CSC 3990 -- Feedback on Paper Summaries

1. In a summary report, do not say "we", "I" or "you". Here are some examples (drawn from your summaries) of what not to do:

“I liked how the research focused on […]” à

“One positive aspect of this research is the focus on […]”

“I have questions about”

“It is unclear whether […]”

“The paper does not address the problem of […]”

“I feel that, with some further research […]” à

“It may be possible that further research […]”

“Personally, I found the proposed solutions to be […] à

“The proposed solutions are […]

2. Avoid uncertainty. Present the results in a clear and unambiguous way. Here are some examples (drawn from your summaries) of what not to do:

“The authors try to show […]”à

“The authors show”

“The authors claim that […]”à

“The authors show that […]”

(No claim ever goes in print without supporting evidence)

“The paper attempts to solve the problem of wireless mobile localization […]”à

“The paper addresses the problem of wireless localization”

(The localization problem is huge; this paper focuses on a small sub-problem)

3. Eliminate comments on the clarity and quality of presentation, or other comments that would fit under reviewer’s tasks. Examples:

This paper was well written

The work is original, and significant.

The paper is well organized

4. Avoid acronyms! You should always first spell out an acronym, and use it only when you find yourself repeating the same thing over and over again. Examples:

“[…] at least n/2 # guards” à what does # stand for?

“[…] required to solve the SGP” à what does SGP stand for?

5. Do not summarize each section in the paper individually. We are interested in the content and contribution of the paper, not its physical form. Examples of what not to do:

“One theorem suggests that […]”

“The paper goes on to establish […]”

“Section 3 deals with the upper bound […]”

“The authors dedicate an entire sixth section […]”

6. Do not use contractions (such as "don't").

7. The problem statement should be as clear and accurate as possible. Someone reading your problem statement should be able to start working on the problem right away. Here is an example of an inaccurate statement (drawn from your summaries):

“The paper focuses on the problem of wireless localization […]”à

“The paper focuses on a particular aspect of the problem of localization”

(The focus is on a smaller sub-problem of the localization problem)

8. Once the problem has been defined, your summary should state the main result up front:

The authors establish an upper bound of n-2 and a lower bound of n/2 for the guard placement problem for arbitrary polygons.

9. Proposed solutions should briefly describe the methods used to solve the problem. Most students have been unable to distinguish the results from the proposed solutions.

Two main solutions have been proposed. The first one handles convex and orthogonal polygons, and places one natural guard at every other vertex (thus establishing the n/2 lower bound). The second solution handles arbitrary polygons and is based on partitioning the polygon into smaller pieces, then guarding each smaller piece in a straightforward manner. This latter solution establishes the n-2 upper bound.

10. In a future work section, identify results that could be pursued further:

It has yet to be determined whether there exists a polygon that requires n-2 angle guards. Closing the gap between the n/2 and n-2 bounds remains open.

11. In a discussion section, identify the strengths and weaknesses of the paper. Several students have been unhappy with the fact that the paper starts with explaining the wireless localization problem, but then it restricts to simply placing guards around a polygonal area. This is a common research technique – to present the bigger context in which a smaller sub-problem fits. The wireless localization problem is huge, and not one single person or even research group could solve it. Several research groups work on different aspects of this problem – some on guard placement, others on key definition and/or implementation, others on key encryption and/or transmission, others in validation protocols, etc. The ultimate solution is a combination of these otherwise isolate results.

If you found that the introduction has been misleading though, you should say so:

“The presentation is somewhat misleading in the sense that the wireless localization problem presented in the introduction is reduced to the simpler problem of placing guards around a polygonal region, and ignores several important issues necessary to solve the original problem.”