Paper Writing

UChicago: The Craft of Writing Effectively

  • Valuable to the audience

    • Valueable is determined by people
    • people in and out over time, knowledge in and out over time.
    • knowledge will be ‘abondoned’ after it is done.
    • What make it be important?
    • Words: neitherless, however, although, inconsistent, reported, …
    • Words creating value to the readers.
      • Words signaling community: widely, reported, …
      • Words for flow/transition: and, but, however,
      • Words for Instability (tension/contradiction): anomaly, inconsistent, but, however,
    • The code shared in the community to create values.
      • Polite way.
    • Ph.D. 4-5 years: learning more things + learning to know your readers.
    • Identify the readers and give what they want.

  • Professional writing is not conveying your ideas to your readers, but changing their ideas.

    • “Why do you think that?” is a bad question to an expert. Teachers ask this because they want to know what the students think.

  • If reader does not understand, do not explain. Why shouldn’t you explain?

    • Explain is to revealing the world the inside of your head. But no one cares about the inside of your head (unless teachers are paid to care about your ideas).

  • Why teachers read student’s text?

    • Paid to care about the students.

  • Why collegues (journals, conference) reading your writing?

    • Valuable >>> Persuasive, Organized, Clear
    • Value is not determined at your Word, but at the Readers.
    • Different people (community/conference/journal) has different ‘value’ judgement.

  • Think — Words, for experts.

    • Think about word in difficult ways.
    • Writing pattern: (Horizontal) Write words to help yourself think. Challenge problems.
    • Reader pattern: (Vertical) Slow down, re-reading; misunderstand; agervated.

  • Rules of writing: good for “low value” writing.

    • short memo every day.
    • think about rules, about readers.

UChicago: More faculty than freshmen.

Tutorial from Derek Dreyer

How to write a paper and give talks that people follow

  • Flow from sentence to sentence: old to new things.

  • Point sentence of a paragraph.

  • Structure:

    • Top-Down: explain multi times, each time deeper with different abstractions. For complex ideas.
    • E.G: abstract -> Intro -> Key ideas -> Tech meat -> related work.

Name your baby. Consistent between collaborators.

Just in time for technical information, not before.

CGI model

Context, Gap , Innovation.

motivation, problem/existing, new stuff.

Understanding how readers processing information.

References:

  • element of style: “Be clear”, but how?
  • Style: Toward clarity and grace. 1990. Jseph M. Williams.
  • How to write a great research paper. Simon

Excellent Samples:

Portable Native Client @ USENIX Security 20101

Styles:

Virtual Ghost 2: introduction = comparison to related + design + implementation + evaluation. No challenges explicitly explained, no contributions explictly listed as bullets.

  • Limitations

    • Limitation Every work has limitations. Some inherits from the design; Some exist because they are just obvious but complex to implement. Reviewers might be interested in this to see how you understand the solution by yourself. Some reviewers will reject you if they pointed out some limitations that you haven’t discussed in the paper. However, if you can discuss the limitation of your work thoroughly, this will show the depth of your understanding in the field and will get your work be understood better.

  1. Adapting Software Fault Isolation to Contemporary CPU Architectures. USENIX SEC, 2010. ↩
  2. Virtual Ghost: Protecting Applications from Hostile Operating Systems. ASPLOS, 2014. ↩
Created Aug 14, 2019 // Last Updated May 18, 2021

If you could revise
the fundmental principles of
computer system design
to improve security...

... what would you change?