How to Write Journal Paper in Computer Networking

Sean Xiaojun Hei

* Reasons to Publish Journal Paper
* Overview of Journal Paper
* Structure of Journal Paper
* Linguistic Characteristics
o How to Create a Niche in the Introduction
o Tense and Voice
o Reference Citation
* Summary
* References
* About this document ...

Reasons to Publish Journal Paper

With the full emphasis upon research activities, postgraduate students ought to contribute in the research community to fulfill the requirements of the degree program. One of the most important way to contribute is to publish paper in research journals, especially in those best journals with the good reputation in the area. Ideas are cheap [1], and the only thing that counts is action. There are more important reasons than merely meeting the degree requirements in the University [1]. First, the writing process provides us with a chance to sit down and put our ideas and evidence on paper, so it is possible to refine them and even find any flaws in the arguments or methodology because the whole writing process forces us to think though our ideas more carefully and structure them more logically. Second, writing helps us revisit our ideas and theories and look at them again in a fresh, more impartial way, therefore, we have the opportunity to make improvements. Besides, the publishing procedure generally involves a refereeing process, which will probably result in feedback. Because the referee is most likely to be a respected leader in our research area, we have the chance to get free valuable advice on how we can improve our work. Last but not least, by publishing our work, we can boost our self-confidence and we can obtain satisfaction and recognition.

Getting published begins with the desire to do so, swiftly followed by action [1]. The actual research contribution is one of the deterministic factors whether a paper can be accepted by a journal. On the other hand, the published journal papers have their own characteristics in terms of language. In this essay, we aim to study the linguistic aspects when writing a journal paper in computer networking by the case study of six papers about a particular area ``Reliable Multicast'' from one of best journals in the area, IEEE/ACM Transaction on Networking.

Overview of Journal Paper

There are distinctions among a report, a thesis and a paper. A report simply relates facts in neutral fashion, while a thesis definitely seeks to draw conclusions and assert an evaluation of the material [3]. Computer Networking belongs to Engineering field, therefore, a report in the area also tends to provide a conclusion instead of a mere collection of fact. Generally, it presents in a more detailed way and has a much longer text. A thesis is written to convince supervisors that one has really done the work, so text space puts no constraint in the writing as long as it is necessary for the presentation. However, the journal editors care about space and footnotes because they have to cost the printers' bills [3], therefore, a journal paper has more submission requirements. Notes to Authors are the first reference for the journal paper publishing.

There are different types of journals even in the same academic area. Some may focus on theoretical aspects while others may be mainly about the application. In order to get the paper published in journals, targeting journals is the very first step.

In the following sections, with the case study for the papers in IEEE/ACM Transaction on Networking, we try to generalize some characteristics.
Structure of Journal Paper

There is a general movement for papers in IEEE/ACM Transaction on Networking, including the following sections: abstraction, introduction, authors' contribution, conclusion, appendix, acknowledgment, reference, authors' bibliography. Authors' contribution is the major part in the paper. It is likely that authors further divide the contribution section into several more sections, in parallel to introduction, related work, etc.. Even in the same research topic ``reliable multicast'', the writing style are very different. Anyway, the papers are written by different writers, therefore, it is natural to observe the heterogeneity. However, there are still common moves in this section. Normally, the system model is presented first, followed by the theoretical analysis. Then simulation study or the actually system implementation is carried out. Thereafter, the discussion or the comparison goes into the framework. Sometimes, the section of related work shows up in the paper as an individual part, in which the literature review is done. Also, the future work is presented in a separated section with much more details instead of the common brief description in the conclusion section at last. In the rest of this section, the writing for several sections is covered.

How to Create a Niche in the Introduction

There are typically three moves in research paper introductions [1]: establishing a research territory, establishing a niche and occupying the niche. The six papers in this study are of no exceptions. It is interesting to see that most papers create the niche by indicating a gap with the negative opening in a sentence. The connector however is commonly used to fulfill the task.

* ``The use of multiple multicast channel is expected ... However, in practice ... In this paper, we explore...'' [4].
* ``...(literature review)... However, latency is not considered.'' [5].
* ``Very little work was done on the analysis ... We give an analytical foundation ...'' [6].
* ``However, each packet loss will result in ... In this paper, we present two different mechanisms ...'' [7].
* ``FEC by itself cannot provide full reliability. However, when ... In this paper we study the effectiveness ...''[8].
* ``The weakness of ... This paper further evolves the principles ... for Scalable Reliable Multicast (SRM)'' [9].

I observed that many authors place the word in the initial position of a sentence; however, Ms. Susanna Ho suggested that it had better to be put in the middle of a sentence concerning the tone problem. By placing however at the beginning, authors seems to be introduce a great discovery or invention; however, it is not true most of the time.

Tense and Voice

It turns out that the present tense and active voice are most commonly used in the introduction section when authors present their work, which may be due to the fact that computer networking belongs to the engineering area, so that the researchers hope that their work appears to be true. However, if the literature review is included in the introduction part, the past tense, the present perfect tense and the present future tense are possible to show up due to the actual situation. Though passive voice is common when other researchers' work is referred, active voice is used most of time when authors are talking about their own work.

* ``In this paper, we examine an approach ... ''[4].
* ``CER and DER protocols with optimizations have been compared regarding ...'' [5].
* ``In the following, we will compare ...'' [5].
* ``We propose a new probabilistic feedback method ...'' [6].
* ``We further evaluate our mechanism...'' [6].
* ``Note that local recovery is a performance optimization, ...'' [7].
* ``We find that a layered approach makes more efficient use ...''[8].
* ``It has been argued that a single dynamically configurable protocol should be used ...'' [9].
* ``In 1990 Clark and Tennenhouse proposed ... '' [9].
* ``ALF was later elaborated with ... '' [9].

Reference Citation

Based on my observations, there are two types of citation ways for the reference, either implicit or explicit. Implicit citation occurs when a concept, a model or a algorithm, etc. is referred, while explicit citation often needs to point out the original presenter of the concept, etc. On the other hand, if the citations show up in an individual section "related work", it is more likely that the citation will be presented in an explicit way. Maybe paper authors would like to acknowledge the contribution in a more direct way.

* ``Reliable multicast could benefit from the use of forward error correction (FEC) [24] ''[4].
* ``Cheung et al. [5], McCanne [23] and Vicisano [32] have proposed to ... ''[4].
* ``Further, there are no conflicts with existing classifications [3], [4].'' [5].
* ``In [8], two types of hybrid ARQ are introduced: ...'' [5].
* ``Ammar has defined the feedback problem as response collection via several cost functions [16].'' [6].
* ``SRM [3] exploits heterogeneous delays for a deterministic suppression ...'' [6].
* ``This limits the scalability of SRM as network and group size increases [6], [7]. '' [7].
* ``Hofmann [23]-[25] proposed a "local group concept". '' [7].

John Swales [1] points out that at least two-thirds of all citing statements use the past tense, the present perfect tense or the present tense. The above examples also support the argument. Note that the past tense refers to single studies, the present perfect tense refers to the areas of inquiry and the present tense refers to the state of current knowledge. Besides, the former two tenses focus on what previous researchers did, while the latter one gives attentions to what has been found [1].

Summary

The essay summarizes the findings during the Lang503 study for the writing by the case study.

References

1
John M. Swales and Christine B. Feak, Academic writing for graduate students : essential tasks and skills : a course for nonnative speakers of English, Ann Arbor : University of Michigan Press, 1994.

2
Abby Day, How to get research published in journals , Aldershot, Hampshire, England : Gower, 1996.

3
Ralph Berry, How to write a research paper, Oxford, New York : Pergamon Press, 1986.

4
S. K. Kasera, G. Hjalmtusson, D. F. Towsley, and J. F. Kurose, ``Scalable reliable multicast using multiple multicast channels'', IEEE/ACM Trans. Networking, vol. 8, pp. 29

About this document ...
How to Write Journal Paper in Computer Networking

This document was generated using the LaTeX2HTML translator Version 97.1 (release) (July 13th, 1997)

Copyright © 1993, 1994, 1995, 1996, 1997, Nikos Drakos, Computer Based Learning Unit, University of Leeds.

The command line arguments were:
latex2html paper.tex.

The translation was initiated by Hei Xiao Jun on 5/9/2001