starter
如果你在最后期限的三天前才开始动笔，而这时实验结果还正在生成，那你的论文根本没有希望完成