The Art of Research Paper Writing and the Research Iceberg Analogy

Tadayoshi Kohno (Yoshi Kohno)
5 min readApr 9, 2022

Here, I discuss the application of the “Research Iceberg Analogy” to the art of writing a research paper.

There are two key topics covered in this post:

  1. During the research process: Many explorations are often needed before fully determining what a paper is about, many of which might end up not being in the final publication.
  2. During the writing process: There is significant value in having a clean, clear, and simple focus for a research paper, which may mean excluding some of the explorations done as part of (1).

This is my fourth post on the “Research Iceberg Analogy.” My first was titled: “Unseen PhD Effort, ‘Failures,’ and the Research Iceberg Analogy.” My third was titled “On Comparing with Others and the Research Iceberg Analogy.” My second was a sidebar to the first.

High-level summary of the “Research Iceberg Analogy.”

For any scientific research project, there is a portion of a research project that is visible to the public (the portion of an iceberg above the surface of the water) and a portion of a project that is invisible (the portion of an iceberg below the water surface).

In order to remain stable, the portion of an iceberg below the surface is much greater than the portion of an iceberg above the surface.

I advise against trying to float an entire iceberg above the water surface.

I have seen the following two related situations:

  1. A researcher is unsure of what to do next as part of their research project. They might be deciding between paths X and Y. Because they are not sure whether X or Y will be in the final research paper, they end up doing neither. They hope to figure out which of X or Y will be in the paper first, before proceeding with either X or Y. As they try to make this determination, their research stalls.
  2. A researcher has done X, Y, and Z (and more) as part of their research. When thinking about writing up their results, they wish to include all of X, Y, and Z (and more) in their paper submission. It feels like a shame to have done so much work and then not include that work in the paper submission.

Situation (1) is analogous to trying to build the visible tip of an iceberg before building the iceberg’s foundation.

Situation (2) is analogous to trying to float an entire iceberg above the surface of the water.

Regarding situation (1): Picking a direction, changing a direction, and learning throughout the process.

If one is truly stuck, my suggestion would (generally) be to just pick either X or Y — let’s say X — and use the process of exploring X to help answer the question of whether or not X will be in the final paper. It might not be. But now, after having made progress on X, the research has progressed and the researcher has learned more about what will be and not be in the paper.

Relatedly, realizing that not all of the work one does needs to appear in the final paper can also make it easier to terminate a specific direction and return to an earlier phase of the research cycle (recall the cycle mentioned in my earlier post). The invisible parts of the iceberg still made a valuable contribution to the project’s overall progress!

The first paragraph above focuses on why doing X can still be valuable for the research even if X is not included in the final paper. The second paragraph observes that it is okay to terminate one research direction, rewind the research cycle, and start a new direction. There is another reason why I think it can be valuable to do X even if X is not included in the final paper: through the process of doing X, the researcher often learns and grows as a researcher.

To me, research is not just about producing research results. Research is a lifelong path of continual learning and discovery. Sometimes we publish what we find; other times we don’t. But whether we publish or not, we grow and mature as researchers.

To quote from my PhD Bubble Diagrams post:

My philosophy is that the addition of new knowledge is a byproduct of the PhD process. To me, the most important part of a PhD is experiencing the PhD process — learning how to….

For more on (1), please see my first Research Iceberg Analogy post. I write more about (2) below.

Regarding situation (2): Taking pride in the unpublished work.

When one keeps the iceberg analogy in mind, I believe that it becomes easier to take pride in the unpublished aspects of one’s work and be okay with those aspects never being visible. It is the hidden part below the surface that made the visible part possible. And the visible part is beautiful.

More on (2): The shape of an iceberg, submission contents, and peer review.

One can continue the above observations in the following way: suppose that an author has somehow managed to include all aspects of their work in a paper submission — somehow the author has succeeded at (2) above and managed to float and balance an entire iceberg above the water. When subjected to external forces — reviewers for papers (or ocean currents for icebergs) — the paper (the iceberg) would likely topple.

In a balanced iceberg (most below the surface of the water), the paper clearly and accurately distills the research. In an unbalanced iceberg (most above the surface of the water), every aspect of the research is crammed into a paper submission. In the balanced case, reviewers have a clear understanding of the contributions.
The Research Iceberg Analogy and the Art of Research Paper Writing.

In my experience, the more simple and focused a paper submission, the more likely it is to get published.

For example, if a paper contributes too much, reviewers may not understand what the paper’s contributions actually are, different reviewers may take away different contributions and start disagreeing with each other over what the paper is about, reviewers may focus their criticism on one tangential contribution and not see the strengths of the other contributions, and so on.

My comments above are not meant as criticisms of reviewers; reviewing papers can be difficult! Rather, my observations are intended to be of help to authors as they navigate writing up their research results.

High-level summary: When possible, I suggest striving for clear, focused, and accurate paper submissions. In writing this, I am also reminded of the following quote: “I have made this longer than usual because I have not had time to make it shorter.” Of course, it is also possible to be too short, so some balance is needed.


I learned so much about the PhD process from my own advisor, Mihir Bellare. After obtaining a faculty position, I continued to develop my own philosophy on advising and the PhD process through the advising my own PhD students. Much of my thoughts on advising and the PhD process have also been shaped through the co-advising PhD students with UW Security Lab co-director Franziska Roesner. Thank you to all the students and postdocs that I have advised, past and present! Thank you also to Kaiming Cheng, Camille Cobb, Ivan Evtimov, Earlence Fernandes, Umar Iqbal, Lucy Simko, and Eric Zeng for comments on my writings about icebergs.



Tadayoshi Kohno (Yoshi Kohno)

Tadayoshi (Yoshi) Kohno is a professor in the UW Paul G. Allen School of Computer Science & Engineering. His homepage: